Predicting Cryptocurrency Prices during Periods of Conflict: A Comparative Sentiment Analysis Approach Using SVM, CNN-LSTM, and PySentimento

Predicting Cryptocurrency Prices during Periods of Conflict: A Comparative Sentiment Analysis Approach Using SVM, CNN-LSTM, and PySentimento | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Predicting Cryptocurrency Prices during Periods of Conflict: A Comparative Sentiment Analysis Approach Using SVM, CNN-LSTM, and PySentimento Muhammad Nabil Rateb, Sameh Alansary, Marwa Khamis Elzouka, Mohamad Galal This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3949248/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Sentiment analysis is a powerful tool for extracting valuable insights from social media data. In this paper, more than one million tweets spanning three months (March, June, and December 2022) regarding three cryptocurrencies: Bitcoin (BTC), Ethereum (ETH), and Binance Coin (BNB) during the Russian-Ukrainian War are considered. Two models, a convolutional neural network with long short-term memory (CNN-LSTM) and a support vector machine (SVM) with GloVe and TF-IDF features, are trained on a labeled dataset of more than fifty thousand tweets about Bitcoin labeled as (positive, negative, and neutral). A pretrained model (Pysentimento) for sentiment analysis is also employed to compare the performances of the three models. The models are tested on the labeled dataset and then evaluated on the unlabeled tweets, revealing that Pysentimento's level of accuracy outperforms the other two models. Google Trends, along with the opening and closing prices, and the volume of the three cryptocurrencies, in addition to the results of Pysentimento sentiment classification, are employed to apply the Pearson correlation coefficient and conduct price prediction analysis using the SARIMA model. It is found that Bitcoin may appeal to those seeking stability and a known record of accomplishment, while Binance Coin and Ethereum may attract investors looking for more diverse opportunities. Sentiment analysis using machine learning is found to provide invaluable information for cryptocurrency price forecasting and trading strategies, especially in the context of geopolitical events and market volatility. Sentiment analysis Google Trends Russian-Ukrainian War Pysentimento CNN-LSTM SVM Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 1 Introduction Cryptocurrency forecasting has become an increasingly prevalent research topic, with sentiment analysis being one of the key techniques implemented to predict cryptocurrency prices. Political events have been shown to significantly impact cryptocurrency prices and sentiment analysis of social media. Google Trends data is an emphatic method of predicting these fluctuations. The Russian-Ukrainian War represents a substantial occurrence that potentially influences the oscillations of cryptocurrency prices. Sentiment analysis of social media and Google Trends data necessitates the employment of computational methods to identify and extract subjective information, such as opinions and attitudes, from these sources of data. In the context of cryptocurrency price forecasting, sentiment analysis can postulate valuable apprehension regarding how the masses perceive a particular cryptocurrency or the cryptocurrency market as a whole. By analyzing the sentiment expressed in social media and Google Trends data, it is possible to gain a better understanding of how the market may behave in the short or long-term (Wołk, 2020 ). However, there are challenges to conducting sentiment analysis of social media and Google Trends data for cryptocurrency price forecasting. These include the need for accurate and representative datasets, the difficulty in accurately classifying sentiment, and the potential for bias in the data. Despite these challenges, sentiment analysis of social media and Google Trends data remains a promising approach for improving the accuracy of cryptocurrency price forecasting (Sattarov et al., 2020 ). Behavioral finance refers to the study of how psychological biases and emotions influence financial decision-making. In relation to cryptocurrency price forecasting, sentiment analysis of social media and Google Trends data can provide insight into the emotions and psychological biases of cryptocurrency market participants. For example, social media sentiment analysis can reveal how investors feel about a particular cryptocurrency and how sentiment fluctuates over time. Google Trends data can uncover search trends related to a particular cryptocurrency, which can offer insights into the level of interest and attention the cryptocurrency is receiving (AL-Mansour, 2020 ). Cryptocurrency prices are highly volatile and can be influenced by various factors, including global events. During times of crisis, such as the Russian-Ukrainian War or the COVID-19 pandemic, traditional market indicators may not produce an accurate reflection of the market sentiment. This is where sentiment analysis of social media and Google Trends data becomes even more significant, as it can capture the real-time sentiment of market participants. This information can be exploited by traders, investors, and analysts to make informed decisions about their investments (Pano & Kashef, 2020 ). Presently, countless subscribers use social media platforms such as Twitter (presently X), Instagram, and Facebook to share their thoughts about their daily lives and express their emotions. X as a social media platform, is distinguished with a number of privileges to be chosen for this research. First of all, according to Matthew Woodward (2022), globally, there are 436 million active Twitter subscribers per month. Secondly, Twitter has varied access accounts using application programming interfaces (APIs). The Twitter API is a collection of programmatic endpoints that can be exploited to comprehend or create the Twitter discourse and scrap tweets (Faret & Reitan, 2015 ). Finally, Drus and Khalid ( 2019 ) maintain that the majority of social media subscribers worldwide use Facebook. However, due to the unstructured nature of the data, its poor organization, frequent usage of short forms, and high rate of spelling errors, sentiment analysis is not commonly conducted using these data, but rather more frequently by means of X. In the context of the Russian-Ukrainian War, sentiment analysis of social media and Google Trends data can reveal how geopolitical tensions are impacting the cryptocurrency market. For instance, if social media sentiment around Bitcoin is overwhelmingly positive during the war, it may be an indication that investors could to it as a safe-haven asset. Conversely, if sentiment is negative, it may indicate that investors are selling off their cryptocurrency holdings and moving to more conventional haven assets such as gold or the US dollar (Balasudarsun et al., 2022 ). This paper is divided into five sections. The first section is dedicated to the objectives and research questions. The second is related to the review of the literature on sentiment in relation to cryptocurrencies using different implementations for different purposes. The third discusses the methodology and procedures utilized in carrying out this study along with the comparison encompassing SVM, CNN-LSTM and Pysentimento. The fourth is concerned with the analysis and discussion of the findings of the study. The final one provides a conclusion of the study and its findings along with implications for further research. 2 Objectives and research questions The present study proposes to investigate how sentiment analysis could be utilized to predict the prices and interest in the cryptocurrency market even during times of war. Employing the NLP tool referred to as sentiment analysis, Twitter (presently X) sentiment, and Google trends, this research will focus on mining and mensuration of the subjective emotions and opinions in the chosen tweets regarding Bitcoin, Ethereum, and Binance. The ultimate goal of this study is to conduct sentiment analysis on the collected tweets to ascertain if the sentiment is typically positive, negative or neutral with regard to the designated cryptocurrencies. Furthermore, Convectional Machine Learning Techniques such as Support Vector Machine (SVM), Deep Learning (Deep Neural Network) Techniques such as Convolutional Neural Networks and Long Short-Term Memory (CNN_LSTM), and finally Transformer Techniques (pre-trained), for example, Pysentimento are compared to determine which training algorithm will produce the most accurate results. Just as with X, the data are compiled from Google Trends; an open-source tool that displays a single search term's frequency in relation to global search volume over a specified time period in which a user logs. The volume index and closing of last day prices of (March, June and December 2022) are collected. After that,, light will be shed on realizing the tendency toward the three types of cryptocurrencies: Bitcoin, Ethereum, and Binance coin with maximum possible accuracy. Between a multitude of predictor variables (such as “Normalized Closing of Last Day,” “Normalized Volume,” “Normalized Sum Total Google Trends”) and target variables (including “Normalized Overall Sentiment,” “Normalized Positive Sum,” “Normalized Negative Sum,” and “Normalized Neutral Sum”), Pearson correlation coefficients are calculated. Another objective of the study is to provide insights into the strength and orientation of the associations that these variables share and to provide a recommendation concerning which of the chosen currencies can be a profitable investment and value storage during the time of war. In an attempt to predict the prices of the three cryptocurrencies during the first three months of 2023, the SARIMA model is employed. Finally, to reveal the accuracy of SARIMA predictions, root mean squared error (RMSE) is calculated. 3 Research questions (A) Can sentiment analysis on tweets and Google Trends produce accurate predictions about the emotional tendencies and prices of Bitcoin, Ethereum, and Binance coin? (B) What are the differences between SVM, CNN-LSTM, and Pysentimento? Which method yields the best level of accuracy? (C) If predictions based on social media are possible, do they correlate with the trend of the actual interest of the chosen cryptocurrency retrieved from Google Trends? 4 Review of literature Despite the growing research interest in sentiment analysis of cryptocurrency price prediction in recent years, both Pysentimento as sentiment analysis classifier and Binance Coin has been paid less attention to by other studies, possibly for being relatively recent. As far as sentiment analysis for cryptocurrency price forecasting is concerned, there are various studies with variant objectives. The present review covers them as they highlight the significant role of sentiment analysis in predicting cryptocurrency and stock market prices. Abraham (2018) employs a sentiment analysis of Twitter to predict cryptocurrency prices. The data is analyzed to determine whether they can useful information to the final model. VADER sentiment analysis finds tweets to be more neutral, which reduces the accuracy of the results if the general public’s sentiment is neutral because neutral sentiment usually does not indicate a buying or selling trend. The ratio of tweets and Google Trends are both strongly related to prices. A linear regression technique is used to estimate the final daily price of Bitcoin. Twitter sentiment on cryptocurrency is often positive despite possible price changes. Future research should use a more complex model than linear regression to improve the results because the data for this publication is collected when prices are only increasing. Another study has conducted specific research on Google Trends data, concludes that basic seasonal autoregressive models that incorporates Google Trends data as input data surpassed other models which do not involve Google Trends data by 5–20% (Hyunyoung & Varian, 2018 ). However, it is challenging for such models to explain situations where the direction changes. In these situations, Google Trends data can be beneficial. A different study introduces a system that forecasts the prices of four chosen cryptocurrencies—Bitcoin, Ethereum, Ripple, and Litecoin—using social media data and machine learning methods. Random forests, support vector machines, and neural networks (NNs) are applied in this model. The results of this model demonstrate the applicability of sentiment analysis and machine learning methods for cryptocurrency prediction. Furthermore, the prices of certain cryptocurrencies, typically those with a substantial follow-up, can be anticipated solely by Twitter data (Valencia et al., 2019 ). Xin Huang proposes an LSTM model for sentiment analysis. Sina-Weibo, a famous social networking platform in China, provides the data to measure the sentiment. A recurrent neural network based on long short-term memory (LSTM) and historical cryptocurrency values is used to predict the price trend for cryptocurrencies. The results show an accuracy rate of 87%. This result is 15.4% better than that of the currently used traditional auto-regression method (Liu et al, 2021 ). 5 Research Methodology 5.1 Data collection A number of data collection methods are dedicated to collecting and analyzing the chosen tweets and Google Trends. These methods include Twitter API, sentiment analysis, and Google Trends. To answer the research questions, the previous methods shall be employed and explained in the following section. The primary data source for this research resides in an academic Twitter Application Programming Interface (API) account, which facilitates the scraping of tweets over three entire months (March, June, and December 2022) pertaining to keywords such as "Bitcoin," "Ethereum," and "Binance." The Twitter API, an assortment of programmatic endpoints, empowers the comprehension and generation of Twitter discourse (Faret & Reitan, 2015 ). Via establishing an academic API account, the researcher secures access to ten million tweets per month, ensuring an ample supply of requisite data. Concurrently, employing the same keywords and time frame, Google Trends data is collected utilizing the PyTrends library in Python (Fonseca, 2020 ), while Tweepy serves as the library for Twitter data scraping (Russell, 2018 ). Two distinct datasets are utilized for training and evaluation purposes. The first dataset consists of 50,859 tweets that have been categorized as ['positive'], ['negative'], and ['neutral']. This dataset is obtained from the Kaggle website, which is a renowned platform and online community for data scientists and machine learning practitioners (Dundee, 2002 ). For the secondary assessment, a distinct dataset is employed, encompassing 100 annotated tweets categorized as positive, negative, and neutral. These tweets are drawn randomly from the researcher's collection of tweets pertaining to the three cryptocurrencies. Additionally, historical price data for the three cryptocurrencies (Bitcoin, Ethereum, and Binance) is obtained from Yahoo Finance, a comprehensive platform offering cryptocurrency market information encompassing market capitalization, trading volume, price trends, and news. It also provides an API that facilitates access to historical and real-time cryptocurrency data, with price data collected for the same periods as the tweets (March, June, and December 2022) along with the real prices for the first three months of 2023 (January, February, and March) for the sake of price forecasting (Yahoo, n.d.). Google Trends, in turn, serves as a tool to gauge the relative popularity of specific search phrases relative to others over time (Johnson et al., 2020 ). These data provide insights into the popularity of cryptocurrencies in 2022, aligning with the price charts and sentiment analysis results for Bitcoin, Ethereum, and Binance. Tweets are further filtered based on language (English) and location (worldwide), with the total number of tweets collected ranging from 700,000 to over a million for each month. The Google Trends data is accessed through the Google Trends API, which offers insights into the relative popularity of Bitcoin, Ethereum, and Binance for the same periods as the tweets (March, June, and December 2022). 5.2 Support Victor Machine (SVM) According to Cortes & Vapnik ( 1995 ), Support Vector Machines (SVMs) are a type of supervised machine learning algorithm that can be used for both classification and regression tasks. SVMs work by finding a hyperplane that separates the data points of different classes with the maximum margin. The data points that are closest to the hyperplane are called support vectors, and they determine the position and orientation of the hyperplane. SVMs can handle both linear and nonlinear problems by using different kernel functions. The chosen model for this study is (SVM) with (GloVe) and Term Frequency-Inverse Document Frequency (TF-IDF). GloVe is a word embedding technique capturing semantic and syntactic information in vector spaces. These word embeddings serve as numerical representations of words for machine learning algorithms, capturing word meanings, contexts, similarities, analogies, and polarities (Pennington, Socher, & Manning, 2014 ). According to Ramos ( 2003 ), TF-IDF is a statistical measure evaluating the importance of a word within a document or corpus. TF-IDF assigns weights to words based on their frequency in a document and rarity across the corpus. Higher TF-IDF scores indicate greater word relevance to the document. TF-IDF is utilized in this study to reduce word vector dimensionality and eliminate noise and stopwords. In SVM, classification into positive, negative, or neutral categories is accomplished by identifying the hyperplane that best separates data points into different classes. 5.3 Advantages and disadvantages of (SVM) One notable advantage of SVMs is their ability to attain high accuracy and exhibit strong generalization performance by identifying a hyperplane that effectively separates data points into distinct classes while maximizing the margin between them (Cortes & Vapnik, 1995 ). This unique attribute enables SVMs to mitigate the issues of overfitting and underfitting often encountered by other machine learning algorithms like neural networks or decision trees. In the context of sentiment analysis within the cryptocurrency domain, characterized by its volatile and unpredictable data, SVM's ability to provide a reliable and robust model becomes crucial for addressing the inherent uncertainty and variability of the data (Bollen et al., 2011 ). SVMs effectively classify text into positive, negative, or neutral categories based on emotional tone; accommodate various text lengths and formats; and employ features such as word counts, (TF-IDF), word embeddings, or n-grams. These methods can synergize with other techniques, such as feature selection, dimensionality reduction, or ensemble methods, to enhance performance and accuracy (Wang & Manning, 2012 ; Medhat et al., 2014 ). Despite their strengths, SVMs face certain limitations within the domain of cryptocurrency sentiment analysis. Primarily, their inherent binary classification nature necessitates extensions for multiclass sentiment classification, often required in cryptocurrency discussions characterized by multifaceted sentiment landscapes. The computational intensity of SVMs poses challenges when dealing with large volumes of real-time cryptocurrency data, where timeliness is crucial (Schölkopf et al., 2001 ). Additionally, tuning hyperparameters and kernel functions can be time-consuming and resource intensive, complicating SVM implementation (Hsu et al., 2003 ). SVMs' sensitivity to different kernel functions and hyperparameter settings introduces variability in performance, necessitating extensive trial-and-error approaches or grid searches to optimize results (Medhat et al., 2014 ). 5.4 CNN-LSTM In the present research, a hybrid model that combines Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM), originally adapted from Kaggle (Dundee, 2002 ), has been employed with the primary aim of enhancing the accuracy of outcomes. For a comprehensive exploration of the CNN component, it is imperative to delve deeper into the characteristics of Convolutional Neural Networks (CNNs). They are a category of deep learning algorithms specifically designed to address intricate tasks, including image recognition and natural language processing (LeCun et al., 1998 ). The fundamental operation of CNNs involves the application of filters or kernels to input data, which can encompass images or textual content, thereby facilitating the extraction of localized features or discernible patterns. Notably, these filters or kernels are acquired through network learning during the training phase, and they possess the capacity to encapsulate diverse aspects or dimensions of the data, encompassing attributes such as edges, shapes, colors, linguistic constructs, n-grams, and sentiment expressions (Kim, 2014 ). It is pertinent to mention that CNNs may additionally employ pooling layers for the purpose of diminishing data intricacy and dimensionality, alongside fully connected layers to cater to classification or regression objectives (Yamashita et al., 2018 ). Within this integrated framework, CNNs assume the role of extracting localized features or patterns, while LSTMs specializes in capturing temporal dynamics and protracted dependencies inherent in the data (Hochreiter & Schmidhuber, 1997 ). This harmonious fusion of layers imparts to CNN-LSTM the ability to deliver superior performance and address more intricate challenges, surpassing the efficacy of deploying each layer in isolation (Wang et al., 2016 ). The CNNs outputs a vector encompassing all these features. LSTM then processes this vector to generate a sentiment score. LSTM learns to utilize the vector's features to compute a numerical sentiment value for the tweet, taking into account the tweet's context and tone. The final outcome is a sentiment score for each tweet, reflecting its positivity or negativity toward a particular cryptocurrency. A high sentiment score suggests optimism about the cryptocurrency, potentially indicating an expected price increase, while a low score signifies pessimism, possibly signaling a price decrease (Zhang et al., 2015 ). 5.5 CNN-LSTM privileges and drawbacks CNN-LSTM has some merits and demerits that affect its performance and applicability for sentiment analysis of cryptocurrency. As for the first, CNN-LSTM can capture both local and global features along with patterns from the data, while other models may only capture one or the other. This means that CNN-LSTM can handle more diverse and heterogeneous data, and handle more fine-grained and nuanced sentiments (Liu et al., 2018 ). It can handle variable-length inputs and outputs, which is useful for dealing with data of different lengths and formats. This means that CNN-LSTM can process texts of different word counts, languages, or styles, while other models may require fixed-length inputs or outputs (Sutskever et al., 2014 ). It can be trained end-to-end, which means that it can learn the optimal parameters for both the convolutional and recurrent layers without requiring any manual feature engineering or preprocessing. This approach can reduce the complexity and cost of the model development and deployment. Regarding the latter, CNN-LSTM requires considerable training data and computational resources to learn effectively, as it has many parameters and operations to optimize (Donahue et al., 2015 ). This means that CNN-LSTM can be slow and expensive to train and test and may need many trails and errors or grid search to find the best settings (Medhat et al., 2014 ). It may not be able to capture the long-range dependencies and context information in the data, as it relies on fixed-size filters or kernels for the convolutional layers. This means that CNN-LSTM may miss some important clues or signals that affect the sentiment of the text, such as the word order, the sentence structure, or the discourse relations (Zhang et al., 2016 ). 5.6 Pysentimento Pysentimento, a Python library designed for accessibility, provides a user-friendly avenue for conducting sentiment analysis and various social (NLP) tasks. The library achieves this by harnessing the capabilities of models such as BERT and others. Within Pysentimento, users can effortlessly load pretrained sentiment analysis models, including BERT, DistilBERT, RoBERTa, XLM-RoBERTa, or custom models tailored to specific domains or languages. Pysentimento's versatility extends to the application of these models on text inputs or data frames, generating sentiment scores that serve as quantitative indicators of the textual content's positivity or negativity. Carrillo et al. ( 2020 ) developed Pysentimento, making it conveniently accessible on both GitHub and PyPI. In the field of sentiment analysis, Pysentimento opts for BERT as its principal model for conducting evaluations in both English and Spanish. The toolkit extends its capabilities by incorporating models that have undergone fine-tuning to address the specific requirements of distinct datasets or languages. A notable example is Robertuito, a specialized variant within the Pysentimento toolkit crafted explicitly for the analysis of sentiment in Spanish tweets sourced from social media. This adaptation is rooted in RoBERTa, an improved version of BERT distinguished by augmented data and extended training iterations. Pérez et al. ( 2022 ) introduced Robertuito to enhance sentiment analysis within the Pysentimento framework. This comprehensive toolkit is designed to provide users with the capability to perform sentiment analysis on their proprietary data or make use of alternative models available in Pysentimento or the Hugging Face Transformers library. In doing so, a versatile and adaptable solution is established (García-Pablos et al., 2020 ; Devlin et al., 2019 ). Pysentimento has been evaluated on several datasets and benchmarks for different languages and tasks. According to its authors, it achieves state-of-the-art results for sentiment analysis in Spanish and English, and competitive results for other languages and tasks. For example, for sentiment analysis in Spanish, Pysentimento achieves an accuracy of 91.3% on the TASS 2017 dataset, which is a collection of tweets annotated with positive, negative, or neutral labels. For sentiment analysis in English, Pysentimento achieves an accuracy of 90.9% on the SemEval 2017 Task 4A dataset, which is a similar collection of tweets (Pérez et al., 2021 ). 5.7 Pysentimento strengths and weaknesses In terms of merits, the utilization of BERT models within Pysentimento enables high accuracy and robustness for sentiment analysis tasks, surpassing previous models on various natural language processing (NLP) benchmarks and tasks (Devlin et al., 2019 ). Furthermore, Pysentimento offers models that are fine-tuned on specific datasets or languages, thereby enhancing performance and adaptability (García-Pablos et al., 2020 ). This approach is particularly advantageous as sentiment analysis can aid in predicting cryptocurrency trends by capturing public sentiment and opinions, shedding light on the emotions and attitudes expressed in cryptocurrency-related text data (Chen et al., 2018 ; Akhtar et al., 2019 ). Additionally, Pysentimento leverages established and well-maintained NLP libraries and frameworks, such as the Hugging Face Transformers library, which supports numerous models and frameworks. Moreover, it utilizes popular libraries like pandas, numpy, and scikit-learn, ensuring reliability and widespread usage (Hugging Face, 2021 ). Pysentimento does have some demerits. Firstly, users are required to install Pysentimento and its dependencies, which may necessitate considerable time and storage space, depending on the user's system and internet connection. Furthermore, compatibility issues or errors may be encountered during the installation process (García-Pablos et al., 2020 ; Hugging Face, 2021 ). 6 Analysis and discussion The primary aim of this section is to conduct a comparative analysis of the performance of (SVMs (LSTMs), and the Pysentimento framework utilizing BERT for sentiment analysis within the domain of cryptocurrencies. Based on accuracy level, the best model is elected for sentiment analysis. Next, Google trends, volume, and the results of sentiment analysis (positive, negative and neutral) are combined in a table and normalized after obtaining overall sentiment sum. After that, Pearson correlation coefficient are applied and time series prediction implementing SARIMA model is conducted to unveil which cryptocurrency is the safest for investment during times of geopolitical tension. 6.1 Data preprocessing and training of (CNN-LSTM) The training and testing dataset is comprised of 50,859 tweets on Bitcoin that have been categorized as ['positive'], ['negative'], and ['neutral']. This dataset has been obtained from the Kaggle website, which is a renowned platform and online community for data scientists and machine learning practitioners (Dundee, 2002 ). Eighty percent of the data are devoted to training and the remaining portion is dedicated for testing where the training set consists of (40687) tweets, while the testing set comprises (10172) tweets. A larger training set allows the model to learn from a substantial amount of data, capturing patterns and relationships in the tweets that are essential for making accurate predictions. The bar plot indicates that 22937 tweets are labeled as positive, 21939 as neutral and 5983 as negative. This exploratory analysis offers an initial understanding of sentiment distribution prior to model development. (Fig. 1 ) conveys such a distribution. Preprocessing initiates with the column "tweet" turning into "clean_tweet" after preprocessing and the column "label" is the one for sentiments. The word count is a common feature employed in text analysis tasks to assess the complexity and content of text data whereas text length column quantifies the length of each tweet in terms of character count. This metric can be valuable for examining the distribution of tweet lengths within the dataset. The mean and standard deviation enable the researcher to determine the correct average of tweet length employing Freedman-Diaconis rule, which was 30. The training and testing shapes become (40687, 30) (10172, 30). (Fig. 2 ) shows the data after preprocessing with both metrics while (Fig. 3 ) uncovers the word cloud, a powerful visual representation of frequently occurring words in text data, generated at this stage since it offers a visual summary, highlighting the most prevalent words within the dataset. It provides insights into the dataset's underlying themes and characteristics. The epochs’ number is 5 and the batch size is 128. The model in (Fig. 4 ) undergoes five processing stages. To begin with, an embedding layer maps each of 30 input tokens to a 200 dimensional vector, forming a three dimensional tensor of shape (30, 200). It involves two crucial parameters: maximum features and embedding dimensions. These parameters grant the model flexibility in encoding word information, making it adaptable to the dataset's specific The model in (Fig. 4 ) undergoes five processing stages. To begin with, an embedding layer maps each of 30 input tokens to a 200 dimensional vector, forming a three dimensional tensor of shape (30, 200). It involves two crucial parameters: maximum features and embedding dimensions. These parameters grant the model flexibility in encoding word information, making it adaptable to the dataset's specific characteristics. Subsequently, the architecture incorporates two convolutional 1D layers, which introduce a spatial understanding of the text data. These layers employ a set of 1D convolutional filters, alike to sliding windows, to detect local patterns and feature representations within the tweet sequences. Throughout utilizing filters with varying receptive field sizes, the model becomes proficient at recognizing both fine-grained details and broader textual features. The Rectified Linear Unit (ReLU) activation function adds a critical element of non-linearity, enabling the model to capture complex relationships between words. Following the convolutional layers, two MaxPooling 1D layers act as a dimensionality reduction mechanism. These layers systematically downsample the output from the convolutional layers preserving the most salient and informative features while mitigating computational complexity. This process allows the model to concentrate on the most relevant elements of the data, enhancing its efficiency and focus. The neural network architecture further evolves with the inclusion of an LSTM layer as it excels at capturing sequential dependencies and long-range contextual information within text data. This layer is vital in modeling the temporal dynamics of tweet sequences, understanding how words relate to each other over time, and capturing intricate patterns that might span across the entire sequence. Finally, the dense layer maps the LSTM output to a three dimensional vector using linear activation, forming a tensor of shape (3) where the softmax function transforms the model's internal representations into probability distributions across the three sentiment classes: negative, neutral, and positive (Gaber et al, 2021 ). 6.2 Classification report of (CNN-LSTM) Based on Sharma and Sharma ( 2022 ), classification reports are essential instruments for assessing the efficacy of models intended to infer attitudes (positive, negative, or neutral) from textual data. These reports provide crucial metrics, each of which provides unique information on how well the model works. The first indicator, precision, measures how well positive predictions are made in comparison to false positives, or how frequently the model accurately predicts positive feelings. The other one, recall, gauges how well the model can recognize real-world positive examples, demonstrating its accuracy in expressing good emotions. Eventually, the third metric, the F1-score, finds a middle ground between recall and accuracy. This middle ground is especially helpful when the distribution of emotion classes is not uniform. (Fig. 5 ) uncovers CNN-LSTM classification report. The results demonstrate the ability of the proposed method to classify tweets into three sentiment categories: positive (pos), negative (neg), and neutral (neu). As for negative (neg), the precision for the negative sentiment class is approximately 0.92. This indicates that when the model predicts a tweet as negative, it is correct proximately 92% of the time. The recall, at around 0.95, demonstrates that the model captures about 95% of the actual negative tweets. The F1-score, which harmonizes these metrics, is approximately 0.93, suggesting a strong balance between precision and recall for the negative sentiment class. This means that the model excels in identifying and correctly classifying tweets expressing negative sentiments. Regarding the neutral sentiment class, the precision is roughly 0.98, indicating that the model's predictions of neutrality are highly accurate. The recall, at approximately 0.97, signifies that the model correctly identifies about 97% of the actual neutral tweets. The F1-score of about 0.98 reaffirms the model's exceptional performance in classifying neutral sentiment, with a balanced combination of precision and recall. This signifies the model's proficiency in distinguishing neutral tweets from others. The positive sentiment class exhibits similar excellence, with a precision of approximately 0.98, indicating highly accurate positive predictions. The recall, around 0.98, indicates that the model captures about 98% of the actual positive tweets. The F1-score, approximately 0.98, reflects the strong balance between precision and recall for the positive sentiment class. This underscores the model's ability to effectively identify and classify positive sentiment in tweets. The overall model performance is impressive, with an accuracy of approximately 97%. This accuracy demonstrates the model's proficiency in classifying tweets across all sentiment categories. The macro-average F1-score, at about 0.96, signifies that the model maintains a robust balance between precision and recall for all sentiment classes, considering their individual support levels. Additionally, the weighted average F1-score, also around 0.97, indicates the model's consistency in performance across different sentiment classes, considering their varying proportions in the dataset. To conclude, the model demonstrates high proficiency in classifying tweets into positive, negative, and neutral sentiment categories, with an overall accuracy of 97%. It upholds robust balanced performance across sentiment classes, as indicated by macro and weighted average F1-scores of 0.96 and 0.97 respectively. The results emphasize the model's effectiveness in sentiment analysis of social media texts. 6.3 Data preprocessing and training of (SVM) All the steps of this phase are similar to those in the previous model apart from minor differences. First of all, tokenization breaks the text into individual words or tokens, enhancing the model's ability to comprehend and analyze the content. Secondly, stop-words removal is essential for eliminating common but uninformative words. Thirdly, lemmatization reduces words to their base or root forms, ensuring consistency in the data. Variations like "running" and "ran" are both transformed to "run," streamlining feature extraction and pattern recognition. Finally, Part-of-Speech Tagging (POS) enriches the data by assigning grammatical labels to each word, allowing for the capture of specific linguistic patterns. This step can aid in discerning verbs, adjectives, or other parts of speech relevant to sentiment analysis. For this model, the Hugging Face 200-dimensional GloVe feature is imported, and it has been widely implemented in NLP tasks. The 200 dimensions of the word vectors refer to the fact that each word is represented by a vector of 200 numerical values to provide a rich and informative representation of words. The feature works by converting words into a continuous vector space, where the similarity between words can be measured using cosine similarity (Pennington et al., 2014 ). On the positive side, this feature is built upon a vast and diverse corpus of tweets, enabling it to effectively capture informal, colloquial language, slang, and even emoticons commonly found in social media conversations and tweets. It excels at capturing both global and local information from the corpus, including word frequency, word order, and word context. Additionally, the model can unveil intriguing linear relationships between words, such as analogies, antonyms, and synonyms. However, there are certain restrictions associated with the Hugging Face GloVe embeddings. Firstly, its performance is contingent upon the vocabulary size and coverage of the underlying corpus, which may not encompass rare or domain-specific (cryptocurrency) words, potentially limiting its applicability in specialized contexts. Secondly, the feature may struggle to capture intricate and nonlinear relationships between words, including polysemy, homonymy, irony, and sarcasm. Lastly, it may encounter difficulties in handling out-of-vocabulary words or common typographical errors commonly encountered in the informal language of tweets. Understanding these merits and demerits is crucial when considering the application of the Hugging Face GloVe embeddings 200d in various natural language processing tasks (Pennington et al., 2014 ). Conversely, (TF-IDF) is a fundamental technique in (NLP). Being part and parcel of this model, TF-IDF addresses a critical challenge in NLP which is how to represent the inherent information and nuances of text data in a way that algorithms can effectively process (Sharma et al., 2023 ). TF-IDF vectorization offers several advantages when compared to alternative text representation methods like bag-of-words or word embeddings. First of all, it addresses the issue of high dimensionality by considering only words that appear in at least one document within the dataset. TF-IDF aids in solving the obstacle of having too many different words to deal with by only looking at words that appear in at least one document in the dataset. This reduction in dimensionality makes it computationally efficient and manageable. Secondly, TF-IDF captures both local and global information about words. It accounts for word frequency within a document and across all documents in the dataset, providing a holistic view of word importance. Thirdly, TF-IDF assigns higher weights to words that carry more informative or distinctive characteristics for a document or topic, while assigning lower weights to common or generic words. This feature is particularly valuable for highlighting the significance of words in context. Finally, TF-IDF is known for its simplicity in implementation and interpretation, as it does not necessitate complex mathematical operations or external resources (Mikolov et al., 2013 ). Based on Bird et al. ( 2009 ), there exist certain limitations and challenges associated with TF-IDF vectorization. To begin with, TF-IDF assumes that words are independent of each other, disregarding their order and context within a document. This can lead to the loss of valuable sequential information. Moreover, TF-IDF does not capture semantic or syntactic relationships between words, such as synonyms, antonyms, or grammatical structures, which limit its ability to understand the deeper meaning of language. Additionally, TF-IDF may assign low weights to words that are relevant but infrequent across all documents, such as proper nouns or domain-specific terms. Eventually, TF-IDF can be sensitive to outliers or noisy data, including spelling errors or typos. To mitigate some of these limitations, the researcher employs complementary techniques like stemming, lemmatization, NLTK stop-word removal, and adding (pos) to tokens with TF-IDF. These enhancements improve its overall performance and accuracy, making it a versatile choice for text analysis tasks and that has already been put into practice at the first step of pre-processing. (Fig. 6 ) uncovers the tweets after preprocessing steps executed. The input data shapes are examined, with train GloVe at 40,687 samples by 200 features and test GloVe at 10,172 by 200, encoding tweets in a semantic space. Meanwhile, train TF-IDF and test TF-IDF have the same sample counts and 9991 features, representing tweets high-dimensionally. This conveys the input data shapes of GloVeand TF-IDF. GridSearchCV is a tool that performs an exhaustive search over specified parameter values for an estimator using cross-validation. Cross-validation is a technique that splits the data into k folds, utilizes one-fold as the test set and the rest as the training set, and repeats this process k times, averaging the results. This approach can provide a more reliable estimate of the performance of the estimator on unseen data. Three-fold cross-validation is employed to evaluate each combination of parameter values. This conveys that data will be split into three parts, and each part is used as a test set once, while the other two parts are employed as a training set. The average score across the three folds is used as the performance metric for each combination. The parameters that are tuned are as follows: C, which is the regularization parameter for SVM that controls the trade-off between margin maximization and error minimization; and kernel, which is the kernel function for SVM that determines the type of transformation applied to the data. After that, the researcher designates 'all' as the value for k, indicating the selection of all features by SelectKBest. The 'SVM C' parameter, which represents several levels of regularization strength, is set to a range of values, including 0.1, 1, and 10, in the SVM classifier. Furthermore, two kernel options "linear" and "RBF" (Radial Basis Function) are defined for the 'SVM kernel' parameter, which affects how the SVM maps input data. The first one finds a linear hyperplane that separates the data points based on their features (positive, negative neutral). The second maps the data points into a higher-dimensional space where a linear hyperplane can separate them better than in the original space. The grid search process evaluates the model's performance in a systematic manner across different sets of these hyperparameter values in order to determine which set best improves model performance. The results are illustrated in (Fig. 7 ). The figure conveys that utilizing "all" features, setting C to 10, and employing the RBF kernel provide the best performance for SVM with GloVeand TF_IDF features on the data. C is a hyperparameter for the SVM classifier that controls the trade-off between margin maximization and error minimization. A higher value of C means that the classifier tries to fit the data more closely, but it may also overfit and generalizes poorly. A lower value of C means that the classifier will allow more errors, but it may also underfit and miss important patterns. C = 10 means that the value is a reasonably high one of C for this SVM classifier with GloVeand TF_IDF features. The RBF kernel is frequently implemented for capturing non-linear relationships in the data, which is an indicative of the data's complexity and non-linearity. The model achieves accuracy scores of approximately 91.94% and 88.22% with GloVeand TF-IDF respectively. Such scores indicate how well the model can classify tweets into their respective sentiment categories. Higher scores normally indicate better model performance. 6.4 Classification reports of (SVM) 6.4.1 Classification report of (SVM-GloVe) As revealed in (Fig. 8 ), with regard to sentiment analysis, the classification report thoroughly provides a thorough summary of the performance data obtained from an established Support Vector Machine (SVM) model that makes use of GloVe features. This assessment methodology includes support, recall, F1-score, and accuracy measures that are specific to the three sentiment categories (positive, neutral, and negative). Central measures of the model's prediction accuracy for every sentiment class are precision values, which are shown as noteworthy percentages. For the negative sentiment category, the accuracy is 91.83%, for the neutral sentiment category, it is 94.46%, and for the positive sentiment category as 91.25%. The recall percentages surface as 95.17% for negative sentiments, 93.54% for neutral sentiments, and 82.39% for positive sentiments. The accuracy, a pivotal metric gauging correctness in classifying cases crosswise all sentiment subclasses, is 92.96%. This overarching accuracy metric provides a holistic viewpoint on the model's effectiveness in sentiment examination. Macro and weighted averages, integral portions of the report, offer nuanced appraisals accounting for stabilizing and circulation of sentiment categories. The macro-averaged precision, recall, and F1-score communicate as 92.51%, 90.37%, and 91.35%, respectively. Conversely, the weighted-averaged precision, recall, and F1-score arise at 92.97%, 92.96%, and 92.92%, correspondingly. These averages offer a more comprehensive assessment regarding the impact of class imbalances on the level of performance held by the model. To summarize, the classification report delivers comprehensive and illuminating information about the SVM model's performance, providing subtle insights into its accuracy, recall, and F1-score for various sentiment categories. While macro and weighted averages offer a comprehensive picture of the model's performance in sentiment analysis tasks, the support metric and overall accuracy add more context. This thorough examination is essential for identifying the model's advantages and possible areas for improvement, enabling well-informed sentiment analysis decision-making. 6.4.2 Classification report of (SVM-TF-IDF) The deployed (SVM) model's sentiment analysis performance is painstakingly evaluated in the classification report in (Fig. 9 ). The assessment is carried out via TF-IDF characteristics, a commonly employed method for determining a word's relevance inside a collection of documents. The report provides a detailed overview of the discriminative skills of the model by covering precision, recall, F1-score, and support metrics for each sentiment category (positive, neutral, and negative). The precision values signify the degree of accuracy with which the model predicts each sentiment class. In the present case, the accuracy of the negative sentiment class is 86.54%, indicating that the model is capable of correctly classifying negative attitudes. The model's accuracy in predicting neutral attitudes is demonstrated by the neutral sentiment class's precision of 94.47%, whereas the positive sentiment class's precision is 90.70%. Metrics for recall offer valuable information on how well the model captures examples from each sentiment category. Recall for the negative class is 95.63%, highlighting the model's capacity to accurately identify a significant percentage of real negative cases. With an 88.49% recall rate, the neutral class demonstrates a strong sensitivity to real-world neutral mood occurrences. Nevertheless, the recall of the positive class is 78.14%, indicating a somewhat lower capture of real positive cases. F1-scores, which exhibit a harmonic mean of recall and precision, shed more light on the model's well-balanced performance. With an F1-score of 90.85%, the negative emotion class exhibits a well-balanced trade-off between recall and precision. Likewise, the class representing neutral sentiments has an F1-score of 91.38%, signifying a well-balanced approach to forecasting neutral sentiments. On the other hand, the F1-score of 83.95% for the positive sentiment class indicates a trade-off between recall and precision for positive sentiments. The SVM model with TF-IDF features has an overall accuracy of 90.35%, which provides a global indicator of how accurate it is at categorizing instances in all sentiment categories. Weighted averages and macro statistics offer a thorough analysis that takes the impact of class disparities into account. The macro-averages for recall, F1-score, and precision are 90.57%, 87.42%, and 88.73%, in that order. On the contrary, the F1-score, weighted-averaged precision, and recall are 90.35%, 90.65%, and 90.70 percent, correspondingly. To conclude, this thorough assessment provides a detailed overview of the SVM model's advantages and disadvantages in sentiment analysis tasks, offering insightful information that can be used to develop the model and make well-informed decisions. The comparative study of SVM model that employs both of TF-IDF and GloVe characteristics has several implications. The TF-IDF feature is regularly outperformed by the SVM model combined with GloVe embeddings in terms of accuracy, precision, and F1-score for a range of attitudes. This consistency implies that GloVe embeddings play a major role in building a more sophisticated and reliable sentiment analysis model due to their capacity to capture complex semantic information. Both features have a high level of competence when it comes to neutral emotions, demonstrating their ability to navigate through a wide range of expressions in this area. The F1-scores accurately reflect the subtle nature of neutral sentiment expressions, which forces models to strike a careful balance between recall and precision. With the highest recall in this category, the SVM model which utilizes TF-IDF features demonstrates a unique ability to efficiently capture occurrences of negative sentiment efficiently. This discovery suggests that TF-IDF could be vastly advantageous in recognizing manifestations of negative emotion, providing insightful information, especially in applications where detecting negative sentiments is critical. These findings highlight the significance of contemplating the intricacies of sentiment expressions and carefully evaluating the trade-offs between precision and recall when selecting a sentiment analysis model. 6.5 Pysentimento This particular model does necessitate neither a preprocessing nor a training phase, as it is a pre-trained model equipped with a tokenizer. The process commences with the installation of the Pysentimento library (version 0.7.2) and its prerequisites. It procures the following dependencies: "accelerate" library (version 0.22.0), "datasets" (version 2.14.5), and "emoji" (version 1.7.0). The model architecture, "robertuito," suggests that this model is based on the RoBERTa architecture. RoBERTa is a variant of the BERT (Bidirectional Encoder Representations from Transformers) model and is known for its effectiveness in a wide range of NLP tasks. The model is fine-tuned specifically for sentiment analysis. Fine-tuning involves training a pre-existing model on a task-specific dataset. In this case, the model has been trained on a dataset of text samples with associated sentiment labels (positive, negative, and neutral). This fine-tuning process supports the model to learn the patterns and features relevant to sentiment analysis. Pre-trained models for sentiment analysis have become increasingly popular due to their effectiveness in capturing nuances in sentiment across various domains (Pérez et al., 2021 ).. The "accelerate" library, designed for Python, focuses on optimizing and expediting computations, particularly within the realms of numerical and scientific computing. Moreover, the "emoji" library, also a Python library, augments the project with capabilities centered on "emoji" processing and management. This library enables tasks such as "emoji" detection, extraction, and manipulation within textual data. Its functionality extends to "emoji" identification, conversion between "emoji" and Unicode representations, and "emoji" visualization. Incorporating the "emoji" library is a sensible decision aimed at enhancing the project's capacity to adeptly manage emojis within the sphere of NLP endeavors, encompassing applications such as sentiment analysis, text classification, and text generation (carpedm20, 2015 ). The dataset dedicated to training purposes consists of 50,859 tweets related to Bitcoin. It is partitioned into an 80% training set and a 20% testing set previously. The testing dataset from the previous models, encompassing 10,174 tweets, is stored in a CSV file. This dataset will be employed to assess the accuracy of the Pysentimento model, ensuring uniformity across all three models under evaluation. (Fig. 10 ) and (Fig. 11 ) demonstrate the stages of Pysentimento model and the results of Pysentimento application on a portion of the testing dataset correspondingly. In (Fig. 8 ), the "label" column represents the classification of the training dataset whereas the "sentiment" is the one for Pysentimento. 6.5.1 Classification report of (Pysentimento) The presented classification report concerns the "Pysentimento" model's performance following its application to test data. This report evaluates the model's capabilities in classifying text-based sentiments into three categories: "positive," "negative," and "neutral." It is crucial to analyze each aspect of the report to gain a comprehensive understanding of the model's performance as conveyed in (Fig. 12 ). The provided classification report pertains to the performance of the "Pysentimento" model after its application to testing data. This report evaluates the model's ability to classify text-based sentiments into three categories: "positive," "negative," and "neutral." It is crucial to analyze each aspect of the report to gain a comprehensive understanding of the model's performance. When it comes to assessing the "positive" emotion category, the model performs admirably. For this category, the precision is 0.89, meaning that 89% of the feelings that the model predicted to be "positive" are indeed true. The recall for the "positive" category is equally impressive, with a score of 0.88, indicating that the model effectively captures 88% of actual positive sentiments present in the dataset. An F1-score of 0.88 in this situation indicates that the model is remarkably well-balanced, with a strong equilibrium between recall and precision. This equilibrium suggests that the model performs well in predicting "positive" feelings and in capturing the majority of genuine positive sentiments seen in the data. The model's precision score in the "negative" sentiment category is 0.83, which means that 83% of the feelings it identified as "negative" are actually "negative." However, the recall for the "negative" category is 0.69, demonstrating that the model only accurately classifies 69% of the real negative attitudes in the dataset. For the "negative" category, the F1-score, a measure of precision and recall equilibrated, is recorded as 0.76. The model appears to be operating in this category quite harmoniously, based on its F1-score of 0.76. With a precision of 0.92 in the "neutral" sentiment category, the "Pysentimento" model does exceptionally well, correctly predicting 92% of the "neutral" feelings among its predictions. Moreover, the recall for the "neutral" category is an impressive 0.97, highlighting the model's capacity to correctly classify 97% of actual neutral sentiments in the dataset. The F1-score for “neutral” is 0.94, denoting a balance between precision and recall. This score highlights the model’s ability to discern neutral sentiments. The “Pysentimento” model has an overall accuracy of 90%, demonstrating its competence in sentiment prediction across all categories. In conclusion, the "Pysentimento" model performs admirably when tested utilizing test data. It maintains a balanced F1-score, demonstrating its efficacy in capturing various sentiment categories, and demonstrates notable strengths in precision, especially in the "neutral" category. The model's overall 90% accuracy rating specifies how consistently it can predict "positive," "negative," and "neutral" sentiments. 7 The Second evaluation The researcher's choice of utilizing a second evaluation stems from a compelling need for rigorous accuracy validation. This imperativeness arises due to substantial doubts surrounding the observed accuracy levels of the three models. The core objective is to ascertain the authenticity of the attained accuracy, safeguarding against potential pitfalls like overfitting or underfitting, which may compromise the credibility of the results. Given the paramount prominence of this objective, the previously constructed models are saved and subsequently recalled to undergo a test against a dataset comprising 100 tweets of the randomly collected data regarding three cryptocurrencies: (BTC), (ETH), and (BNB), which represents unseen data manually categorized by the researcher into positive, negative, and neutral sentiments. (Fig. 13 ) unveils the results of the second evaluation. The performance metrics, including precision, recall, F1 rating, and accuracy, for the four distinct NLP models become conveyed. They are 'CNN-LSTM,' 'SVM (GloVe),' 'SVM (TF-IDF),' and 'Pysentimento.' Such metrics function as fundamental indicators of the models' proficiency in accurately identifying positive, negative, and neutral sentiments within a specified dataset. The precision values reveal observable modifications across every model under examination. These indices serve as indications of the percentage of correctly expected positive cases among all cases expected as positive. 'CNN-LSTM' uncovers a noteworthy precision of 90%, emphasizing its dexterity in productively categorizing positive sentiments with precision. In contrast, 'SVM (GloVe)' and 'SVM (TF-IDF)' demonstrate comparatively reduced precision values of 80% and 78%, respectively, indicating a relatively smaller accuracy in correctly distinguishing positive cases. Contrastingly, 'Pysentimento' separates itself with a remarkable precision of 97%, symptomatic of its heightened ability to outwit incorrect positives and discern positive sentiments with extraordinary accuracy. This discrepancy in precision values underscores the nuanced execution divergences among the models, thereby offering crucial insights into their efficacy in positive sentiment characterization. The recall values, which signify the proportion of actual positive instances properly recognized by the models, unveil intricate patterns that shed light on the models' efficacy in capturing positive sentiments. 'CNN-LSTM' achieves a commendable recall of 88%, affirming its ability to capture a considerable part of genuine positive sentiments within the dataset. In contrast, both 'SVM (GloVe)' and 'SVM (TF-IDF)' present dissimilar recall values of 95% and 75%, respectively. This delineates varied abilities in correctly identifying positive instances, with 'SVM (GloVe)' demonstrating a notably higher recall than 'SVM (TF-IDF).' Noteworthy is the exceptional execution of 'Pysentimento,' boasting a perfect recall of 100%, indicating its unparalleled capability to comprehensively identify all positive sentiments present in the dataset. This divergence in recall values among the models accentuates nuanced distinctions in their ability to effectively identify and capture actual positive sentiments, providing valuable insights into their respective strengths in positive sentiment recognition. The examination of the F1 rating provides a thorough assessment of the models accomplishment. A harmonic average derived from recall and precision is the F1 rating. 'CNN-LSTM' becomes distinguished by an F1 rating of 89%, indicating a tuneful balance between precision and recall, thereby proposing a well-rounded effectiveness. In contrast, 'SVM (GloVe)' and 'SVM (TF-IDF)' present F1 scores of 87% and 77%, respectively, accentuating nuanced variations in their precision-recall equilibrium. The heightened F1 rating of 98% accomplished by 'Pysentimento' attests to its exceptional proficiency in accomplishing a brilliant amalgamation of precision and recall, underscoring its intensified effectiveness in sentiment analysis tasks. 8 Comparative analysis Below is a comprehensive analysis of sentiment classification models, namely "CNN-LSTM," "SVM (GloVe)," "SVM (TF-IDF)," and "Pysentimento." The focus is on F1 score values, accuracy, precision, and recall in both the first and second assessments. Subtle variations in the models' output that provide insight into how well they perform in sentiment analysis applications will be revealed through careful inspection of these metrics. The precision values for sentiment classification models in the first and second evaluations exhibit slight variations. 'CNN-LSTM' obtains a precision of 0.90 in the first evaluation, but in the second evaluation, it slightly drops to 0.89. Conversely, "SVM (GloVe)" exhibits a more significant decline from 0.80 to 0.87, indicating an increase in its capacity to accurately detect pleasant emotions. The precision of "SVM (TF-IDF)" decreases from 0.78 to 0.77, suggesting a slight deterioration in its precise performance. In contrast, 'Pysentimento' displays consistency in both evaluations, with a precision value of 0.97, highlighting its ability to prevent false positives. Similarly, recall values for each of the sentiment categorization models disclose significant alterations between the two assessments. The recall of the "CNN-LSTM" model falls slightly from 0.88 to 0.87, suggesting a marginal decline in the model's ability to identify true positive cases. In contrast, "SVM (GloVe)" reveals a boost in its recall from 0.95 to 0.98, indicating an elevated degree of accuracy in identifying positive sentiments. "SVM (TF-IDF)" has a more substantial reduction in recall from 0.75 to 0.68, indicating a shift in its proficiency in recognizing positive instances. It is noteworthy that 'Pysentimento' consistently scores a perfect recall of 1.00 in both evaluations, demonstrating its unwavering ability to fully detect all positive sentiments in the dataset. The F1 score offers a comprehensive view of model performance as a well-balanced combination of recall and precision. With an F1 score of 0.89 in both assessments, "CNN-LSTM" performs consistently and represents an equitable trade-off between recall and precision. The F1 score of "SVM (GloVe)" slightly increases from 0.87 to 0.88, demonstrating a better equilibrium between recall and precision. However, "SVM (TF-IDF)" experiences a more significant decrease in F1 score, from 0.77 to 0.73, suggesting a change in the precision-recall balance. With a consistent F1 score of 0.98 in both tests, "Pysentimento" performs exceptionally well, demonstrating its ongoing ability to achieve a pleasing balance between recall and precision. The sentiment classification models' accuracy values reflect significant variations between the first and second assessments. In both evaluations, "CNN-LSTM" consistently scores an accuracy of 0.86, demonstrating stability of its overall prediction ability. From 0.82 to 0.81, "SVM (GloVe)" shows a slight decline, indicating a minor change in accuracy performance. SVM (TF-IDF) experiences a more notable decline from 0.79 to 0.71, signifying a notable change in its overall accuracy. Conversely, 'Pysentimento' has a remarkable consistency, exhibiting a constant accuracy of 0.92, highlighting its continuous ability to anticipate sentiments correctly in all classes. Potential overfitting or underfitting factors when examining performance fluctuations in sentiment analysis models such as "CNN-LSTM," "SVM (GloVe)," "SVM (TF-IDF)," and "Pysentimento" must be taken into account. When a model becomes overfitted to the training data, it captures irrelevant details and exhibits poorer performance on new, unseen data. This phenomenon is known as overfitting. Conversely, underfitting occurs when a model accomplishes poorly than expected across measurements due to its oversimplification. Overfitting may be indicated by abrupt declines in accuracy, recall, F1 score, or precision, while underfitting, which exposes the model's simplicity in capturing sentiment classifications, may be designated by persistent underperformance across metrics. In conclusion, subtle variances in the performance metrics of the sentiment analysis models "CNN-LSTM," "SVM (GloVe)," "SVM (TF-IDF)," and "Pysentimento" are exposed through comparing their analyses between the first and second evaluations. In both assessments, "CNN-LSTM" continuously maintains lofted values for accuracy, recall, F1 score, and precision, indicating a strong and reliable performance. Recall and F1 score improvements for "SVM (GloVe)" demonstrate the system's capacity to recognize positive sentiments. A number of measures indicate a possible shift in the precision-recall balance of 'SVM (TF-IDF)'. It is noteworthy that 'Pysentimento' consistently exhibits exceptional precision, recall, and F1 score, confirming its effectiveness regarding sentiment analysis accurate prediction and it will be employed on the data concerning the three cryptocurrencies. 9 Pysentimento and Google Trends This implementation of Pysentimento is preceded by a pre-processing step, which involves the removal of irrelevant tweets from the dataset pertaining to the three currencies. A pattern of words is designated and it serves the purpose of recognizing tweets that pertain to the context of cryptocurrencies, encompassing terms such as Bitcoin, Ethereum, and Binance, along with a multitude of associated hash tags and keywords. Tweets that deviate from this pattern, commonly referred to as "mismatches," are systematically identified and segregated, with a dedicated output file housing these mismatched tweets. As for Google Trends, (Fig. 14 ) illustrates total searches for (BTC), (ETH), (BNB) respectively in March, June and December, 2022. The total searches on Google Trends for the three cryptocurrencies, Bitcoin, Ethereum, and Binance Coin in 2022 demonstrate varying levels of attention cross the given months. In March, Bitcoin reached 2075 searches, Ethereum garnered 2155 searches, and Binance Coin observed 2621 searches. In June, Bitcoin declined to 1506 searches, Ethereum reduced to 1371 searches, and Binance Coin held a substantial interest with 2150 searches. December notably saw a significant increase in searches for Bitcoin and Ethereum, reaching 2558 searches for each, indicating heightened fascination in these cryptocurrencies toward the end of the year. In contrast, Binance Coin observed a decrease in searches to 1899. These search volume fluctuations imply dynamic shifts in public sentiment and interest regarding these cryptocurrencies over the course of the year 2022. The collective data reveal complex patterns in the online attention for these digital assets, potentially swayed by external elements such as market trends, regulatory developments, and overall market sentiment. After the application of Pysentimento to the data, the sentiments sum total is collected for the three previously designated cryptocurrencies and months in 2022. The opening and closing of last days for each month in addition to the trading volume of the three cryptocurrencies are provided using yahoo finance. The trading volume is the total quantity of a cryptocurrency that is traded across all exchanges over a specific timeframe. It exhibits the total quantity of cryptocurrency units that have been acquired and sold on the market during that period. Trading volume is a crucial metric in the cryptocurrency market since it offers insights into the market activity and liquidity. Higher trading volumes commonly indicate increased market interest, liquidity, and the potential for more accurate price discovery. On the contrary, lower trading volumes could suggest reduced interest, potentially leading to higher price volatility and less reliable price information (Bitsgap, 2021 ). (Table. 1) displays how the collected data are organized. Table 1 Collected data organization Sentiment Prices Month Coin Total Tweets % Positive % Negative % Neutral % Overall Sentiment % Opening ( $ ) Closing ( $ ) Volume (billions) Google index MAR-22 BNB 51.78 43.19 5.03 51.78 38.22 395.62 428.92 54 2621 JUN-22 BNB 57.92 33.79 8.32 57.92 42.08 320.47 219.30 39 2150 DEC-22 BNB 54.74 32.46 12.83 54.74 45.26 300.75 246.35 21 1899 MAR-22 BTC 57.02 33.25 9.78 57.02 42.98 43,194.50 45,538.68 831 2075 JUN-22 BTC 60.56 23.98 15.44 60.56 39.44 31,792.55 19,784.73 924 1506 DEC-22 BTC 57.66 29.68 12.66 57.66 42.34 17,168.00 16,547.50 541 2558 MAR-22 ETH 59.50 36.81 3.68 59.50 40.50 2,919.78 3,281.64 437 2155 JUN-22 ETH 71.07 23.47 5.33 71.07 28.93 1,942.05 1,067.30 562 1371 DEC-22 ETH 60.07 37.00 2.93 60.07 39.93 1,295.77 1,196.77 161 2558 The table provides a detailed overview of cryptocurrency data for March, June, and December 2022. The "Month" column categorizes data based on the recording month, allowing for a closer look at cryptocurrency performance during specific periods. The "Year" column adds chronological context, clarifying data for the year 2022. Each entry in the "Coin" column acts as a unique identifier, indicating the cryptocurrency (e.g., BNB-USD, BTC-USD, ETH-USD) and the recorded month. This simplifies the analysis process. Sentiment-related columns like "Positive %," "Negative %," "Neutral %," and "Overall Sentiment %" break down sentiment percentages for each cryptocurrency. The "Total Tweets%" column provides insights into social media activity. For sentiment analysis, the "Positive %," "Negative %," "Neutral %," and "Overall Sentiment %" columns are crucial, revealing the percentage of positive, negative, neutral sentiments, and an overall sentiment indicator. "Opening ( $ )" shows the opening price on the first day of the month, while "Closing ( $ )" indicates the closing price on the last day. "Volume (billions)" records the total trading volume, and the "Google index" reflects Google searches. Normalization is applied to several columns, including "Positive %," "Negative %," "Neutral %," "Overall Sentiment %," "Volume (billions)," and "Google index." This ensures comparability and meaningful analysis across different data scales. The dataset is a valuable resource for analysts, investors, and researchers, offering insights into cryptocurrency behavior, performance, and sentiment analysis. It enhances understanding in correlation and time series forecasting, providing valuable indicators of market sentiment. 10 Correlation analysis The Pearson correlation coefficient serves as an instrument for calculating the linear association between two variables. It assumes a pivotal role in the assessment and quantification of relationships existing amidst diverse predictor variables and target variables found within the dataset. Precisely, this statistical measure aids in determining the extent of correlation between two variables, signifying whether their association is robust or feeble. The output of the Pearson correlation coefficient is a numerical value that falls within the range of (-1 to 1), with distinct interpretations: First of all, a coefficient of 1 indicates a perfect positive linear relationship, implying that as one variable exhibits an increase, the other demonstrates a synchronous increase at a constant rate. Secondly, a coefficient of -1 signifies a perfect negative linear relationship, suggesting that as one variable experiences growth; the other exhibits a parallel and consistent decrease. Thirdly, when the coefficient assumes a value of 0, it conveys the absence of a linear relationship, denoting that the variables under consideration lack correlation. The calculation of Pearson correlation coefficients transpires between a multitude of predictor variables (such as "Normalized Closing of Last Day," "Normalized Volume," "Normalized SumTotal Google Trends") and target variables (including "Normalized Overall Sentiment," "Normalized Positive Sum," "Normalized Negative Sum," and "Normalized Neutral Sum"). The importance of computing Pearson connection coefficients in this specific setting stems from the need to comprehend how the anticipating variables interrelate with the target variables. Such an investigation demonstrates valuable in numerous regards: Fundamentally, it assists in recognizing anticipating - target relationships, offering clearness on which anticipating variables exert a huge impact on the target variables. After that, inside the area of predictive displaying and machine learning, understanding the connections between elements including anticipating variables and the target variable is of most extreme significance. Elements showing strong connections regularly emerge as likely parts for prescient models. Next, it adds to the creation of bits of knowledge identifying with the potential outcomes of changes in particular anticipating variables on the target variables. These bits of learning hold worth for educated dynamic producing and the elucidation of information examples. To realize such a purpose, a heatmap illustrating the correlation coefficients among various sentiment measures for three specific cryptocurrencies: BNB, BTC, and ETH. Within each cryptocurrency, sentiment measures encompass sub-measures related to predictor variables, while columns represent target variables in (Fig. 15 ). Such coefficients result in priceless insights into the interplay between sentiment and the three cryptocurrencies market indicators can be comprehended. As for (ETH), Normalized Overall Sentiment exhibits a correlation coefficient of (0.3773) with "Normalized Closing of Last Day." This indicates a correlation where Ethereum's overall sentiment tends to rise modestly with an increase in its closing price at the end of each month. Conversely, a substantial negative correlation (-0.8531) between "Normalized Volume" and Normalized Overall Sentiment implies that higher Ethereum trading volumes coincide with decreased overall sentiment. A robust positive correlation (0.9885) between "Normalized SumTotal Google Trends" and Normalized Overall Sentiment reveals that heightened Google search activity corresponds to increased overall sentiment for Ethereum. Normalized Positive Sum proves a moderate positive correlation (0.3811) with "Normalized Closing of Last Day," indicating that as Ethereum's closing price rises at month-end, positive sentiment similarly tends to increase, though modestly. "Normalized Volume" exhibits a strong negative correlation (-0.851) with Normalized Positive Sum, showing that increased Ethereum trading volumes coincide with a significant decrease in positive sentiment. There's a solid positive correlation (0.9879) between "Normalized SumTotal Google Trends" and Normalized Positive Sum, illustrating that increased Google search activity for Ethereum aligns with elevated positive sentiment. Weak negative correlation (-0.3507) observed between Normalized Negative Sum and "Normalized Closing of Last Day." There's a solid positive correlation (0.9879) between "Normalized SumTotal Google Trends" and Normalized Positive Sum, illustrating that increased Google search activity for Ethereum aligns with elevated positive sentiment. Weak negative correlation (-0.3507) observed between Normalized Negative Sum and "Normalized Closing of Last Day." In essence, as Ethereum's month-end closing price increases, negative sentiment slightly decreases, indicating minor shifts in sentiment with closing price. "Normalized Volume" and Normalized Negative Sum illustrate a notably strong correlation (0.8677), demonstrating that increased Ethereum trading volumes align with a significant rise in negative sentiment. "Normalized SumTotal Google Trends" and Normalized Negative Sum have a strong negative correlation (-0.9925), indicating that increased Google search activity for Ethereum aligns with a significant decrease in negative sentiment. "Normalized Closing of Last Day" and Normalized Neutral Sum exhibit a strong negative correlation (-0.8295). "Normalized Volume" and Normalized Neutral Sum show a less pronounced correlation (0.4174), implying a relatively weak positive relationship. In conclusion, the substantial influence of public interest on Ethereum's market dynamics is highlighted by the robust correlation between heightened Google search activity and diminished neutral sentiment. The nuanced nature of Ethereum's sentiment dynamics is emphasized by the varied impacts of predictor variables on different sentiment aspects. The significance of factors such as closing prices and trading volume in shaping market sentiment is suggested by positive and negative correlations, respectively. The strong association between Google search trends and sentiment implies the importance of information-seeking behavior. These collective findings indicate that monitoring Google Trendsdata, closing prices, and trading volume is imperative for insightful market analysis, enabling informed investment decisions and forecasting. Regarding (BTC), "Normalized Closing of Last Day" strongly correlates (0.867) with overall sentiment, signifying a noteworthy association with Bitcoin's closing price. "Normalized Volume" has a negative correlation (-0.1305) with "Normalized Overall Sentiment," suggesting that fluctuations in trading volume have a limited negative impact on Bitcoin's overall sentiment. Moreover, "Normalized SumTotal Google Trends" reveals a moderate positive correlation of (0.45) with "Normalized Overall Sentiment." Positive sentiment, particularly with "Normalized Closing of Last Day," shows a notably strong correlation of (0.8883). This robust connection implies that positive sentiment in the cryptocurrency market is highly responsive to Bitcoin's closing price, with a higher closing price corresponding to overwhelmingly positive sentiment. In opposition, "Normalized Volume" exhibits a weak negative correlation of (-0.0864) with "Normalized Positive Sum." Furthermore, "Normalized SumTotal Google Trends" proves a moderate positive correlation of (0.4099) with "Normalized Positive Sum." Negative sentiment, especially with "Normalized Closing of Last Day," exhibits a significant negative correlation of (-0.4232). A rise in closing price tends to reduce negative sentiment, emphasizing the pivotal influence of price performance (increases) on negative sentiment in the cryptocurrency market."Normalized Volume" shows a significant positive correlation of (0.6766) with "Normalized Negative Sum." "Normalized SumTotal Google Trends" shows a significant negative correlation of (-0.8814) with "Normalized Negative Sum." High Google search interest in Bitcoin corresponds to a notable decrease in negative sentiment. Neutral sentiment, particularly starting with "Normalized Closing of Last Day," displays a remarkably strong positive correlation of (0.9751). Additionally, "Normalized Volume" demonstrates a modest positive correlation of (0.1663) with "Normalized Neutral Sum." "Normalized SumTotal Google Trends" shows a similar modest positive correlation of (0.1679) with "Normalized Neutral Sum." To conclude, the correlation analysis stresses a consistent and substantial positive relationship between Bitcoin's closing price and all sentiment aspects, with correlation coefficients ranging from 0.867 to 0.9751. This implies that as Bitcoin's price increases, sentiment across positive, negative, and neutral aspects becomes notably more positive. In contrast, the impact of "Normalized Volume" on sentiment is relatively weak, suggesting a minor and inconsistent effect on overall sentiment and its components. Concerning (BNB), the coefficient between "Normalized Closing of Last Day" and "Normalized Overall Sentiment" is (0.9914), indicating a remarkably strong positive linear relationship. Likewise, the correlation coefficient between "Normalized Volume" and "Normalized Overall Sentiment" is (0.6927), demonstrating a positive correlation, though not as powerful as the closing price. The coefficient (0.8254) between "Normalized SumTotal Google Trends" and "Normalized Overall Sentiment" indicates a strong positive relationship. Increased Google search volume for Binance corresponds to a considerable rise in overall sentiment. The correlation coefficient (0.7434) between "Normalized Closing of Last Day" and "Normalized Positive Sum" suggests a moderately strong positive linear relationship. The correlation coefficient between "Normalized Volume" and "Normalized Positive Sum" is (0.1635), denoting a positive correlation, though relatively weak. The coefficient (0.3614) between "Normalized SumTotal Google Trends" and "Normalized Positive Sum" indicates a positive relationship. Higher Google search volume for Binance corresponds to an increase in positive sentiment. The correlation coefficient (-0.4831) between "Normalized Closing of Last Day" and "Normalized Negative Sum" suggests a moderate negative relation. An increase in Binance's closing price on the last day of the month corresponds to a noticeable decrease in negative sentiment. Similarly, a burly negative correlation is indicated by the coefficient (-0.9239) between "Normalized Volume" and "Normalized Negative Sum." A substantial decrease in negative sentiment towards Binance is robustly linked to an increase in the trading volume. The coefficient (-0.8263) between "Normalized SumTotal Google Trends" and "Normalized Negative Sum" specifies a firm negative relationship. Higher Google search volume for Binance correlates strongly with decreased negative sentiment. The correlation coefficient (-0.4831) between "Normalized Closing of Last Day" and "Normalized Negative Sum" suggests a moderate negative relationship. The coefficient (-0.9239) between "Normalized Volume" and "Normalized Negative Sum" conveys a solid negative correlation. Higher Binance trading volume is stoutly linked to a notable decrease in negative sentiment. The correlation coefficient (-0.8263) between "Normalized SumTotal Google Trends" and "Normalized Negative Sum" denotes a glaring negative relationship. In conclusion, the noticeable patterns across these correlations indicate that the closing price, trading volume, and Google search trends are influential factors in shaping sentiment towards Binance. While overall sentiment and positive sentiment display positive relationships with these predictors, negative and neutral sentiments exhibit negative or weaker associations. This comprehensive understanding of correlations provides valuable insights for stakeholders in Binance, aiding in strategic decision-making, market analysis, and sentiment trend predictions. In considering investment decisions amid the Russian-Ukrainian war, Bitcoin emerges as an appealing choice due to its positive correlations with market variables, providing stability sought by investors during geopolitical uncertainties. Binance Coin, with its mixed correlations, demands a nuanced approach, considering potential impacts on its distinctive dynamics. Ethereum, showing positive associations akin to Bitcoin, may attract those seeking diverse opportunities. The choice among these cryptocurrencies, amidst the complexities of the war, should align with investors' risk tolerance and goals, necessitating thorough research, consideration of additional factors, and staying informed within the dynamic cryptocurrency market. Price Forcasting analysis integration remains pertinent in navigating the evolving geopolitical landscape and making well-informed investment decisions. 11 Price forecasting (SARIMA) Cryptocurrency price forecasting has drawn escalating attention, as researchers and practitioners deploy advanced techniques to boost the meticulousness of predictions. The SARIMA (Seasonal Autoregressive Integrated Moving Average) library, a Python package that actualizes the SARIMA model for time series forecasting, is utilized for this purpose. Founded on the statsmodels library, which furnishes an extensive array of tools for statistical analysis in Python, the SARIMA library is instrumental in modeling and forecasting univariate time series data that exhibit non-stationarity (Brownlee, 2019; Aggarwal, 2020 ). The SARIMA model extends the ARIMA model, standing for Autoregressive Integrated Moving Average, and introduces a seasonal component to capture periodic fluctuations occurring at fixed intervals, such as daily, weekly, monthly, or yearly. SARIMA excels in capturing seasonal patterns, enhancing forecast precision. Addressing non-stationary data, it adeptly handles cyclical fluctuations and variations. SARIMA provides confidence intervals and diagnostic plots for forecasts, aiding in assessing uncertainty and reliability. Despite its benefits, SARIMA requires substantial historical data for accurate models and faces challenges in parameter determination and computational intensity (Brownlee, 2019; Hyndman & Athanasopoulos, 2021). (Fig. 16 ) clarifies the predictions based on the prices spinning March, June, and December 2022. The prices in (Table. 1) are utilized in the training The figure illustrates price predictions for three cryptocurrencies—Ethereum (ETH), Bitcoin (BTC), and Binance Coin (BNB)—spanning from January to March 2023. Two lines are depicted: the blue line denotes actual prices, and the orange dotted line reflects predicted prices. Predictive outcomes are delineated for each cryptocurrency, encompassing Ethereum (ETH), Binance Coin (BNB), and Bitcoin (BTC) as shown above. These predictions offer valuable insights into anticipated price trends. For Ethereum, the model forecasts $ 1727.21 on January 31, 2023, $ 1603.38 on February 28, 2023, and $ 1632.29 on March 31, 2023. Binance Coin is expected to reach $ 286.71 on January 31, 2023, $ 299.61 on February 28, 2023, and $ 303.72 on March 31, 2023. Bitcoin's predicted prices are $ 20,739.52 on January 31, 2023, $ 20,630.29 on February 28, 2023, and $ 20,633.13 on March 31, 2023. These projections offer a detailed insight into expected price movements, derived from the SARIMA model's analysis of historical data and temporal patterns. To gauge prediction accuracy, the Root Mean Squared Error (RMSE) metric is utilized. It is applied to the model's predictions to compare them with the actual prices for the three cryptocurrencies in 2023 in (Fig. 17 ). The aforementioned image comprises three line graphs, each delineating the price trends of distinct cryptocurrencies over a specified duration. The blue and orange lines in each graph represent the actual and predicted prices respectively. The first graph illustrates the Root Mean Square Error (RMSE) for Binance Coin (BNB) over January, February, and March 2023. The second graph, following a similar pattern, represents the RMSE for Ethereum (ETH) during the same period. The third graph, meanwhile, depicts the RMSE for Bitcoin (BTC). The divergence between the actual and predicted prices in each graph signifies the prediction error. A greater divergence implies a higher RMSE, indicating a less accurate prediction, while a smaller divergence suggests a lower RMSE, indicative of a more accurate prediction. The model predicts BTC closing prices at $ 20,739.52, $ 20,630.29, and $ 20,633.13 for January 31, 2023, February 28, 2023, and March 31, 2023, respectively. BNB forecasts are $ 286.71, $ 299.61, and $ 303.72, while ETH projections are $ 1,727.21, $ 1,603.38, and $ 1,632.29 during the same period. Actual closing prices for BTC, BNB, and ETH on these dates are $ 22,840.14, $ 23,522.87, $ 28,033.56, $ 307.07, $ 304.86, $ 316.57, $ 1,567.33, $ 1,634.33, and $ 1,792.74, respectively. RMSE values, assessing predictive accuracy, are 4745.03 for BTC, 131.99 for ETH, and 14.22 for BNB. Elevated BTC RMSE indicates challenges capturing its complex price patterns. ETH's lower RMSE suggests better predictive performance, while BNB's notably low RMSE indicates accurate predictions, hinting at more foreseeable price movements. In summary, the distinct RMSE values underscore the variability in predicting different cryptocurrencies with the SARIMA model. The intricate field of predicting cryptocurrency prices requires continuous examination and adjustments to align with the ever-changing dynamics of the market. This suggests that the predictability of each cryptocurrency differs, emphasizing the ongoing need to fine-tune model parameters for reliable predictions. Given the geopolitical uncertainties introduced by the Russian-Ukrainian war, accurate cryptocurrency price forecasting becomes increasingly crucial. While the SARIMA model holds promise, its effectiveness hinges on continuous adaptation to the unique characteristics of each cryptocurrency. Persistent efforts to improve forecasting techniques are essential, particularly in response to geopolitical shifts and market fluctuations linked to the ongoing conflict. Declarations *Funding: This research is not supported by any institution or foundation *Conflicts of interest/Competing interests - (N/A) *Ethics approval - (N/A) *Consent to participate - (N/A) *Consent for publication - (N/A) *Availability of data and material: (The following link provides access to the a Kaggle website where first model employed in my research along with the first dataset, which consists of 50,859 tweets that have been categorized as ['positive'], ['negative'], and ['neutral']. This dataset is obtained from the Kaggle website, which is a renowned platform and online community for data scientists and machine learning practitioners. The link is mentioned in the references, and it has the following in-text citation in data collection section in my research (Dundee, 2002 ). Bitcoin Tweets Sentiment Analysis: cnn-lstm | Kaggle *Code availability: codes cannot be shared openly, to protect the author’s privacy due to the fact that some of such codes are directly involved in my Ph.D. thesis, which has not yet been defended. *Authors' contributions: As for the roles of the authors, they are provided as follows: 1-M.R. authored the manuscript and conducted the research experiments and analysis. 2-M.G. provided consultation on machine learning and deep learning approaches, guided the experimental methods and analysis, and reviewed the coding. 3-M.K. and S.A. served as research supervisors. References Abraham, J., Higdon, D., Nelson, J., & Ibarra, J. (2018). Cryptocurrency price prediction using tweet volumes and sentiment analysis. SMU Data Science Review, 1(3), 1. Aggarwal, P. K. (2020). Powerful learning is all about retrieval. ASCD Education Update, 62(1), 1-5. https://www.ascd.org/el/articles/powerful-learning-is-all-about-retrieval Akhtar, W., Kumaraguru, P., and Joshi, A. (2019). Sentiment analysis for cryptocurrencies using roBERTa transformer model with self-attention mechanism. In Proceedings of the Third Workshop on Blockchain Technologies and Applications (pp. 1-10). AL-MANSOUR, B. Y. (2020). Cryptocurrency market: Behavioral finance perspective. The Journal of Asian Finance, Economics and Business , 7 (12), 159–168. https://doi.org/10.13106/jafeb.2020.vol7.no12.159 Balasudarsun, N. L., Ghosh, B., & Mahendran, S. (2022). Impact of negative tweets on diverse assets during stressful events: An investigation through time-varying connectedness. Journal of Risk and Financial Management , 15 (6), 260. https://doi.org/10.3390/jrfm15060260 Bird, S., Klein, E., & Loper, E. (2009). Tokenization for Natural Language Processing. In Natural Language Processing with Python (pp. 39-57). O'Reilly Media, Inc. Bitsgap. (2021, February 9). What is a trading volume in cryptocurrency and why is it important? https://bitsgap.com/blog/what-is-a-trading-volume-in-cryptocurrency-and-why-is-it-important Bollen, J., Mao, H., & Zeng, X. (2011). Twitter mood predicts the stock market. Journal of Computational Science , 2 (1), 1–8. https://doi.org/10.1016/j.jocs.2010.12.007 Carpedm20. (2015). Emoji: Emoji for Python (Version 1.7.0) [Software]. Retrieved from https://github.com/carpedm20/emoji Carrillo, J., Carrillo, A., & Carrillo, A. (2020). Pysentimiento: A Python Toolkit for Sentiment Analysis and Social NLP Tasks. In Proceedings of the 28th International Conference on Computational Linguistics: System Demonstrations. Chen, T., Li, Z., Zhang, Y., and Li, T. (2018). Neural sentiment classification with user and product attention. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (pp. 1650-1659). Cortes, C., & Vapnik, V. N. (1995). Support-vector networks. Machine learning, 20(3), 273-297. Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of NAACL-HLT 2019 (pp. 4171-4186). Donahue, J., Anne Hendricks, L., Guadarrama, S., Venugopalan, S., Saenko, K., & Darrell, T. (2015). Long-term recurrent convolutional networks for visual recognition and description. IEEE transactions on pattern analysis and machine intelligence, 39(4), 677-691. https://doi.org/10.1109/TPAMI.2015.2500292. Drus, Z., & Khalid, H. (2019). Sentiment analysis in social media and its application: Systematic literature review. Procedia Computer Science , 161 , 707-714. Dundee. (2002). Bitcoin Tweets Sentiment Analysis: cnn-lstm. Kaggle . https://www.kaggle.com/code/dundee2002/bitcoin-tweets-sentiment-analysis-GloVe-cnn-lstm/log Faret, J., & Reitan, J. (2015). Twitter Sentiment Analysis-Exploring the Effects of Linguistic Negation (Master's thesis, NTNU). Fonseca, J. (2020). PyTrends (Version 4.9.1) [Computer software]. GitHub. https://github.com/GeneralMills/PyTrends Gaber, M., Ezzat, M., & Mokhtar, M. (2021). Hyperparameter Optimization for Deep Learning-Based Sentiment Analysis. IEEE Access, 9, 78030-78047. García-Pablos, A., Cuadros, M., Rigau, G., and Agirre, E. (2020). Pysentimiento: A Python Toolkit for Sentiment Analysis and Social NLP Tasks. In Proceedings of The 12th Language Resources and Evaluation Conference (pp. 7215-7223). Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780 Hsu, C.-W., Chang, C.-C., & Lin, C.-J. (2003). A practical guide to support vector classification. National Taiwan University. Hugging Face. (2021). Transformers: State-of-the-art Natural Language Processing for Pytorch and TensorFlow 2.0. Retrieved May 28, 2021, from https://huggingface.co/transformers/ Hyunyoung, C., & Varian, H. (2018). Replicating “predicting the present with google trends” by Hyunyoung Choi and Hal Varian (the economic record, 2012). Economics, 12(1). https://doi.org/10.5018/economics-ejournal.ja.2018-34 Johnson, R., Smith, T., Williams, K., & Davis, M. (2020). Using Google Trendsdata to explore public interest in breast cancer screening. BMC Public Health, 20(1), 1-6. Kim, Y. (2014). Convolutional neural networks for sentence classification. arXiv preprint arXiv:1408.5882. LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324. Liu, X., Zhang, P., Liu, L., & Zhou, G. (2018). Empower sequence labeling with task-aware neural language model. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 1234-1243). Liu, Z., Lin, W., Shi, Y., & Zhao, J. (2021). A robustly optimized Bert pre-training approach with post-training. Lecture Notes in Computer Science , 471–484. https://doi.org/10.1007/978-3-030-84186-7_31 Medhat, W., Hassan, A., & Korashy, H. (2014). Sentiment analysis algorithms and applications: A survey. Ain Shams Engineering Journal, 5(4), 1093-1113. Mikolov T., Sutskever I., Chen K., Corrado G.S., Dean J. (2013) Distributed Representations of Words and Phrases and their Compositionality. In: Advances in Neural Information Processing Systems. ownlee, J. (2020). How to develop LSTM models for time series forecasting. Machine Learning Mastery. https://machinelearningmastery.com/how-to-develop-lstm-models-for-time-series-forecasting/ Pano, T., & Kashef, R. (2020). A complete vader-based sentiment analysis of Bitcoin (BTC) tweets during the era of covid-19. Big Data and Cognitive Computing , 4 (4), 33. https://doi.org/10.3390/bdcc4040033 Pennington, J., Socher, R., & Manning, C. D. (2014). GloVe: Global vectors for word representation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP) (pp. 1532-1543). Pérez, J. M., Furman, D. A., Alonso Alemany, L., & Luque, F. M. (2022). RoBERTuito: a pre-trained language model for social media text in Spanish. Proceedings of the Thirteenth Language Resources and Evaluation Conference. Pérez, J. M., Giudici, J. C., & Luque, F. (2021). pysentimiento: A Python toolkit for sentiment analysis and social NLP tasks. arXiv preprint arXiv:2106.09462 Ramos, J. (2003). Using tf-idf to determine word relevance in document queries. In Proceedings of the first instructional conference on machine learning (Vol. 242, pp. 133-142). Russell, M. A. (2018). Mining the Social Web: Data Mining Facebook, Twitter, LinkedIn, Instagram, GitHub, and More (3rd ed.). O'Reilly Media, Inc. Sattarov, O., Jeon, H. S., Oh, R., & Lee, J. D. (2020). Forecasting bitcoin price fluctuation by Twitter sentiment analysis. 2020 International Conference on Information Science and Communications Technologies (ICISCT) . https://doi.org/10.1109/icisct50599.2020.9351527 Schölkopf, B., Platt, J. C., Shawe-Taylor, J., Smola, A. J., & Williamson, R. C. (2001). Estimating the dimension of a kernel subspace. In Proceedings of the Fourteenth Annual Conference on Computational Learning Theory (COLT'01), ACM. Sharma, N., Khosla, A., Kim, T., Gade, A., & Pagh, R. (2023). TF-IDF: A fundamental technique in natural language processing. GitHub Repository. https://github.com/GeneralMills/PyTrends. Sharma, P., & Sharma, D. (2022). Classification Reports: Essential Tools for Sentiment Analysis Model Evaluation. arXiv preprint arXiv:2208.03906. Sutskever, I., Vinyals, O., & Le, Q. V. (2014). Sequence to sequence learning with neural networks. In Advances in neural information processing systems (pp. 3104-3112). Valencia, F., Gómez-Espinosa, A., & Valdés-Aguirre, B. (2019). Price movement prediction of cryptocurrencies using sentiment analysis and machine learning. Entropy , 21 (6), 589. https://doi.org/10.3390/e21060589 Wang, S., & Manning, C. D. (2012). Baselines and bigrams: Simple, good sentiment and topic classification. In Proceedings of the 50th annual meeting of the association for computational linguistics: Short papers-volume 2 (pp. 90-94). Wang, Y., Sun, Y., Liu, T., & Huang, X. (2016). A CNN-LSTM based model for text classification. arXiv preprint Wołk, K. (2020). Advanced social media sentiment analysis for short‐term cryptocurrency price prediction. Expert Systems , 37 (2), e12493. Yahoo! (n.d.). Yahoo Finance - Stock Market Live, quotes, Business & Finance News . Yahoo! Finance. https://finance.yahoo.com/ Yamashita, R., Nishida, Y., Kido, R., & Akita, K. (2018). Convolutional neural networks: an overview and applications in medical image analysis. In Medical Imaging Informatics (pp. 449-483). Springer, Cham. Zhang, X., Lai, L., Xu, C., & Liu, J. (2015). CNN-LSTM neural networks for sentence-level sentiment classification. arXiv preprint arXiv:1511.05352. Zhang, Y., Wallace, B., & Wang, D. (2016). Rationale-augmented convolutional neural networks for text classification. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing (pp. 795-804). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3949248","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":272574659,"identity":"2d27384e-a6b6-4d9c-9e55-ff07266b8da7","order_by":0,"name":"Muhammad Nabil Rateb","email":"data:image/png;base64,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","orcid":"","institution":"Alexandria University","correspondingAuthor":true,"prefix":"","firstName":"Muhammad","middleName":"Nabil","lastName":"Rateb","suffix":""},{"id":272574660,"identity":"83df11d8-9c1e-4222-a924-4fc708cc5fab","order_by":1,"name":"Sameh Alansary","email":"","orcid":"","institution":"Alexandria University","correspondingAuthor":false,"prefix":"","firstName":"Sameh","middleName":"","lastName":"Alansary","suffix":""},{"id":272574661,"identity":"1f9662cb-6b66-4188-a1ee-6e9f02de4413","order_by":2,"name":"Marwa Khamis Elzouka","email":"","orcid":"","institution":"Alexandria University","correspondingAuthor":false,"prefix":"","firstName":"Marwa","middleName":"Khamis","lastName":"Elzouka","suffix":""},{"id":272574662,"identity":"08b8f4ea-edff-43c4-8936-31736d874642","order_by":3,"name":"Mohamad Galal","email":"","orcid":"","institution":"Nile University","correspondingAuthor":false,"prefix":"","firstName":"Mohamad","middleName":"","lastName":"Galal","suffix":""}],"badges":[],"createdAt":"2024-02-11 19:59:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3949248/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3949248/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":51104088,"identity":"20cd13bb-765d-4332-831d-150713899b8c","added_by":"auto","created_at":"2024-02-14 07:28:49","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":25147,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of sentiment labels\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/9f557303d52197d8dc737914.png"},{"id":51104396,"identity":"59553ff9-8cd3-47fe-8992-f1836d4509ca","added_by":"auto","created_at":"2024-02-14 07:36:49","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":171482,"visible":true,"origin":"","legend":"\u003cp\u003ePre-processing with text length and word count\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/c6f93ea391863faa153af48b.png"},{"id":51104395,"identity":"3085d493-4515-489e-b5f4-7e40a77f4499","added_by":"auto","created_at":"2024-02-14 07:36:49","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":318649,"visible":true,"origin":"","legend":"\u003cp\u003eWord cloud\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/619ec2281d7487576214df6c.png"},{"id":51104086,"identity":"f1325daf-8d0f-4a06-8392-39afb36e140c","added_by":"auto","created_at":"2024-02-14 07:28:49","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":24955,"visible":true,"origin":"","legend":"\u003cp\u003eCNN-LSTM sequential model components\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/1a7a4f2a03676545b0861bfc.png"},{"id":51104091,"identity":"04f31ea8-1b4f-476c-9997-29521cfcd043","added_by":"auto","created_at":"2024-02-14 07:28:49","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":38327,"visible":true,"origin":"","legend":"\u003cp\u003eCNN-LSTM classification report\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/eee5db3d7bee6d4e7c9b53c4.png"},{"id":51104390,"identity":"adf81cab-8ec3-4305-b7fe-4ee7f11e6c26","added_by":"auto","created_at":"2024-02-14 07:36:49","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":44073,"visible":true,"origin":"","legend":"\u003cp\u003eTweets after\u003cstrong\u003e \u003c/strong\u003eSVM preprocessing\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/b6c8f4f8395d0c29b44e3abc.png"},{"id":51104084,"identity":"2c11173a-a6b3-4655-9cce-0fc914c2ca4a","added_by":"auto","created_at":"2024-02-14 07:28:49","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":20498,"visible":true,"origin":"","legend":"\u003cp\u003eBest Hyperparameters for SVM\u003cstrong\u003e \u003c/strong\u003ewith GloVe and TF-IDF\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/b40ddb5d7e47d9b7c0053098.png"},{"id":51104392,"identity":"30e0afcf-ec26-4bc1-b021-70fe0cc12236","added_by":"auto","created_at":"2024-02-14 07:36:49","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":33722,"visible":true,"origin":"","legend":"\u003cp\u003eClassification report of (SVM-GloVe)\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/62e8a2a855a07cb1f3ccae05.png"},{"id":51104736,"identity":"3ecf052d-ce47-4d42-83e3-c883e03bbd16","added_by":"auto","created_at":"2024-02-14 07:44:49","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":75977,"visible":true,"origin":"","legend":"\u003cp\u003eClassification report of (SVM-TF-IDF)\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/307da8b1e93ee703c532bf09.png"},{"id":51104087,"identity":"beac1cbd-94c5-4eab-95af-f6101cb69edf","added_by":"auto","created_at":"2024-02-14 07:28:49","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":61477,"visible":true,"origin":"","legend":"\u003cp\u003ePysentimento model stages\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/66a2ad8acd9796751f03330f.png"},{"id":51104397,"identity":"c938c5d1-7b15-4524-9331-13e98331f0d9","added_by":"auto","created_at":"2024-02-14 07:36:50","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":257841,"visible":true,"origin":"","legend":"\u003cp\u003ePysentimento application results\u003c/p\u003e","description":"","filename":"11.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/ffae00affdba9e3e4e96dc18.png"},{"id":51104393,"identity":"418c0c10-938f-402f-89de-17bdeb208322","added_by":"auto","created_at":"2024-02-14 07:36:49","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":76425,"visible":true,"origin":"","legend":"\u003cp\u003ePysentimento classification report\u003c/p\u003e","description":"","filename":"12.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/a6e7aca4c14401102db1e577.png"},{"id":51104095,"identity":"7a5c619b-2f75-426d-93af-fbfb62d140ed","added_by":"auto","created_at":"2024-02-14 07:28:49","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":47855,"visible":true,"origin":"","legend":"\u003cp\u003eSecond evaluation scores\u003c/p\u003e","description":"","filename":"13.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/a9d908569c1f58e7b3279290.png"},{"id":51104391,"identity":"21b53308-a419-41cf-8e53-6996b1871373","added_by":"auto","created_at":"2024-02-14 07:36:49","extension":"png","order_by":14,"title":"Figure 14","display":"","copyAsset":false,"role":"figure","size":46960,"visible":true,"origin":"","legend":"\u003cp\u003eTotal searches of Google Trends\u003c/p\u003e","description":"","filename":"14.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/e8f718f75d57e2033c1756df.png"},{"id":51104097,"identity":"5b50f5e5-da71-482a-aa31-e6a13ad850d0","added_by":"auto","created_at":"2024-02-14 07:28:49","extension":"png","order_by":15,"title":"Figure 15","display":"","copyAsset":false,"role":"figure","size":74935,"visible":true,"origin":"","legend":"\u003cp\u003eHeatmap of correlation coefficients\u003c/p\u003e","description":"","filename":"15.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/a425292374b1e151c1699fed.png"},{"id":51104098,"identity":"8ee63737-fe47-4d73-b0d5-0dc8bd1434f8","added_by":"auto","created_at":"2024-02-14 07:28:50","extension":"png","order_by":16,"title":"Figure 16","display":"","copyAsset":false,"role":"figure","size":76424,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted vs. historical prices\u003c/p\u003e","description":"","filename":"16.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/fbcae936094e3933eb5c7b29.png"},{"id":51104099,"identity":"1d33612f-fd53-4137-91f3-df1264e64d1b","added_by":"auto","created_at":"2024-02-14 07:28:50","extension":"png","order_by":17,"title":"Figure 17","display":"","copyAsset":false,"role":"figure","size":123565,"visible":true,"origin":"","legend":"\u003cp\u003eReal vs. predicted prices RMSE\u003c/p\u003e","description":"","filename":"17.png","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/fd05465437b7b50f178244d6.png"},{"id":51354970,"identity":"30271921-8400-4093-a5cb-de5bc8142e80","added_by":"auto","created_at":"2024-02-20 06:30:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1841828,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3949248/v1/9a2c3980-b108-4ded-8b71-ef213952189f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Predicting Cryptocurrency Prices during Periods of Conflict: A Comparative Sentiment Analysis Approach Using SVM, CNN-LSTM, and PySentimento","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eCryptocurrency forecasting has become an increasingly prevalent research topic, with sentiment analysis being one of the key techniques implemented to predict cryptocurrency prices. Political events have been shown to significantly impact cryptocurrency prices and sentiment analysis of social media. Google Trends data is an emphatic method of predicting these fluctuations. The Russian-Ukrainian War represents a substantial occurrence that potentially influences the oscillations of cryptocurrency prices.\u003c/p\u003e \u003cp\u003eSentiment analysis of social media and Google Trends data necessitates the employment of computational methods to identify and extract subjective information, such as opinions and attitudes, from these sources of data. In the context of cryptocurrency price forecasting, sentiment analysis can postulate valuable apprehension regarding how the masses perceive a particular cryptocurrency or the cryptocurrency market as a whole. By analyzing the sentiment expressed in social media and Google Trends data, it is possible to gain a better understanding of how the market may behave in the short or long-term (Wołk, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). However, there are challenges to conducting sentiment analysis of social media and Google Trends data for cryptocurrency price forecasting. These include the need for accurate and representative datasets, the difficulty in accurately classifying sentiment, and the potential for bias in the data. Despite these challenges, sentiment analysis of social media and Google Trends data remains a promising approach for improving the accuracy of cryptocurrency price forecasting (Sattarov et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBehavioral finance refers to the study of how psychological biases and emotions influence financial decision-making. In relation to cryptocurrency price forecasting, sentiment analysis of social media and Google Trends data can provide insight into the emotions and psychological biases of cryptocurrency market participants. For example, social media sentiment analysis can reveal how investors feel about a particular cryptocurrency and how sentiment fluctuates over time. Google Trends data can uncover search trends related to a particular cryptocurrency, which can offer insights into the level of interest and attention the cryptocurrency is receiving (AL-Mansour, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCryptocurrency prices are highly volatile and can be influenced by various factors, including global events. During times of crisis, such as the Russian-Ukrainian War or the COVID-19 pandemic, traditional market indicators may not produce an accurate reflection of the market sentiment. This is where sentiment analysis of social media and Google Trends data becomes even more significant, as it can capture the real-time sentiment of market participants. This information can be exploited by traders, investors, and analysts to make informed decisions about their investments (Pano \u0026amp; Kashef, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePresently, countless subscribers use social media platforms such as Twitter (presently X), Instagram, and Facebook to share their thoughts about their daily lives and express their emotions. X as a social media platform, is distinguished with a number of privileges to be chosen for this research. First of all, according to Matthew Woodward (2022), globally, there are 436\u0026nbsp;million active Twitter subscribers per month. Secondly, Twitter has varied access accounts using application programming interfaces (APIs). The Twitter API is a collection of programmatic endpoints that can be exploited to comprehend or create the Twitter discourse and scrap tweets (Faret \u0026amp; Reitan, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Finally, Drus and Khalid (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) maintain that the majority of social media subscribers worldwide use Facebook. However, due to the unstructured nature of the data, its poor organization, frequent usage of short forms, and high rate of spelling errors, sentiment analysis is not commonly conducted using these data, but rather more frequently by means of X.\u003c/p\u003e \u003cp\u003eIn the context of the Russian-Ukrainian War, sentiment analysis of social media and Google Trends data can reveal how geopolitical tensions are impacting the cryptocurrency market. For instance, if social media sentiment around Bitcoin is overwhelmingly positive during the war, it may be an indication that investors could to it as a safe-haven asset. Conversely, if sentiment is negative, it may indicate that investors are selling off their cryptocurrency holdings and moving to more conventional haven assets such as gold or the US dollar (Balasudarsun et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis paper is divided into five sections. The first section is dedicated to the objectives and research questions. The second is related to the review of the literature on sentiment in relation to cryptocurrencies using different implementations for different purposes. The third discusses the methodology and procedures utilized in carrying out this study along with the comparison encompassing SVM, CNN-LSTM and Pysentimento. The fourth is concerned with the analysis and discussion of the findings of the study. The final one provides a conclusion of the study and its findings along with implications for further research.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"2 Objectives and research questions","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe present study proposes to investigate how sentiment analysis could be utilized to predict the prices and interest in the cryptocurrency market even during times of war. Employing the NLP tool referred to as sentiment analysis, Twitter (presently X) sentiment, and Google trends, this research will focus on mining and mensuration of the subjective emotions and opinions in the chosen tweets regarding Bitcoin, Ethereum, and Binance. The ultimate goal of this study is to conduct sentiment analysis on the collected tweets to ascertain if the sentiment is typically positive, negative or neutral with regard to the designated cryptocurrencies. Furthermore, Convectional Machine Learning Techniques such as Support Vector Machine (SVM), Deep Learning (Deep Neural Network) Techniques such as Convolutional Neural Networks and Long Short-Term Memory (CNN_LSTM), and finally Transformer Techniques (pre-trained), for example, Pysentimento are compared to determine which training algorithm will produce the most accurate results.\u003c/p\u003e \u003cp\u003eJust as with X, the data are compiled from Google Trends; an open-source tool that displays a single search term's frequency in relation to global search volume over a specified time period in which a user logs. The volume index and closing of last day prices of (March, June and December 2022) are collected. After that,, light will be shed on realizing the tendency toward the three types of cryptocurrencies: Bitcoin, Ethereum, and Binance coin with maximum possible accuracy. Between a multitude of predictor variables (such as \u0026ldquo;Normalized Closing of Last Day,\u0026rdquo; \u0026ldquo;Normalized Volume,\u0026rdquo; \u0026ldquo;Normalized Sum Total Google Trends\u0026rdquo;) and target variables (including \u0026ldquo;Normalized Overall Sentiment,\u0026rdquo; \u0026ldquo;Normalized Positive Sum,\u0026rdquo; \u0026ldquo;Normalized Negative Sum,\u0026rdquo; and \u0026ldquo;Normalized Neutral Sum\u0026rdquo;), Pearson correlation coefficients are calculated. Another objective of the study is to provide insights into the strength and orientation of the associations that these variables share and to provide a recommendation concerning which of the chosen currencies can be a profitable investment and value storage during the time of war. In an attempt to predict the prices of the three cryptocurrencies during the first three months of 2023, the SARIMA model is employed. Finally, to reveal the accuracy of SARIMA predictions, root mean squared error (RMSE) is calculated.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"3 Research questions","content":"\u003cp\u003e(A) Can sentiment analysis on tweets and Google Trends produce accurate predictions about the emotional tendencies and prices of Bitcoin, Ethereum, and Binance coin?\u003c/p\u003e \u003cp\u003e(B) What are the differences between SVM, CNN-LSTM, and Pysentimento? Which method yields the best level of accuracy?\u003c/p\u003e \u003cp\u003e(C) If predictions based on social media are possible, do they correlate with the trend of the actual interest of the chosen cryptocurrency retrieved from Google Trends?\u003c/p\u003e"},{"header":"4 Review of literature","content":"\u003cp\u003eDespite the growing research interest in sentiment analysis of cryptocurrency price prediction in recent years, both Pysentimento as sentiment analysis classifier and Binance\u003c/p\u003e \u003cp\u003eCoin has been paid less attention to by other studies, possibly for being relatively recent. As far as sentiment analysis for cryptocurrency price forecasting is concerned, there are various studies with variant objectives. The present review covers them as they highlight the significant role of sentiment analysis in predicting cryptocurrency and stock market prices.\u003c/p\u003e \u003c/p\u003e \u003cp\u003eAbraham (2018) employs a sentiment analysis of Twitter to predict cryptocurrency prices. The data is analyzed to determine whether they can useful information to the final model. VADER sentiment analysis finds tweets to be more neutral, which reduces the accuracy of the results if the general public\u0026rsquo;s sentiment is neutral because neutral sentiment usually does not indicate a buying or selling trend. The ratio of tweets and Google Trends are both strongly related to prices. A linear regression technique is used to estimate the final daily price of Bitcoin. Twitter sentiment on cryptocurrency is often positive despite possible price changes. Future research should use a more complex model than linear regression to improve the results because the data for this publication is collected when prices are only increasing.\u003c/p\u003e \u003cp\u003eAnother study has conducted specific research on Google Trends data, concludes that basic seasonal autoregressive models that incorporates Google Trends data as input data surpassed other models which do not involve Google Trends data by 5\u0026ndash;20% (Hyunyoung \u0026amp; Varian, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). However, it is challenging for such models to explain situations where the direction changes. In these situations, Google Trends data can be beneficial.\u003c/p\u003e \u003cp\u003eA different study introduces a system that forecasts the prices of four chosen cryptocurrencies\u0026mdash;Bitcoin, Ethereum, Ripple, and Litecoin\u0026mdash;using social media data and machine learning methods. Random forests, support vector machines, and neural networks (NNs) are applied in this model. The results of this model demonstrate the applicability of sentiment analysis and machine learning methods for cryptocurrency prediction. Furthermore, the prices of certain cryptocurrencies, typically those with a substantial follow-up, can be anticipated solely by Twitter data (Valencia et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eXin Huang proposes an LSTM model for sentiment analysis. Sina-Weibo, a famous social networking platform in China, provides the data to measure the sentiment. A recurrent neural network based on long short-term memory (LSTM) and historical cryptocurrency values is used to predict the price trend for cryptocurrencies. The results show an accuracy rate of 87%. This result is 15.4% better than that of the currently used traditional auto-regression method (Liu et al, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e"},{"header":"5 Research Methodology","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e5.1 Data collection\u003c/h2\u003e \u003cp\u003eA number of data collection methods are dedicated to collecting and analyzing the chosen tweets and Google Trends. These methods include Twitter API, sentiment analysis, and Google Trends. To answer the research questions, the previous methods shall be employed and explained in the following section. The primary data source for this research resides in an academic Twitter Application Programming Interface (API) account, which facilitates the scraping of tweets over three entire months (March, June, and December 2022) pertaining to keywords such as \"Bitcoin,\" \"Ethereum,\" and \"Binance.\" The Twitter API, an assortment of programmatic endpoints, empowers the comprehension and generation of Twitter discourse (Faret \u0026amp; Reitan, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Via establishing an academic API account, the researcher secures access to ten million tweets per month, ensuring an ample supply of requisite data. Concurrently, employing the same keywords and time frame, Google Trends data is collected utilizing the PyTrends library in Python (Fonseca, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), while Tweepy serves as the library for Twitter data scraping (Russell, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTwo distinct datasets are utilized for training and evaluation purposes. The first dataset consists of 50,859 tweets that have been categorized as ['positive'], ['negative'], and ['neutral']. This dataset is obtained from the Kaggle website, which is a renowned platform and online community for data scientists and machine learning practitioners (Dundee, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). For the secondary assessment, a distinct dataset is employed, encompassing 100 annotated tweets categorized as positive, negative, and neutral. These tweets are drawn randomly from the researcher's collection of tweets pertaining to the three cryptocurrencies. Additionally, historical price data for the three cryptocurrencies (Bitcoin, Ethereum, and Binance) is obtained from Yahoo Finance, a comprehensive platform offering cryptocurrency market information encompassing market capitalization, trading volume, price trends, and news. It also provides an API that facilitates access to historical and real-time cryptocurrency data, with price data collected for the same periods as the tweets (March, June, and December 2022) along with the real prices for the first three months of 2023 (January, February, and March) for the sake of price forecasting (Yahoo, n.d.).\u003c/p\u003e \u003cp\u003eGoogle Trends, in turn, serves as a tool to gauge the relative popularity of specific search phrases relative to others over time (Johnson et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). These data provide insights into the popularity of cryptocurrencies in 2022, aligning with the price charts and sentiment analysis results for Bitcoin, Ethereum, and Binance. Tweets are further filtered based on language (English) and location (worldwide), with the total number of tweets collected ranging from 700,000 to over a million for each month. The Google Trends data is accessed through the Google Trends API, which offers insights into the relative popularity of Bitcoin, Ethereum, and Binance for the same periods as the tweets (March, June, and December 2022).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e5.2 Support Victor Machine (SVM)\u003c/h2\u003e \u003cp\u003eAccording to Cortes \u0026amp; Vapnik (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e1995\u003c/span\u003e), Support Vector Machines (SVMs) are a type of supervised machine learning algorithm that can be used for both classification and regression tasks. SVMs work by finding a hyperplane that separates the data points of different classes with the maximum margin. The data points that are closest to the hyperplane are called support vectors, and they determine the position and orientation of the hyperplane. SVMs can handle both linear and nonlinear problems by using different kernel functions. The chosen model for this study is (SVM) with (GloVe) and Term Frequency-Inverse Document Frequency (TF-IDF). GloVe is a word embedding technique capturing semantic and syntactic information in vector spaces. These word embeddings serve as numerical representations of words for machine learning algorithms, capturing word meanings, contexts, similarities, analogies, and polarities (Pennington, Socher, \u0026amp; Manning, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAccording to Ramos (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), TF-IDF is a statistical measure evaluating the importance of a word within a document or corpus. TF-IDF assigns weights to words based on their frequency in a document and rarity across the corpus. Higher TF-IDF scores indicate greater word relevance to the document. TF-IDF is utilized in this study to reduce word vector dimensionality and eliminate noise and stopwords. In SVM, classification into positive, negative, or neutral categories is accomplished by identifying the hyperplane that best separates data points into different classes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e5.3 Advantages and disadvantages of (SVM)\u003c/h2\u003e \u003cp\u003eOne notable advantage of SVMs is their ability to attain high accuracy and exhibit strong generalization performance by identifying a hyperplane that effectively separates data points into distinct classes while maximizing the margin between them (Cortes \u0026amp; Vapnik, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e1995\u003c/span\u003e). This unique attribute enables SVMs to mitigate the issues of overfitting and underfitting often encountered by other machine learning algorithms like neural networks or decision trees. In the context of sentiment analysis within the cryptocurrency domain, characterized by its volatile and unpredictable data, SVM's ability to provide a reliable and robust model becomes crucial for addressing the inherent uncertainty and variability of the data (Bollen et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). SVMs effectively classify text into positive, negative, or neutral categories based on emotional tone; accommodate various text lengths and formats; and employ features such as word counts, (TF-IDF), word embeddings, or n-grams. These methods can synergize with other techniques, such as feature selection, dimensionality reduction, or ensemble methods, to enhance performance and accuracy (Wang \u0026amp; Manning, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Medhat et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite their strengths, SVMs face certain limitations within the domain of cryptocurrency sentiment analysis. Primarily, their inherent binary classification nature necessitates extensions for multiclass sentiment classification, often required in cryptocurrency discussions characterized by multifaceted sentiment landscapes. The computational intensity of SVMs poses challenges when dealing with large volumes of real-time cryptocurrency data, where timeliness is crucial (Sch\u0026ouml;lkopf et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). Additionally, tuning hyperparameters and kernel functions can be time-consuming and resource intensive, complicating SVM implementation (Hsu et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). SVMs' sensitivity to different kernel functions and hyperparameter settings introduces variability in performance, necessitating extensive trial-and-error approaches or grid searches to optimize results (Medhat et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e5.4 CNN-LSTM\u003c/h2\u003e \u003cp\u003eIn the present research, a hybrid model that combines Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM), originally adapted from Kaggle (Dundee, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2002\u003c/span\u003e), has been employed with the primary aim of enhancing the accuracy of outcomes. For a comprehensive exploration of the CNN component, it is imperative to delve deeper into the characteristics of Convolutional Neural Networks (CNNs). They are a category of deep learning algorithms specifically designed to address intricate tasks, including image recognition and natural language processing (LeCun et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e1998\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe fundamental operation of CNNs involves the application of filters or kernels to input data, which can encompass images or textual content, thereby facilitating the extraction of localized features or discernible patterns. Notably, these filters or kernels are acquired through network learning during the training phase, and they possess the capacity to encapsulate diverse aspects or dimensions of the data, encompassing attributes such as edges, shapes, colors, linguistic constructs, n-grams, and sentiment expressions (Kim, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). It is pertinent to mention that CNNs may additionally employ pooling layers for the purpose of diminishing data intricacy and dimensionality, alongside fully connected layers to cater to classification or regression objectives (Yamashita et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWithin this integrated framework, CNNs assume the role of extracting localized features or patterns, while LSTMs specializes in capturing temporal dynamics and protracted dependencies inherent in the data (Hochreiter \u0026amp; Schmidhuber, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). This harmonious fusion of layers imparts to CNN-LSTM the ability to deliver superior performance and address more intricate challenges, surpassing the efficacy of deploying each layer in isolation (Wang et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The CNNs outputs a vector encompassing all these features. LSTM then processes this vector to generate a sentiment score. LSTM learns to utilize the vector's features to compute a numerical sentiment value for the tweet, taking into account the tweet's context and tone. The final outcome is a sentiment score for each tweet, reflecting its positivity or negativity toward a particular cryptocurrency. A high sentiment score suggests optimism about the cryptocurrency, potentially indicating an expected price increase, while a low score signifies pessimism, possibly signaling a price decrease (Zhang et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e5.5 CNN-LSTM privileges and drawbacks\u003c/h2\u003e \u003cp\u003eCNN-LSTM has some merits and demerits that affect its performance and applicability for sentiment analysis of cryptocurrency. As for the first, CNN-LSTM can capture both local and global features along with patterns from the data, while other models may only capture one or the other. This means that CNN-LSTM can handle more diverse and heterogeneous data, and handle more fine-grained and nuanced sentiments (Liu et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). It can handle variable-length inputs and outputs, which is useful for dealing with data of different lengths and formats. This means that CNN-LSTM can process texts of different word counts, languages, or styles, while other models may require fixed-length inputs or outputs (Sutskever et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). It can be trained end-to-end, which means that it can learn the optimal parameters for both the convolutional and recurrent layers without requiring any manual feature engineering or preprocessing. This approach can reduce the complexity and cost of the model development and deployment.\u003c/p\u003e \u003cp\u003eRegarding the latter, CNN-LSTM requires considerable training data and computational resources to learn effectively, as it has many parameters and operations to optimize (Donahue et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). This means that CNN-LSTM can be slow and expensive to train and test and may need many trails and errors or grid search to find the best settings (Medhat et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). It may not be able to capture the long-range dependencies and context information in the data, as it relies on fixed-size filters or kernels for the convolutional layers. This means that CNN-LSTM may miss some important clues or signals that affect the sentiment of the text, such as the word order, the sentence structure, or the discourse relations (Zhang et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e5.6 Pysentimento\u003c/h2\u003e \u003cp\u003ePysentimento, a Python library designed for accessibility, provides a user-friendly avenue for conducting sentiment analysis and various social (NLP) tasks. The library achieves this by harnessing the capabilities of models such as BERT and others. Within Pysentimento, users can effortlessly load pretrained sentiment analysis models, including BERT, DistilBERT, RoBERTa, XLM-RoBERTa, or custom models tailored to specific domains or languages. Pysentimento's versatility extends to the application of these models on text inputs or data frames, generating sentiment scores that serve as quantitative indicators of the textual content's positivity or negativity. Carrillo et al. (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) developed Pysentimento, making it conveniently accessible on both GitHub and PyPI.\u003c/p\u003e \u003cp\u003eIn the field of sentiment analysis, Pysentimento opts for BERT as its principal model for conducting evaluations in both English and Spanish. The toolkit extends its capabilities by incorporating models that have undergone fine-tuning to address the specific requirements of distinct datasets or languages. A notable example is Robertuito, a specialized variant within the Pysentimento toolkit crafted explicitly for the analysis of sentiment in Spanish tweets sourced from social media. This adaptation is rooted in RoBERTa, an improved version of BERT distinguished by augmented data and extended training iterations. P\u0026eacute;rez et al. (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) introduced Robertuito to enhance sentiment analysis within the Pysentimento framework. This comprehensive toolkit is designed to provide users with the capability to perform sentiment analysis on their proprietary data or make use of alternative models available in Pysentimento or the Hugging Face Transformers library. In doing so, a versatile and adaptable solution is established (Garc\u0026iacute;a-Pablos et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Devlin et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Pysentimento has been evaluated on several datasets and benchmarks for different languages and tasks. According to its authors, it achieves state-of-the-art results for sentiment analysis in Spanish and English, and competitive results for other languages and tasks. For example, for sentiment analysis in Spanish, Pysentimento achieves an accuracy of 91.3% on the TASS 2017 dataset, which is a collection of tweets annotated with positive, negative, or neutral labels. For sentiment analysis in English, Pysentimento achieves an accuracy of 90.9% on the SemEval 2017 Task 4A dataset, which is a similar collection of tweets (P\u0026eacute;rez et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e5.7 Pysentimento strengths and weaknesses\u003c/h2\u003e \u003cp\u003eIn terms of merits, the utilization of BERT models within Pysentimento enables high accuracy and robustness for sentiment analysis tasks, surpassing previous models on various natural language processing (NLP) benchmarks and tasks (Devlin et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Furthermore, Pysentimento offers models that are fine-tuned on specific datasets or languages, thereby enhancing performance and adaptability (Garc\u0026iacute;a-Pablos et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This approach is particularly advantageous as sentiment analysis can aid in predicting cryptocurrency trends by capturing public sentiment and opinions, shedding light on the emotions and attitudes expressed in cryptocurrency-related text data (Chen et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Akhtar et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Additionally, Pysentimento leverages established and well-maintained NLP libraries and frameworks, such as the Hugging Face Transformers library, which supports numerous models and frameworks. Moreover, it utilizes popular libraries like pandas, numpy, and scikit-learn, ensuring reliability and widespread usage (Hugging Face, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Pysentimento does have some demerits. Firstly, users are required to install Pysentimento and its dependencies, which may necessitate considerable time and storage space, depending on the user's system and internet connection. Furthermore, compatibility issues or errors may be encountered during the installation process (Garc\u0026iacute;a-Pablos et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Hugging Face, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"6 Analysis and discussion","content":"\u003cp\u003eThe primary aim of this section is to conduct a comparative analysis of the performance of (SVMs (LSTMs), and the Pysentimento framework utilizing BERT for sentiment analysis within the domain of cryptocurrencies. Based on accuracy level, the best model is elected for sentiment analysis. Next, Google trends, volume, and the results of sentiment analysis (positive, negative and neutral) are combined in a table and normalized after obtaining overall sentiment sum.\u003c/p\u003e \u003cp\u003eAfter that, Pearson correlation coefficient are applied and time series prediction implementing SARIMA model is conducted to unveil which cryptocurrency is the safest for investment during times of geopolitical tension.\u003c/p\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e6.1 Data preprocessing and training of (CNN-LSTM)\u003c/h2\u003e \u003cp\u003eThe training and testing dataset is comprised of 50,859 tweets on Bitcoin that have been categorized as ['positive'], ['negative'], and ['neutral']. This dataset has been obtained from the Kaggle website, which is a renowned platform and online community for data scientists and machine learning practitioners (Dundee, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). Eighty percent of the data are devoted to training and the remaining portion is dedicated for testing where the training set consists of (40687) tweets, while the testing set comprises (10172) tweets. A larger training set allows the model to learn from a substantial amount of data, capturing patterns and relationships in the tweets that are essential for making accurate predictions. The bar plot indicates that 22937 tweets are labeled as positive, 21939 as neutral and 5983 as negative. This exploratory analysis offers an initial understanding of sentiment distribution prior to model development. (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) conveys such a distribution.\u003c/p\u003e \u003cp\u003ePreprocessing initiates with the column \"tweet\" turning into \"clean_tweet\" after preprocessing and the column \"label\" is the one for sentiments. The word count is a common feature employed in text analysis tasks to assess the complexity and content of text data whereas text length column quantifies the length of each tweet in terms of character count. This metric can be valuable for examining the distribution of tweet lengths within the dataset. The mean and standard deviation enable the researcher to determine the correct average of tweet length employing Freedman-Diaconis rule, which was 30.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe training and testing shapes become (40687, 30) (10172, 30). (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) shows the data after preprocessing with both metrics while (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) uncovers the word cloud, a powerful visual representation of frequently occurring words in text data, generated at this stage since it offers a visual summary, highlighting the most prevalent words within the dataset. It provides insights into the dataset's underlying themes and characteristics. The epochs\u0026rsquo; number is 5 and the batch size is 128.\u003c/p\u003e \u003cp\u003eThe model in (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) undergoes five processing stages. To begin with, an embedding layer maps each of 30 input tokens to a 200 dimensional vector, forming a three dimensional tensor of shape (30, 200). It involves two crucial parameters: maximum features and embedding dimensions. These parameters grant the model flexibility in encoding word information, making it adaptable to the dataset's specific The model in (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) undergoes five processing stages. To begin with, an embedding layer maps each of 30 input tokens to a 200 dimensional vector, forming a three dimensional tensor of shape (30, 200). It involves two crucial parameters: maximum features and embedding dimensions. These parameters grant the model flexibility in encoding word information, making it adaptable to the dataset's specific characteristics. Subsequently, the architecture incorporates two convolutional 1D layers, which introduce a spatial understanding of the text data.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThese layers employ a set of 1D convolutional filters, alike to sliding windows, to detect local patterns and feature representations within the tweet sequences. Throughout utilizing filters with varying receptive field sizes, the model becomes proficient at recognizing both fine-grained details and broader textual features. The Rectified Linear Unit (ReLU) activation function adds a critical element of non-linearity, enabling the model to capture complex relationships between words. Following the convolutional layers, two MaxPooling 1D layers act as a dimensionality reduction mechanism. These layers systematically downsample the output from the convolutional layers preserving the most salient and informative features while mitigating computational complexity. This process allows the model to concentrate on the most relevant elements of the data, enhancing its efficiency and focus.\u003c/p\u003e \u003cp\u003eThe neural network architecture further evolves with the inclusion of an LSTM layer as it excels at capturing sequential dependencies and long-range contextual information within text data. This layer is vital in modeling the temporal dynamics of tweet sequences, understanding how words relate to each other over time, and capturing intricate patterns that might span across the entire sequence. Finally, the dense layer maps the LSTM output to a three dimensional vector using linear activation, forming a tensor of shape (3) where the softmax function transforms the model's internal representations into probability distributions across the three sentiment classes: negative, neutral, and positive (Gaber et al, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e6.2 Classification report of (CNN-LSTM)\u003c/h2\u003e \u003cp\u003eBased on Sharma and Sharma (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), classification reports are essential instruments for assessing the efficacy of models intended to infer attitudes (positive, negative, or neutral) from textual data. These reports provide crucial metrics, each of which provides unique information on how well the model works. The first indicator, precision, measures how well positive predictions are made in comparison to false positives, or how frequently the model accurately predicts positive feelings. The other one, recall, gauges how well the model can recognize real-world positive examples, demonstrating its accuracy in expressing good emotions. Eventually, the third metric, the F1-score, finds a middle ground between recall and accuracy. This middle ground is especially helpful when the distribution of emotion classes is not uniform. (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) uncovers CNN-LSTM classification report.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe results demonstrate the ability of the proposed method to classify tweets into three sentiment categories: positive (pos), negative (neg), and neutral (neu). As for negative (neg), the precision for the negative sentiment class is approximately 0.92. This indicates that when the model predicts a tweet as negative, it is correct proximately 92% of the time. The recall, at around 0.95, demonstrates that the model captures about 95% of the actual negative tweets. The F1-score, which harmonizes these metrics, is approximately 0.93, suggesting a strong balance between precision and recall for the negative sentiment class. This means that the model excels in identifying and correctly classifying tweets expressing negative sentiments.\u003c/p\u003e \u003cp\u003eRegarding the neutral sentiment class, the precision is roughly 0.98, indicating that the model's predictions of neutrality are highly accurate. The recall, at approximately 0.97, signifies that the model correctly identifies about 97% of the actual neutral tweets. The F1-score of about 0.98 reaffirms the model's exceptional performance in classifying neutral sentiment, with a balanced combination of precision and recall. This signifies the model's proficiency in distinguishing neutral tweets from others.\u003c/p\u003e \u003cp\u003eThe positive sentiment class exhibits similar excellence, with a precision of approximately 0.98, indicating highly accurate positive predictions. The recall, around 0.98, indicates that the model captures about 98% of the actual positive tweets. The F1-score, approximately 0.98, reflects the strong balance between precision and recall for the positive sentiment class. This underscores the model's ability to effectively identify and classify positive sentiment in tweets. The overall model performance is impressive, with an accuracy of approximately 97%. This accuracy demonstrates the model's proficiency in classifying tweets across all sentiment categories.\u003c/p\u003e \u003cp\u003eThe macro-average F1-score, at about 0.96, signifies that the model maintains a robust balance between precision and recall for all sentiment classes, considering their individual support levels. Additionally, the weighted average F1-score, also around 0.97, indicates the model's consistency in performance across different sentiment classes, considering their varying proportions in the dataset.\u003c/p\u003e \u003cp\u003eTo conclude, the model demonstrates high proficiency in classifying tweets into positive, negative, and neutral sentiment categories, with an overall accuracy of 97%. It upholds robust balanced performance across sentiment classes, as indicated by macro and weighted average F1-scores of 0.96 and 0.97 respectively. The results emphasize the model's effectiveness in sentiment analysis of social media texts.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e6.3 Data preprocessing and training of (SVM)\u003c/h2\u003e \u003cp\u003eAll the steps of this phase are similar to those in the previous model apart from minor differences. First of all, tokenization breaks the text into individual words or tokens, enhancing the model's ability to comprehend and analyze the content. Secondly, stop-words removal is essential for eliminating common but uninformative words. Thirdly, lemmatization reduces words to their base or root forms, ensuring consistency in the data. Variations like \"running\" and \"ran\" are both transformed to \"run,\" streamlining feature extraction and pattern recognition. Finally, Part-of-Speech Tagging (POS) enriches the data by assigning grammatical labels to each word, allowing for the capture of specific linguistic patterns. This step can aid in discerning verbs, adjectives, or other parts of speech relevant to sentiment analysis.\u003c/p\u003e \u003cp\u003eFor this model, the Hugging Face 200-dimensional GloVe feature is imported, and it has been widely implemented in NLP tasks. The 200 dimensions of the word vectors refer to the fact that each word is represented by a vector of 200 numerical values to provide a rich and informative representation of words. The feature works by converting words into a continuous vector space, where the similarity between words can be measured using cosine similarity (Pennington et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOn the positive side, this feature is built upon a vast and diverse corpus of tweets, enabling it to effectively capture informal, colloquial language, slang, and even emoticons commonly found in social media conversations and tweets. It excels at capturing both global and local information from the corpus, including word frequency, word order, and word context. Additionally, the model can unveil intriguing linear relationships between words, such as analogies, antonyms, and synonyms. However, there are certain restrictions associated with the Hugging Face GloVe embeddings. Firstly, its performance is contingent upon the vocabulary size and coverage of the underlying corpus, which may not encompass rare or domain-specific (cryptocurrency) words, potentially limiting its applicability in specialized contexts. Secondly, the feature may struggle to capture intricate and nonlinear relationships between words, including polysemy, homonymy, irony, and sarcasm. Lastly, it may encounter difficulties in handling out-of-vocabulary words or common typographical errors commonly encountered in the informal language of tweets. Understanding these merits and demerits is crucial when considering the application of the Hugging Face GloVe embeddings 200d in various natural language processing tasks (Pennington et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eConversely, (TF-IDF) is a fundamental technique in (NLP). Being part and parcel of this model, TF-IDF addresses a critical challenge in NLP which is how to represent the inherent information and nuances of text data in a way that algorithms can effectively process (Sharma et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). TF-IDF vectorization offers several advantages when compared to alternative text representation methods like bag-of-words or word embeddings.\u003c/p\u003e \u003cp\u003eFirst of all, it addresses the issue of high dimensionality by considering only words that appear in at least one document within the dataset. TF-IDF aids in solving the obstacle of having too many different words to deal with by only looking at words that appear in at least one document in the dataset. This reduction in dimensionality makes it computationally efficient and manageable. Secondly, TF-IDF captures both local and global information about words. It accounts for word frequency within a document and across all documents in the dataset, providing a holistic view of word importance. Thirdly, TF-IDF assigns higher weights to words that carry more informative or distinctive characteristics for a document or topic, while assigning lower weights to common or generic words. This feature is particularly valuable for highlighting the significance of words in context. Finally, TF-IDF is known for its simplicity in implementation and interpretation, as it does not necessitate complex mathematical operations or external resources (Mikolov et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBased on Bird et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2009\u003c/span\u003e), there exist certain limitations and challenges associated with TF-IDF vectorization. To begin with, TF-IDF assumes that words are independent of each other, disregarding their order and context within a document. This can lead to the loss of valuable sequential information. Moreover, TF-IDF does not capture semantic or syntactic relationships between words, such as synonyms, antonyms, or grammatical structures, which limit its ability to understand the deeper meaning of language. Additionally, TF-IDF may assign low weights to words that are relevant but infrequent across all documents, such as proper nouns or domain-specific terms. Eventually, TF-IDF can be sensitive to outliers or noisy data, including spelling errors or typos. To mitigate some of these limitations, the researcher employs complementary techniques like stemming, lemmatization, NLTK stop-word removal, and adding (pos) to tokens with TF-IDF.\u003c/p\u003e \u003cp\u003eThese enhancements improve its overall performance and accuracy, making it a versatile choice for text analysis tasks and that has already been put into practice at the first step of pre-processing. (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) uncovers the tweets after preprocessing steps executed. The input data shapes are examined, with train GloVe at 40,687 samples by 200 features and test GloVe at 10,172 by 200, encoding tweets in a semantic space. Meanwhile, train TF-IDF and test TF-IDF have the same sample counts and 9991 features, representing tweets high-dimensionally. This conveys the input data shapes of GloVeand TF-IDF. GridSearchCV is a tool that performs an exhaustive search over specified parameter values for an estimator using cross-validation. Cross-validation is a technique that splits the data into k folds, utilizes one-fold as the test set and the rest as the training set, and repeats this process k times, averaging the results. This approach can provide a more reliable estimate of the performance of the estimator on unseen data.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThree-fold cross-validation is employed to evaluate each combination of parameter values. This conveys that data will be split into three parts, and each part is used as a test set once, while the other two parts are employed as a training set. The average score across the three folds is used as the performance metric for each combination. The parameters that are tuned are as follows: C, which is the regularization parameter for SVM that controls the trade-off between margin maximization and error minimization; and kernel, which is the kernel function for SVM that determines the type of transformation applied to the data.\u003c/p\u003e \u003cp\u003eAfter that, the researcher designates 'all' as the value for k, indicating the selection of all features by SelectKBest. The 'SVM C' parameter, which represents several levels of regularization strength, is set to a range of values, including 0.1, 1, and 10, in the SVM classifier. Furthermore, two kernel options \"linear\" and \"RBF\" (Radial Basis Function) are defined for the 'SVM kernel' parameter, which affects how the SVM maps input data. The first one finds a linear hyperplane that separates the data points based on their features (positive, negative neutral). The second maps the data points into a higher-dimensional space where a linear hyperplane can separate them better than in the original space. The grid search process evaluates the model's performance in a systematic manner across different sets of these hyperparameter values in order to determine which set best improves model performance. The results are illustrated in (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe figure conveys that utilizing \"all\" features, setting C to 10, and employing the RBF kernel provide the best performance for SVM with GloVeand TF_IDF features on the data. C is a hyperparameter for the SVM classifier that controls the trade-off between margin maximization and error minimization. A higher value of C means that the classifier tries to fit the data more closely, but it may also overfit and generalizes poorly. A lower value of C means that the classifier will allow more errors, but it may also underfit and miss important patterns. C\u0026thinsp;=\u0026thinsp;10 means that the value is a reasonably high one of C for this SVM classifier with GloVeand TF_IDF features. The RBF kernel is frequently implemented for capturing non-linear relationships in the data, which is an indicative of the data's complexity and non-linearity. The model achieves accuracy scores of approximately 91.94% and 88.22% with GloVeand TF-IDF respectively. Such scores indicate how well the model can classify tweets into their respective sentiment categories. Higher scores normally indicate better model performance.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e6.4 Classification reports of (SVM)\u003c/h2\u003e \u003cdiv id=\"Sec18\" class=\"Section3\"\u003e \u003ch2\u003e6.4.1 Classification report of (SVM-GloVe)\u003c/h2\u003e \u003cp\u003eAs revealed in (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e), with regard to sentiment analysis, the classification report thoroughly provides a thorough summary of the performance data obtained from an established Support Vector Machine (SVM) model that makes use of GloVe features. This assessment methodology includes support, recall, F1-score, and accuracy measures that are specific to the three sentiment categories (positive, neutral, and negative). Central measures of the model's prediction accuracy for every sentiment class are precision values, which are shown as noteworthy percentages. For the negative sentiment category, the accuracy is 91.83%, for the neutral sentiment category, it is 94.46%, and for the positive sentiment category as 91.25%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe recall percentages surface as 95.17% for negative sentiments, 93.54% for neutral sentiments, and 82.39% for positive sentiments. The accuracy, a pivotal metric gauging correctness in classifying cases crosswise all sentiment subclasses, is 92.96%. This overarching accuracy metric provides a holistic viewpoint on the model's effectiveness in sentiment examination. Macro and weighted averages, integral portions of the report, offer nuanced appraisals accounting for stabilizing and circulation of sentiment categories. The macro-averaged precision, recall, and F1-score communicate as 92.51%, 90.37%, and 91.35%, respectively. Conversely, the weighted-averaged precision, recall, and F1-score arise at 92.97%, 92.96%, and 92.92%, correspondingly. These averages offer a more comprehensive assessment regarding the impact of class imbalances on the level of performance held by the model.\u003c/p\u003e \u003cp\u003eTo summarize, the classification report delivers comprehensive and illuminating information about the SVM model's performance, providing subtle insights into its accuracy, recall, and F1-score for various sentiment categories. While macro and weighted averages offer a comprehensive picture of the model's performance in sentiment analysis tasks, the support metric and overall accuracy add more context. This thorough examination is essential for identifying the model's advantages and possible areas for improvement, enabling well-informed sentiment analysis decision-making.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section3\"\u003e \u003ch2\u003e6.4.2 Classification report of (SVM-TF-IDF)\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe deployed (SVM) model's sentiment analysis performance is painstakingly evaluated in the classification report in (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e). The assessment is carried out via TF-IDF characteristics, a commonly employed method for determining a word's relevance inside a collection of documents. The report provides a detailed overview of the discriminative skills of the model by covering precision, recall, F1-score, and support metrics for each sentiment category (positive, neutral, and negative).\u003c/p\u003e \u003cp\u003eThe precision values signify the degree of accuracy with which the model predicts each sentiment class. In the present case, the accuracy of the negative sentiment class is 86.54%, indicating that the model is capable of correctly classifying negative attitudes. The model's accuracy in predicting neutral attitudes is demonstrated by the neutral sentiment class's precision of 94.47%, whereas the positive sentiment class's precision is 90.70%.\u003c/p\u003e \u003cp\u003eMetrics for recall offer valuable information on how well the model captures examples from each sentiment category. Recall for the negative class is 95.63%, highlighting the model's capacity to accurately identify a significant percentage of real negative cases. With an 88.49% recall rate, the neutral class demonstrates a strong sensitivity to real-world neutral mood occurrences. Nevertheless, the recall of the positive class is 78.14%, indicating a somewhat lower capture of real positive cases.\u003c/p\u003e \u003cp\u003eF1-scores, which exhibit a harmonic mean of recall and precision, shed more light on the model's well-balanced performance. With an F1-score of 90.85%, the negative emotion class exhibits a well-balanced trade-off between recall and precision. Likewise, the class representing neutral sentiments has an F1-score of 91.38%, signifying a well-balanced approach to forecasting neutral sentiments. On the other hand, the F1-score of 83.95% for the positive sentiment class indicates a trade-off between recall and precision for positive sentiments.\u003c/p\u003e \u003cp\u003eThe SVM model with TF-IDF features has an overall accuracy of 90.35%, which provides a global indicator of how accurate it is at categorizing instances in all sentiment categories. Weighted averages and macro statistics offer a thorough analysis that takes the impact of class disparities into account. The macro-averages for recall, F1-score, and precision are 90.57%, 87.42%, and 88.73%, in that order. On the contrary, the F1-score, weighted-averaged precision, and recall are 90.35%, 90.65%, and 90.70 percent, correspondingly.\u003c/p\u003e \u003cp\u003eTo conclude, this thorough assessment provides a detailed overview of the SVM model's advantages and disadvantages in sentiment analysis tasks, offering insightful information that can be used to develop the model and make well-informed decisions.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe comparative study of SVM model that employs both of TF-IDF and GloVe characteristics has several implications. The TF-IDF feature is regularly outperformed by the SVM model combined with GloVe embeddings in terms of accuracy, precision, and F1-score for a range of attitudes. This consistency implies that GloVe embeddings play a major role in building a more sophisticated and reliable sentiment analysis model due to their capacity to capture complex semantic information.\u003c/p\u003e \u003cp\u003eBoth features have a high level of competence when it comes to neutral emotions, demonstrating their ability to navigate through a wide range of expressions in this area. The F1-scores accurately reflect the subtle nature of neutral sentiment expressions, which forces models to strike a careful balance between recall and precision. With the highest recall in this category, the SVM model which utilizes TF-IDF features demonstrates a unique ability to efficiently capture occurrences of negative sentiment efficiently. This discovery suggests that TF-IDF could be vastly advantageous in recognizing manifestations of negative emotion, providing insightful information, especially in applications where detecting negative sentiments is critical. These findings highlight the significance of contemplating the intricacies of sentiment expressions and carefully evaluating the trade-offs between precision and recall when selecting a sentiment analysis model.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e6.5 Pysentimento\u003c/h2\u003e \u003cp\u003eThis particular model does necessitate neither a preprocessing nor a training phase, as it is a pre-trained model equipped with a tokenizer. The process commences with the installation of the Pysentimento library (version 0.7.2) and its prerequisites. It procures the following dependencies: \"accelerate\" library (version 0.22.0), \"datasets\" (version 2.14.5), and \"emoji\" (version 1.7.0). The model architecture, \"robertuito,\" suggests that this model is based on the RoBERTa architecture. RoBERTa is a variant of the BERT (Bidirectional Encoder Representations from Transformers) model and is known for its effectiveness in a wide range of NLP tasks. The model is fine-tuned specifically for sentiment analysis. Fine-tuning involves training a pre-existing model on a task-specific dataset. In this case, the model has been trained on a dataset of text samples with associated sentiment labels (positive, negative, and neutral). This fine-tuning process supports the model to learn the patterns and features relevant to sentiment analysis. Pre-trained models for sentiment analysis have become increasingly popular due to their effectiveness in capturing nuances in sentiment across various domains (P\u0026eacute;rez et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)..\u003c/p\u003e \u003cp\u003eThe \"accelerate\" library, designed for Python, focuses on optimizing and expediting computations, particularly within the realms of numerical and scientific computing. Moreover, the \"emoji\" library, also a Python library, augments the project with capabilities centered on \"emoji\" processing and management. This library enables tasks such as \"emoji\" detection, extraction, and manipulation within textual data. Its functionality extends to \"emoji\" identification, conversion between \"emoji\" and Unicode representations, and \"emoji\" visualization. Incorporating the \"emoji\" library is a sensible decision aimed at enhancing the project's capacity to adeptly manage emojis within the sphere of NLP endeavors, encompassing applications such as sentiment analysis, text classification, and text generation (carpedm20, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). The dataset dedicated to training purposes consists of 50,859 tweets related to Bitcoin. It is partitioned into an 80% training set and a 20% testing set previously. The testing dataset from the previous models, encompassing 10,174 tweets, is stored in a CSV file. This dataset will be employed to assess the accuracy of the Pysentimento model, ensuring uniformity across all three models under evaluation. (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e) and (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e) demonstrate the stages of Pysentimento model and the results of Pysentimento application on a portion of the testing dataset correspondingly. In (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e), the \"label\" column represents the classification of the training dataset whereas the \"sentiment\" is the one for Pysentimento.\u003c/p\u003e \u003cdiv id=\"Sec21\" class=\"Section3\"\u003e \u003ch2\u003e6.5.1 Classification report of (Pysentimento)\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe presented classification report concerns the \"Pysentimento\" model's performance following its application to test data. This report evaluates the model's capabilities in classifying text-based sentiments into three categories: \"positive,\" \"negative,\" and \"neutral.\" It is crucial to analyze each aspect of the report to gain a comprehensive understanding of the model's performance as conveyed in (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e). The provided classification report pertains to the performance of the \"Pysentimento\" model after its application to testing data. This report evaluates the model's ability to classify text-based sentiments into three categories: \"positive,\" \"negative,\" and \"neutral.\" It is crucial to analyze each aspect of the report to gain a comprehensive understanding of the model's performance.\u003c/p\u003e \u003cp\u003eWhen it comes to assessing the \"positive\" emotion category, the model performs admirably. For this category, the precision is 0.89, meaning that 89% of the feelings that the model predicted to be \"positive\" are indeed true. The recall for the \"positive\" category is equally impressive, with a score of 0.88, indicating that the model effectively captures 88% of actual positive sentiments present in the dataset. An F1-score of 0.88 in this situation indicates that the model is remarkably well-balanced, with a strong equilibrium between recall and precision. This equilibrium suggests that the model performs well in predicting \"positive\" feelings and in capturing the majority of genuine positive sentiments seen in the data.\u003c/p\u003e \u003cp\u003eThe model's precision score in the \"negative\" sentiment category is 0.83, which means that 83% of the feelings it identified as \"negative\" are actually \"negative.\" However, the recall for the \"negative\" category is 0.69, demonstrating that the model only accurately classifies 69% of the real negative attitudes in the dataset. For the \"negative\" category, the F1-score, a measure of precision and recall equilibrated, is recorded as 0.76. The model appears to be operating in this category quite harmoniously, based on its F1-score of 0.76. With a precision of 0.92 in the \"neutral\" sentiment category, the \"Pysentimento\" model does exceptionally well, correctly predicting 92% of the \"neutral\" feelings among its predictions.\u003c/p\u003e \u003cp\u003eMoreover, the recall for the \"neutral\" category is an impressive 0.97, highlighting the model's capacity to correctly classify 97% of actual neutral sentiments in the dataset. The F1-score for \u0026ldquo;neutral\u0026rdquo; is 0.94, denoting a balance between precision and recall. This score highlights the model\u0026rsquo;s ability to discern neutral sentiments. The \u0026ldquo;Pysentimento\u0026rdquo; model has an overall accuracy of 90%, demonstrating its competence in sentiment prediction across all categories.\u003c/p\u003e \u003cp\u003eIn conclusion, the \"Pysentimento\" model performs admirably when tested utilizing test data. It maintains a balanced F1-score, demonstrating its efficacy in capturing various sentiment categories, and demonstrates notable strengths in precision, especially in the \"neutral\" category. The model's overall 90% accuracy rating specifies how consistently it can predict \"positive,\" \"negative,\" and \"neutral\" sentiments.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"7 The Second evaluation","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe researcher's choice of utilizing a second evaluation stems from a compelling need for rigorous accuracy validation. This imperativeness arises due to substantial doubts surrounding the observed accuracy levels of the three models. The core objective is to ascertain the authenticity of the attained accuracy, safeguarding against potential pitfalls like overfitting or underfitting, which may compromise the credibility of the results. Given the paramount prominence of this objective, the previously constructed models are saved and subsequently recalled to undergo a test against a dataset comprising 100 tweets of the randomly collected data regarding three cryptocurrencies: (BTC), (ETH), and (BNB), which represents unseen data manually categorized by the researcher into positive, negative, and neutral sentiments. (Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e13\u003c/span\u003e) unveils the results of the second evaluation.\u003c/p\u003e \u003cp\u003eThe performance metrics, including precision, recall, F1 rating, and accuracy, for the four distinct NLP models become conveyed. They are 'CNN-LSTM,' 'SVM (GloVe),' 'SVM (TF-IDF),' and 'Pysentimento.' Such metrics function as fundamental indicators of the models' proficiency in accurately identifying positive, negative, and neutral sentiments within a specified dataset. The precision values reveal observable modifications across every model under examination. These indices serve as indications of the percentage of correctly expected positive cases among all cases expected as positive. 'CNN-LSTM' uncovers a noteworthy precision of 90%, emphasizing its dexterity in productively categorizing positive sentiments with precision. In contrast, 'SVM (GloVe)' and 'SVM (TF-IDF)' demonstrate comparatively reduced precision values of 80% and 78%, respectively, indicating a relatively smaller accuracy in correctly distinguishing positive cases.\u003c/p\u003e \u003cp\u003eContrastingly, 'Pysentimento' separates itself with a remarkable precision of 97%, symptomatic of its heightened ability to outwit incorrect positives and discern positive sentiments with extraordinary accuracy. This discrepancy in precision values underscores the nuanced execution divergences among the models, thereby offering crucial insights into their efficacy in positive sentiment characterization.\u003c/p\u003e \u003cp\u003eThe recall values, which signify the proportion of actual positive instances properly recognized by the models, unveil intricate patterns that shed light on the models' efficacy in capturing positive sentiments. 'CNN-LSTM' achieves a commendable recall of 88%, affirming its ability to capture a considerable part of genuine positive sentiments within the dataset. In contrast, both 'SVM (GloVe)' and 'SVM (TF-IDF)' present dissimilar recall values of 95% and 75%, respectively. This delineates varied abilities in correctly identifying positive instances, with 'SVM (GloVe)' demonstrating a notably higher recall than 'SVM (TF-IDF).' Noteworthy is the exceptional execution of 'Pysentimento,' boasting a perfect recall of 100%, indicating its unparalleled capability to comprehensively identify all positive sentiments present in the dataset. This divergence in recall values among the models accentuates nuanced distinctions in their ability to effectively identify and capture actual positive sentiments, providing valuable insights into their respective strengths in positive sentiment recognition.\u003c/p\u003e \u003cp\u003eThe examination of the F1 rating provides a thorough assessment of the models accomplishment. A harmonic average derived from recall and precision is the F1 rating. 'CNN-LSTM' becomes distinguished by an F1 rating of 89%, indicating a tuneful balance between precision and recall, thereby proposing a well-rounded effectiveness. In contrast, 'SVM (GloVe)' and 'SVM (TF-IDF)' present F1 scores of 87% and 77%, respectively, accentuating nuanced variations in their precision-recall equilibrium. The heightened F1 rating of 98% accomplished by 'Pysentimento' attests to its exceptional proficiency in accomplishing a brilliant amalgamation of precision and recall, underscoring its intensified effectiveness in sentiment analysis tasks.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"8 Comparative analysis","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eBelow is a comprehensive analysis of sentiment classification models, namely \"CNN-LSTM,\" \"SVM (GloVe),\" \"SVM (TF-IDF),\" and \"Pysentimento.\" The focus is on F1 score values, accuracy, precision, and recall in both the first and second assessments. Subtle variations in the models' output that provide insight into how well they perform in sentiment analysis applications will be revealed through careful inspection of these metrics.\u003c/p\u003e \u003cp\u003eThe precision values for sentiment classification models in the first and second evaluations exhibit slight variations. 'CNN-LSTM' obtains a precision of 0.90 in the first evaluation, but in the second evaluation, it slightly drops to 0.89. Conversely, \"SVM (GloVe)\" exhibits a more significant decline from 0.80 to 0.87, indicating an increase in its capacity to accurately detect pleasant emotions. The precision of \"SVM (TF-IDF)\" decreases from 0.78 to 0.77, suggesting a slight deterioration in its precise performance. In contrast, 'Pysentimento' displays consistency in both evaluations, with a precision value of 0.97, highlighting its ability to prevent false positives.\u003c/p\u003e \u003cp\u003eSimilarly, recall values for each of the sentiment categorization models disclose significant alterations between the two assessments. The recall of the \"CNN-LSTM\" model falls slightly from 0.88 to 0.87, suggesting a marginal decline in the model's ability to identify true positive cases. In contrast, \"SVM (GloVe)\" reveals a boost in its recall from 0.95 to 0.98, indicating an elevated degree of accuracy in identifying positive sentiments. \"SVM (TF-IDF)\" has a more substantial reduction in recall from 0.75 to 0.68, indicating a shift in its proficiency in recognizing positive instances. It is noteworthy that 'Pysentimento' consistently scores a perfect recall of 1.00 in both evaluations, demonstrating its unwavering ability to fully detect all positive sentiments in the dataset.\u003c/p\u003e \u003cp\u003eThe F1 score offers a comprehensive view of model performance as a well-balanced combination of recall and precision. With an F1 score of 0.89 in both assessments, \"CNN-LSTM\" performs consistently and represents an equitable trade-off between recall and precision. The F1 score of \"SVM (GloVe)\" slightly increases from 0.87 to 0.88, demonstrating a better equilibrium between recall and precision. However, \"SVM (TF-IDF)\" experiences a more significant decrease in F1 score, from 0.77 to 0.73, suggesting a change in the precision-recall balance. With a consistent F1 score of 0.98 in both tests, \"Pysentimento\" performs exceptionally well, demonstrating its ongoing ability to achieve a pleasing balance between recall and precision.\u003c/p\u003e \u003cp\u003eThe sentiment classification models' accuracy values reflect significant variations between the first and second assessments. In both evaluations, \"CNN-LSTM\" consistently scores an accuracy of 0.86, demonstrating stability of its overall prediction ability. From 0.82 to 0.81, \"SVM (GloVe)\" shows a slight decline, indicating a minor change in accuracy performance. SVM (TF-IDF) experiences a more notable decline from 0.79 to 0.71, signifying a notable change in its overall accuracy. Conversely, 'Pysentimento' has a remarkable consistency, exhibiting a constant accuracy of 0.92, highlighting its continuous ability to anticipate sentiments correctly in all classes.\u003c/p\u003e \u003cp\u003ePotential overfitting or underfitting factors when examining performance fluctuations in sentiment analysis models such as \"CNN-LSTM,\" \"SVM (GloVe),\" \"SVM (TF-IDF),\" and \"Pysentimento\" must be taken into account. When a model becomes overfitted to the training data, it captures irrelevant details and exhibits poorer performance on new, unseen data. This phenomenon is known as overfitting. Conversely, underfitting occurs when a model accomplishes poorly than expected across measurements due to its oversimplification. Overfitting may be indicated by abrupt declines in accuracy, recall, F1 score, or precision, while underfitting, which exposes the model's simplicity in capturing sentiment classifications, may be designated by persistent underperformance across metrics.\u003c/p\u003e \u003cp\u003eIn conclusion, subtle variances in the performance metrics of the sentiment analysis models \"CNN-LSTM,\" \"SVM (GloVe),\" \"SVM (TF-IDF),\" and \"Pysentimento\" are exposed through comparing their analyses between the first and second evaluations. In both assessments, \"CNN-LSTM\" continuously maintains lofted values for accuracy, recall, F1 score, and precision, indicating a strong and reliable performance. Recall and F1 score improvements for \"SVM (GloVe)\" demonstrate the system's capacity to recognize positive sentiments. A number of measures indicate a possible shift in the precision-recall balance of 'SVM (TF-IDF)'. It is noteworthy that 'Pysentimento' consistently exhibits exceptional precision, recall, and F1 score, confirming its effectiveness regarding sentiment analysis accurate prediction and it will be employed on the data concerning the three cryptocurrencies.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"9 Pysentimento and Google Trends","content":"\u003cp\u003eThis implementation of Pysentimento is preceded by a pre-processing step, which involves the removal of irrelevant tweets from the dataset pertaining to the three currencies. A pattern of words is designated and it serves the purpose of recognizing tweets that pertain to the context of cryptocurrencies, encompassing terms such as Bitcoin, Ethereum, and Binance, along with a multitude of associated hash tags and keywords. Tweets that deviate from this pattern, commonly referred to as \"mismatches,\" are systematically identified and segregated, with a dedicated output file housing these mismatched tweets. As for Google Trends, (Fig.\u0026nbsp;\u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e14\u003c/span\u003e) illustrates total searches for (BTC), (ETH), (BNB) respectively in March, June and December, 2022. The total searches on Google Trends for the three cryptocurrencies, Bitcoin, Ethereum, and Binance Coin in 2022 demonstrate varying levels of attention cross the given months. In March, Bitcoin reached 2075 searches, Ethereum garnered 2155 searches, and Binance Coin observed 2621 searches. In June, Bitcoin declined to 1506 searches, Ethereum reduced to 1371 searches, and Binance Coin held a substantial interest with 2150 searches. December notably saw a significant increase in searches for Bitcoin and Ethereum, reaching 2558 searches for each, indicating heightened fascination in these cryptocurrencies toward the end of the year. In contrast, Binance Coin observed a decrease in searches to 1899.\u003c/p\u003e \u003cp\u003eThese search volume fluctuations imply dynamic shifts in public sentiment and interest regarding these cryptocurrencies over the course of the year 2022. The collective data reveal complex patterns in the online attention for these digital assets, potentially swayed by external elements such as market trends, regulatory developments, and overall market sentiment.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eAfter the application of Pysentimento to the data, the sentiments sum total is collected for the three previously designated cryptocurrencies and months in 2022. The opening and closing of last days for each month in addition to the trading volume of the three cryptocurrencies are provided using yahoo finance.\u003c/p\u003e \u003cp\u003eThe trading volume is the total quantity of a cryptocurrency that is traded across all exchanges over a specific timeframe. It exhibits the total quantity of cryptocurrency units that have been acquired and sold on the market during that period. Trading volume is a crucial metric in the cryptocurrency market since it offers insights into the market activity and liquidity. Higher trading volumes commonly indicate increased market interest, liquidity, and the potential for more accurate price discovery. On the contrary, lower trading volumes could suggest reduced interest, potentially leading to higher price volatility and less reliable price information (Bitsgap, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). (Table. 1) displays how the collected data are organized.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003ctable id=\"Tab1\" border=\"1\" style=\"margin-right: calc(0%); width: 100%;\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eCollected data organization\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" style=\"width: 5.4985%;\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" style=\"width: 4.2766%;\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colspan=\"3\" style=\"width: 20.3647%;\"\u003e\n \u003cp\u003eSentiment\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" style=\"width: 6.1094%;\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" style=\"width: 8.9605%;\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colspan=\"5\" style=\"width: 31.0562%;\"\u003e\n \u003cp\u003ePrices\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eMonth\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eCoin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003eTotal Tweets\u003c/p\u003e\n \u003cp\u003e%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003ePositive %\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003eNegative %\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003eNeutral\u003c/p\u003e\n \u003cp\u003e%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003eOverall Sentiment\u003c/p\u003e\n \u003cp\u003e%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003eOpening (\u003cspan\u003e$\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003eClosing (\u003cspan\u003e$\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003eVolume (billions)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003eGoogle index\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eMAR-22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eBNB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003e51.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003e43.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003e5.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003e51.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003e38.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e395.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e428.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003e54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003e2621\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eJUN-22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eBNB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003e57.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003e33.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003e8.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003e57.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003e42.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e320.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e219.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003e39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003e2150\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eDEC-22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eBNB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003e54.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003e32.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003e12.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003e54.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003e45.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e300.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e246.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003e21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003e1899\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eMAR-22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eBTC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003e57.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003e33.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003e9.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003e57.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003e42.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e43,194.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e45,538.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003e831\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003e2075\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eJUN-22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eBTC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003e60.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003e23.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003e15.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003e60.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003e39.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e31,792.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e19,784.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003e924\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003e1506\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eDEC-22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eBTC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003e57.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003e29.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003e12.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003e57.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003e42.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e17,168.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e16,547.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003e541\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003e2558\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eMAR-22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eETH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003e59.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003e36.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003e3.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003e59.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003e40.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e2,919.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e3,281.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003e437\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003e2155\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eJUN-22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eETH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003e71.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003e23.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003e5.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003e71.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003e28.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e1,942.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e1,067.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003e562\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003e1371\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" style=\"width: 5.4985%;\"\u003e\n \u003cp\u003eDEC-22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 4.2766%;\"\u003e\n \u003cp\u003eETH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.4149%;\"\u003e\n \u003cp\u003e60.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.6185%;\"\u003e\n \u003cp\u003e37.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 7.3313%;\"\u003e\n \u003cp\u003e2.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 6.1094%;\"\u003e\n \u003cp\u003e60.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.9605%;\"\u003e\n \u003cp\u003e39.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e1,295.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 8.0441%;\"\u003e\n \u003cp\u003e1,196.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" style=\"width: 10.8951%;\"\u003e\n \u003cp\u003e161\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\" style=\"width: 8.1459%;\"\u003e\n \u003cp\u003e2558\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe table provides a detailed overview of cryptocurrency data for March, June, and December 2022. The \"Month\" column categorizes data based on the recording month, allowing for a closer look at cryptocurrency performance during specific periods. The \"Year\" column adds chronological context, clarifying data for the year 2022. Each entry in the \"Coin\" column acts as a unique identifier, indicating the cryptocurrency (e.g., BNB-USD, BTC-USD, ETH-USD) and the recorded month. This simplifies the analysis process. Sentiment-related columns like \"Positive %,\" \"Negative %,\" \"Neutral %,\" and \"Overall Sentiment %\" break down sentiment percentages for each cryptocurrency. The \"Total Tweets%\" column provides insights into social media activity.\u003c/p\u003e \u003cp\u003eFor sentiment analysis, the \"Positive %,\" \"Negative %,\" \"Neutral %,\" and \"Overall Sentiment %\" columns are crucial, revealing the percentage of positive, negative, neutral sentiments, and an overall sentiment indicator. \"Opening (\u003cspan\u003e$\u003c/span\u003e)\" shows the opening price on the first day of the month, while \"Closing (\u003cspan\u003e$\u003c/span\u003e)\" indicates the closing price on the last day. \"Volume (billions)\" records the total trading volume, and the \"Google index\" reflects Google searches. Normalization is applied to several columns, including \"Positive %,\" \"Negative %,\" \"Neutral %,\" \"Overall Sentiment %,\" \"Volume (billions),\" and \"Google index.\" This ensures comparability and meaningful analysis across different data scales. The dataset is a valuable resource for analysts, investors, and researchers, offering insights into cryptocurrency behavior, performance, and sentiment analysis. It enhances understanding in correlation and time series forecasting, providing valuable indicators of market sentiment.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"10 Correlation analysis","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe Pearson correlation coefficient serves as an instrument for calculating the linear association between two variables. It assumes a pivotal role in the assessment and quantification of relationships existing amidst diverse predictor variables and target variables found within the dataset. Precisely, this statistical measure aids in determining the extent of correlation between two variables, signifying whether their association is robust or feeble. The output of the Pearson correlation coefficient is a numerical value that falls within the range of (-1 to 1), with distinct interpretations:\u003c/p\u003e \u003cp\u003eFirst of all, a coefficient of 1 indicates a perfect positive linear relationship, implying that as one variable exhibits an increase, the other demonstrates a synchronous increase at a constant rate. Secondly, a coefficient of -1 signifies a perfect negative linear relationship, suggesting that as one variable experiences growth; the other exhibits a parallel and consistent decrease. Thirdly, when the coefficient assumes a value of 0, it conveys the absence of a linear relationship, denoting that the variables under consideration lack correlation.\u003c/p\u003e \u003cp\u003eThe calculation of Pearson correlation coefficients transpires between a multitude of predictor variables (such as \"Normalized Closing of Last Day,\" \"Normalized Volume,\" \"Normalized SumTotal Google Trends\") and target variables (including \"Normalized Overall Sentiment,\" \"Normalized Positive Sum,\" \"Normalized Negative Sum,\" and \"Normalized Neutral Sum\"). The importance of computing Pearson connection coefficients in this specific setting stems from the need to comprehend how the anticipating variables interrelate with the target variables. Such an investigation demonstrates valuable in numerous regards:\u003c/p\u003e \u003cp\u003eFundamentally, it assists in recognizing anticipating - target relationships, offering clearness on which anticipating variables exert a huge impact on the target variables. After that, inside the area of predictive displaying and machine learning, understanding the connections between elements including anticipating variables and the target variable is of most extreme significance. Elements showing strong connections regularly emerge as likely parts for prescient models. Next, it adds to the creation of bits of knowledge identifying with the potential outcomes of changes in particular anticipating variables on the target variables. These bits of learning hold worth for educated dynamic producing and the elucidation of information examples.\u003c/p\u003e \u003cp\u003eTo realize such a purpose, a heatmap illustrating the correlation coefficients among various sentiment measures for three specific cryptocurrencies: BNB, BTC, and ETH. Within each cryptocurrency, sentiment measures encompass sub-measures related to predictor variables, while columns represent target variables in (Fig.\u0026nbsp;\u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e15\u003c/span\u003e). Such coefficients result in priceless insights into the interplay between sentiment and the three cryptocurrencies market indicators can be comprehended. As for (ETH), Normalized Overall Sentiment exhibits a correlation coefficient of (0.3773) with \"Normalized Closing of Last Day.\" This indicates a correlation where Ethereum's overall sentiment tends to rise modestly with an increase in its closing price at the end of each month. Conversely, a substantial negative correlation (-0.8531) between \"Normalized Volume\" and Normalized Overall Sentiment implies that higher Ethereum trading volumes coincide with decreased overall sentiment. A robust positive correlation (0.9885) between \"Normalized SumTotal Google Trends\" and Normalized Overall Sentiment reveals that heightened Google search activity corresponds to increased overall sentiment for Ethereum.\u003c/p\u003e \u003cp\u003eNormalized Positive Sum proves a moderate positive correlation (0.3811) with \"Normalized Closing of Last Day,\" indicating that as Ethereum's closing price rises at month-end, positive sentiment similarly tends to increase, though modestly. \"Normalized Volume\" exhibits a strong negative correlation (-0.851) with Normalized Positive Sum, showing that increased Ethereum trading volumes coincide with a significant decrease in positive sentiment. There's a solid positive correlation (0.9879) between \"Normalized SumTotal Google Trends\" and Normalized Positive Sum, illustrating that increased Google search activity for Ethereum aligns with elevated positive sentiment. Weak negative correlation (-0.3507) observed between Normalized Negative Sum and \"Normalized Closing of Last Day.\" There's a solid positive correlation (0.9879) between \"Normalized SumTotal Google Trends\" and Normalized Positive Sum, illustrating that increased Google search activity for Ethereum aligns with elevated positive sentiment. Weak negative correlation (-0.3507) observed between Normalized Negative Sum and \"Normalized Closing of Last Day.\"\u003c/p\u003e \u003cp\u003eIn essence, as Ethereum's month-end closing price increases, negative sentiment slightly decreases, indicating minor shifts in sentiment with closing price. \"Normalized Volume\" and Normalized Negative Sum illustrate a notably strong correlation (0.8677), demonstrating that increased Ethereum trading volumes align with a significant rise in negative sentiment. \"Normalized SumTotal Google Trends\" and Normalized Negative Sum have a strong negative correlation (-0.9925), indicating that increased Google search activity for Ethereum aligns with a significant decrease in negative sentiment. \"Normalized Closing of Last Day\" and Normalized Neutral Sum exhibit a strong negative correlation (-0.8295). \"Normalized Volume\" and Normalized Neutral Sum show a less pronounced correlation (0.4174), implying a relatively weak positive relationship.\u003c/p\u003e \u003cp\u003eIn conclusion, the substantial influence of public interest on Ethereum's market dynamics is highlighted by the robust correlation between heightened Google search activity and diminished neutral sentiment. The nuanced nature of Ethereum's sentiment dynamics is emphasized by the varied impacts of predictor variables on different sentiment aspects. The significance of factors such as closing prices and trading volume in shaping market sentiment is suggested by positive and negative correlations, respectively. The strong association between Google search trends and sentiment implies the importance of information-seeking behavior. These collective findings indicate that monitoring Google Trendsdata, closing prices, and trading volume is imperative for insightful market analysis, enabling informed investment decisions and forecasting.\u003c/p\u003e \u003cp\u003eRegarding (BTC), \"Normalized Closing of Last Day\" strongly correlates (0.867) with overall sentiment, signifying a noteworthy association with Bitcoin's closing price. \"Normalized Volume\" has a negative correlation (-0.1305) with \"Normalized Overall Sentiment,\" suggesting that fluctuations in trading volume have a limited negative impact on Bitcoin's overall sentiment. Moreover, \"Normalized SumTotal Google Trends\" reveals a moderate positive correlation of (0.45) with \"Normalized Overall Sentiment.\"\u003c/p\u003e \u003cp\u003ePositive sentiment, particularly with \"Normalized Closing of Last Day,\" shows a notably strong correlation of (0.8883). This robust connection implies that positive sentiment in the cryptocurrency market is highly responsive to Bitcoin's closing price, with a higher closing price corresponding to overwhelmingly positive sentiment. In opposition, \"Normalized Volume\" exhibits a weak negative correlation of (-0.0864) with \"Normalized Positive Sum.\" Furthermore, \"Normalized SumTotal Google Trends\" proves a moderate positive correlation of (0.4099) with \"Normalized Positive Sum.\"\u003c/p\u003e \u003cp\u003eNegative sentiment, especially with \"Normalized Closing of Last Day,\" exhibits a significant negative correlation of (-0.4232). A rise in closing price tends to reduce negative sentiment, emphasizing the pivotal influence of price performance (increases) on negative sentiment in the cryptocurrency market.\"Normalized Volume\" shows a significant positive correlation of (0.6766) with \"Normalized Negative Sum.\" \"Normalized SumTotal Google Trends\" shows a significant negative correlation of (-0.8814) with \"Normalized Negative Sum.\" High Google search interest in Bitcoin corresponds to a notable decrease in negative sentiment. Neutral sentiment, particularly starting with \"Normalized Closing of Last Day,\" displays a remarkably strong positive correlation of (0.9751). Additionally, \"Normalized Volume\" demonstrates a modest positive correlation of (0.1663) with \"Normalized Neutral Sum.\" \"Normalized SumTotal Google Trends\" shows a similar modest positive correlation of (0.1679) with \"Normalized Neutral Sum.\"\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003eTo conclude, the correlation analysis stresses a consistent and substantial positive relationship between Bitcoin's closing price and all sentiment aspects, with correlation coefficients ranging from 0.867 to 0.9751. This implies that as Bitcoin's price increases, sentiment across positive, negative, and neutral aspects becomes notably more positive. In contrast, the impact of \"Normalized Volume\" on sentiment is relatively weak, suggesting a minor and inconsistent effect on overall sentiment and its components.\u003c/p\u003e \u003cp\u003eConcerning (BNB), the coefficient between \"Normalized Closing of Last Day\" and \"Normalized Overall Sentiment\" is (0.9914), indicating a remarkably strong positive linear relationship. Likewise, the correlation coefficient between \"Normalized Volume\" and \"Normalized Overall Sentiment\" is (0.6927), demonstrating a positive correlation, though not as powerful as the closing price. The coefficient (0.8254) between \"Normalized SumTotal Google Trends\" and \"Normalized Overall Sentiment\" indicates a strong positive relationship. Increased Google search volume for Binance corresponds to a considerable rise in overall sentiment. The correlation coefficient (0.7434) between \"Normalized Closing of Last Day\" and \"Normalized Positive Sum\" suggests a moderately strong positive linear relationship. The correlation coefficient between \"Normalized Volume\" and \"Normalized Positive Sum\" is (0.1635), denoting a positive correlation, though relatively weak.\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eThe coefficient (0.3614) between \"Normalized SumTotal Google Trends\" and \"Normalized Positive Sum\" indicates a positive relationship. Higher Google search volume for Binance corresponds to an increase in positive sentiment. The correlation coefficient (-0.4831) between \"Normalized Closing of Last Day\" and \"Normalized Negative Sum\" suggests a moderate negative relation. An increase in Binance's closing price on the last day of the month corresponds to a noticeable decrease in negative sentiment.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSimilarly, a burly negative correlation is indicated by the coefficient (-0.9239) between \"Normalized Volume\" and \"Normalized Negative Sum.\" A substantial decrease in negative sentiment towards Binance is robustly linked to an increase in the trading volume. The coefficient (-0.8263) between \"Normalized SumTotal Google Trends\" and \"Normalized Negative Sum\" specifies a firm negative relationship. Higher Google search volume for Binance correlates strongly with decreased negative sentiment. The correlation coefficient (-0.4831) between \"Normalized Closing of Last Day\" and \"Normalized Negative Sum\" suggests a moderate negative relationship.\u003c/p\u003e \u003cp\u003eThe coefficient (-0.9239) between \"Normalized Volume\" and \"Normalized Negative Sum\" conveys a solid negative correlation. Higher Binance trading volume is stoutly linked to a notable decrease in negative sentiment. The correlation coefficient (-0.8263) between \"Normalized SumTotal Google Trends\" and \"Normalized Negative Sum\" denotes a glaring negative relationship.\u003c/p\u003e \u003cp\u003eIn conclusion, the noticeable patterns across these correlations indicate that the closing price, trading volume, and Google search trends are influential factors in shaping sentiment towards Binance. While overall sentiment and positive sentiment display positive relationships with these predictors, negative and neutral sentiments exhibit negative or weaker associations. This comprehensive understanding of correlations provides valuable insights for stakeholders in Binance, aiding in strategic decision-making, market analysis, and sentiment trend predictions. In considering investment decisions amid the Russian-Ukrainian war, Bitcoin emerges as an appealing choice due to its positive correlations with market variables, providing stability sought by investors during geopolitical uncertainties. Binance Coin, with its mixed correlations, demands a nuanced approach, considering potential impacts on its distinctive dynamics. Ethereum, showing positive associations akin to Bitcoin, may attract those seeking diverse opportunities. The choice among these cryptocurrencies, amidst the complexities of the war, should align with investors' risk tolerance and goals, necessitating thorough research, consideration of additional factors, and staying informed within the dynamic cryptocurrency market. Price Forcasting analysis integration remains pertinent in navigating the evolving geopolitical landscape and making well-informed investment decisions.\u003c/p\u003e"},{"header":"11 Price forecasting (SARIMA)","content":"\u003cp\u003eCryptocurrency price forecasting has drawn escalating attention, as researchers and practitioners deploy advanced techniques to boost the meticulousness of predictions. The SARIMA (Seasonal Autoregressive Integrated Moving Average) library, a Python package that actualizes the SARIMA model for time series forecasting, is utilized for this purpose. Founded on the statsmodels library, which furnishes an extensive array of tools for statistical analysis in Python, the SARIMA library is instrumental in modeling and forecasting univariate time series data that exhibit non-stationarity (Brownlee, 2019; Aggarwal, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The SARIMA model extends the ARIMA model, standing for Autoregressive Integrated Moving Average, and introduces a seasonal component to capture periodic fluctuations occurring at fixed intervals, such as daily, weekly, monthly, or yearly.\u003c/p\u003e \u003cp\u003eSARIMA excels in capturing seasonal patterns, enhancing forecast precision. Addressing non-stationary data, it adeptly handles cyclical fluctuations and variations. SARIMA provides confidence intervals and diagnostic plots for forecasts, aiding in assessing uncertainty and reliability. Despite its benefits, SARIMA requires substantial historical data for accurate models and faces challenges in parameter determination and computational intensity (Brownlee, 2019; Hyndman \u0026amp; Athanasopoulos, 2021). (Fig.\u0026nbsp;\u003cspan refid=\"Fig16\" class=\"InternalRef\"\u003e16\u003c/span\u003e) clarifies the predictions based on the prices spinning March, June, and December 2022. The prices in (Table. 1) are utilized in the training\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe figure illustrates price predictions for three cryptocurrencies\u0026mdash;Ethereum (ETH), Bitcoin (BTC), and Binance Coin (BNB)\u0026mdash;spanning from January to March 2023. Two lines are depicted: the blue line denotes actual prices, and the orange dotted line reflects predicted prices. Predictive outcomes are delineated for each cryptocurrency, encompassing Ethereum (ETH), Binance Coin (BNB), and Bitcoin (BTC) as shown above. These predictions offer valuable insights into anticipated price trends.\u003c/p\u003e \u003cp\u003eFor Ethereum, the model forecasts \u003cspan\u003e$\u003c/span\u003e1727.21 on January 31, 2023, \u003cspan\u003e$\u003c/span\u003e1603.38 on February 28, 2023, and \u003cspan\u003e$\u003c/span\u003e1632.29 on March 31, 2023. Binance Coin is expected to reach \u003cspan\u003e$\u003c/span\u003e286.71 on January 31, 2023, \u003cspan\u003e$\u003c/span\u003e299.61 on February 28, 2023, and \u003cspan\u003e$\u003c/span\u003e303.72 on March 31, 2023. Bitcoin's predicted prices are \u003cspan\u003e$\u003c/span\u003e20,739.52 on January 31, 2023, \u003cspan\u003e$\u003c/span\u003e20,630.29 on February 28, 2023, and \u003cspan\u003e$\u003c/span\u003e20,633.13 on March 31, 2023. These projections offer a detailed insight into expected price movements, derived from the SARIMA model's analysis of historical data and temporal patterns.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo gauge prediction accuracy, the Root Mean Squared Error (RMSE) metric is utilized. It is applied to the model's predictions to compare them with the actual prices for the three cryptocurrencies in 2023 in (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e17\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe aforementioned image comprises three line graphs, each delineating the price trends of distinct cryptocurrencies over a specified duration. The blue and orange lines in each graph represent the actual and predicted prices respectively. The first graph illustrates the Root Mean Square Error (RMSE) for Binance Coin (BNB) over January, February, and March 2023. The second graph, following a similar pattern, represents the RMSE for Ethereum (ETH) during the same period. The third graph, meanwhile, depicts the RMSE for Bitcoin (BTC). The divergence between the actual and predicted prices in each graph signifies the prediction error. A greater divergence implies a higher RMSE, indicating a less accurate prediction, while a smaller divergence suggests a lower RMSE, indicative of a more accurate prediction.\u003c/p\u003e \u003cp\u003eThe model predicts BTC closing prices at \u003cspan\u003e$\u003c/span\u003e20,739.52, \u003cspan\u003e$\u003c/span\u003e20,630.29, and \u003cspan\u003e$\u003c/span\u003e20,633.13 for January 31, 2023, February 28, 2023, and March 31, 2023, respectively. BNB forecasts are \u003cspan\u003e$\u003c/span\u003e286.71, \u003cspan\u003e$\u003c/span\u003e299.61, and \u003cspan\u003e$\u003c/span\u003e303.72, while ETH projections are \u003cspan\u003e$\u003c/span\u003e1,727.21, \u003cspan\u003e$\u003c/span\u003e1,603.38, and \u003cspan\u003e$\u003c/span\u003e1,632.29 during the same period. Actual closing prices for BTC, BNB, and ETH on these dates are \u003cspan\u003e$\u003c/span\u003e22,840.14, \u003cspan\u003e$\u003c/span\u003e23,522.87, \u003cspan\u003e$\u003c/span\u003e28,033.56, \u003cspan\u003e$\u003c/span\u003e307.07, \u003cspan\u003e$\u003c/span\u003e304.86, \u003cspan\u003e$\u003c/span\u003e316.57, \u003cspan\u003e$\u003c/span\u003e1,567.33, \u003cspan\u003e$\u003c/span\u003e1,634.33, and \u003cspan\u003e$\u003c/span\u003e1,792.74, respectively. RMSE values, assessing predictive accuracy, are 4745.03 for BTC, 131.99 for ETH, and 14.22 for BNB. Elevated BTC RMSE indicates challenges capturing its complex price patterns. ETH's lower RMSE suggests better predictive performance, while BNB's notably low RMSE indicates accurate predictions, hinting at more foreseeable price movements.\u003c/p\u003e \u003cp\u003eIn summary, the distinct RMSE values underscore the variability in predicting different cryptocurrencies with the SARIMA model. The intricate field of predicting cryptocurrency prices requires continuous examination and adjustments to align with the ever-changing dynamics of the market. This suggests that the predictability of each cryptocurrency differs, emphasizing the ongoing need to fine-tune model parameters for reliable predictions. Given the geopolitical uncertainties introduced by the Russian-Ukrainian war, accurate cryptocurrency price forecasting becomes increasingly crucial. While the SARIMA model holds promise, its effectiveness hinges on continuous adaptation to the unique characteristics of each cryptocurrency. Persistent efforts to improve forecasting techniques are essential, particularly in response to geopolitical shifts and market fluctuations linked to the ongoing conflict.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e*Funding: This research is not supported by any institution or foundation\u003c/p\u003e \u003cp\u003e*Conflicts of interest/Competing interests - (N/A)\u003c/p\u003e \u003cp\u003e*Ethics approval - (N/A)\u003c/p\u003e \u003cp\u003e*Consent to participate - (N/A)\u003c/p\u003e \u003cp\u003e*Consent for publication - (N/A)\u003c/p\u003e \u003cp\u003e*Availability of data and material: (The following link provides access to the a Kaggle website where first model employed in my research along with the first dataset, which consists of 50,859 tweets that have been categorized as ['positive'], ['negative'], and ['neutral']. This dataset is obtained from the Kaggle website, which is a renowned platform and online community for data scientists and machine learning practitioners. The link is mentioned in the references, and it has the following in-text citation in \u003cspan refid=\"Sec6\" class=\"InternalRef\"\u003edata collection\u003c/span\u003e section in my research (Dundee, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2002\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBitcoin Tweets Sentiment Analysis: cnn-lstm | Kaggle\u003c/p\u003e \u003cp\u003e*Code availability: codes cannot be shared openly, to protect the author\u0026rsquo;s privacy due to the fact that some of such codes are directly involved in my Ph.D. thesis, which has not yet been defended.\u003c/p\u003e \u003cp\u003e*Authors' contributions:\u003c/p\u003e \u003cp\u003eAs for the roles of the authors, they are provided as follows:\u003c/p\u003e\n\u003cp\u003e1-M.R. authored the manuscript and conducted the research experiments and analysis.\u003c/p\u003e\n\u003cp\u003e2-M.G. provided consultation on machine learning and deep learning approaches, guided the experimental methods and analysis, and reviewed the coding.\u003c/p\u003e\n\u003cp\u003e3-M.K. and S.A. served as research supervisors.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbraham, J., Higdon, D., Nelson, J., \u0026amp; Ibarra, J. (2018). Cryptocurrency price prediction using tweet volumes and sentiment analysis. SMU Data Science Review, 1(3), 1.\u003c/li\u003e\n\u003cli\u003eAggarwal, P. K. (2020). Powerful learning is all about retrieval. ASCD Education Update, 62(1), 1-5. https://www.ascd.org/el/articles/powerful-learning-is-all-about-retrieval\u003c/li\u003e\n\u003cli\u003eAkhtar, W., Kumaraguru, P., and Joshi, A. (2019). Sentiment analysis for cryptocurrencies using roBERTa transformer model with self-attention mechanism. In Proceedings of the Third Workshop on Blockchain Technologies and Applications (pp. 1-10).\u003c/li\u003e\n\u003cli\u003eAL-MANSOUR, B. Y. (2020). Cryptocurrency market: Behavioral finance perspective. \u003cem\u003eThe Journal of Asian Finance, Economics and Business\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(12), 159\u0026ndash;168. https://doi.org/10.13106/jafeb.2020.vol7.no12.159\u003c/li\u003e\n\u003cli\u003eBalasudarsun, N. L., Ghosh, B., \u0026amp; Mahendran, S. (2022). Impact of negative tweets on diverse assets during stressful events: An investigation through time-varying connectedness. \u003cem\u003eJournal of Risk and Financial Management\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(6), 260. https://doi.org/10.3390/jrfm15060260\u003c/li\u003e\n\u003cli\u003eBird, S., Klein, E., \u0026amp; Loper, E. (2009). Tokenization for Natural Language Processing. In Natural Language Processing with Python (pp. 39-57). O\u0026apos;Reilly Media, Inc.\u003c/li\u003e\n\u003cli\u003eBitsgap. (2021, February 9). What is a trading volume in cryptocurrency and why is it important? https://bitsgap.com/blog/what-is-a-trading-volume-in-cryptocurrency-and-why-is-it-important\u003c/li\u003e\n\u003cli\u003eBollen, J., Mao, H., \u0026amp; Zeng, X. (2011). Twitter mood predicts the stock market. \u003cem\u003eJournal of Computational Science\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(1), 1\u0026ndash;8. https://doi.org/10.1016/j.jocs.2010.12.007 \u003c/li\u003e\n\u003cli\u003eCarpedm20. (2015). \u003cem\u003eEmoji: Emoji for Python\u003c/em\u003e (Version 1.7.0) [Software]. Retrieved from https://github.com/carpedm20/emoji\u003c/li\u003e\n\u003cli\u003eCarrillo, J., Carrillo, A., \u0026amp; Carrillo, A. (2020). Pysentimiento: A Python Toolkit for Sentiment Analysis and Social NLP Tasks. In Proceedings of the 28th International Conference on Computational Linguistics: System Demonstrations.\u003c/li\u003e\n\u003cli\u003eChen, T., Li, Z., Zhang, Y., and Li, T. (2018). Neural sentiment classification with user and product attention. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (pp. 1650-1659).\u003c/li\u003e\n\u003cli\u003eCortes, C., \u0026amp; Vapnik, V. N. (1995). Support-vector networks. Machine learning, 20(3), 273-297.\u003c/li\u003e\n\u003cli\u003eDevlin, J., Chang, M. W., Lee, K., \u0026amp; Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of NAACL-HLT 2019 (pp. 4171-4186).\u003c/li\u003e\n\u003cli\u003eDonahue, J., Anne Hendricks, L., Guadarrama, S., Venugopalan, S., Saenko, K., \u0026amp; Darrell, T. (2015). Long-term recurrent convolutional networks for visual recognition and description. IEEE transactions on pattern analysis and machine intelligence, 39(4), 677-691. https://doi.org/10.1109/TPAMI.2015.2500292.\u003c/li\u003e\n\u003cli\u003eDrus, Z., \u0026amp; Khalid, H. (2019). Sentiment analysis in social media and its application: Systematic literature review. \u003cem\u003eProcedia Computer Science\u003c/em\u003e, \u003cem\u003e161\u003c/em\u003e, 707-714.\u003c/li\u003e\n\u003cli\u003eDundee. (2002). Bitcoin Tweets Sentiment Analysis: cnn-lstm. \u003cem\u003eKaggle\u003c/em\u003e. https://www.kaggle.com/code/dundee2002/bitcoin-tweets-sentiment-analysis-GloVe-cnn-lstm/log\u003c/li\u003e\n\u003cli\u003eFaret, J., \u0026amp; Reitan, J. (2015). \u003cem\u003eTwitter Sentiment Analysis-Exploring the Effects of Linguistic Negation\u003c/em\u003e (Master\u0026apos;s thesis, NTNU).\u003c/li\u003e\n\u003cli\u003eFonseca, J. (2020). PyTrends (Version 4.9.1) [Computer software]. GitHub. https://github.com/GeneralMills/PyTrends\u003c/li\u003e\n\u003cli\u003eGaber, M., Ezzat, M., \u0026amp; Mokhtar, M. (2021). Hyperparameter Optimization for Deep Learning-Based Sentiment Analysis. IEEE Access, 9, 78030-78047.\u003c/li\u003e\n\u003cli\u003eGarc\u0026iacute;a-Pablos, A., Cuadros, M., Rigau, G., and Agirre, E. (2020). Pysentimiento: A Python Toolkit for Sentiment Analysis and Social NLP Tasks. In Proceedings of The 12th Language Resources and Evaluation Conference (pp. 7215-7223).\u003c/li\u003e\n\u003cli\u003eHochreiter, S., \u0026amp; Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780\u003c/li\u003e\n\u003cli\u003eHsu, C.-W., Chang, C.-C., \u0026amp; Lin, C.-J. (2003). A practical guide to support vector classification. National Taiwan University.\u003c/li\u003e\n\u003cli\u003eHugging Face. (2021). Transformers: State-of-the-art Natural Language Processing for Pytorch and TensorFlow 2.0. Retrieved May 28, 2021, from https://huggingface.co/transformers/\u003c/li\u003e\n\u003cli\u003eHyunyoung, C., \u0026amp; Varian, H. (2018). Replicating \u0026ldquo;predicting the present with google trends\u0026rdquo; by Hyunyoung Choi and Hal Varian (the economic record, 2012). Economics, 12(1). https://doi.org/10.5018/economics-ejournal.ja.2018-34\u003c/li\u003e\n\u003cli\u003eJohnson, R., Smith, T., Williams, K., \u0026amp; Davis, M. (2020). Using Google Trendsdata to explore public interest in breast cancer screening. BMC Public Health, 20(1), 1-6.\u003c/li\u003e\n\u003cli\u003eKim, Y. (2014). Convolutional neural networks for sentence classification. arXiv preprint arXiv:1408.5882.\u003c/li\u003e\n\u003cli\u003eLeCun, Y., Bottou, L., Bengio, Y., \u0026amp; Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324.\u003c/li\u003e\n\u003cli\u003eLiu, X., Zhang, P., Liu, L., \u0026amp; Zhou, G. (2018). Empower sequence labeling with task-aware neural language model. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 1234-1243).\u003c/li\u003e\n\u003cli\u003eLiu, Z., Lin, W., Shi, Y., \u0026amp; Zhao, J. (2021). A robustly optimized Bert pre-training approach with post-training. \u003cem\u003eLecture Notes in Computer Science\u003c/em\u003e, 471\u0026ndash;484. https://doi.org/10.1007/978-3-030-84186-7_31\u003c/li\u003e\n\u003cli\u003eMedhat, W., Hassan, A., \u0026amp; Korashy, H. (2014). Sentiment analysis algorithms and applications: A survey. Ain Shams Engineering Journal, 5(4), 1093-1113.\u003c/li\u003e\n\u003cli\u003eMikolov T., Sutskever I., Chen K., Corrado G.S., Dean J. (2013) Distributed Representations of Words and Phrases and their Compositionality. In: Advances in Neural Information Processing Systems.\u003c/li\u003e\n\u003cli\u003eownlee, J. (2020). How to develop LSTM models for time series forecasting. Machine Learning Mastery. https://machinelearningmastery.com/how-to-develop-lstm-models-for-time-series-forecasting/\u003c/li\u003e\n\u003cli\u003ePano, T., \u0026amp; Kashef, R. (2020). A complete vader-based sentiment analysis of Bitcoin (BTC) tweets during the era of covid-19. \u003cem\u003eBig Data and Cognitive Computing\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e(4), 33. https://doi.org/10.3390/bdcc4040033\u003c/li\u003e\n\u003cli\u003ePennington, J., Socher, R., \u0026amp; Manning, C. D. (2014). GloVe: Global vectors for word representation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP) (pp. 1532-1543).\u003c/li\u003e\n\u003cli\u003eP\u0026eacute;rez, J. M., Furman, D. A., Alonso Alemany, L., \u0026amp; Luque, F. M. (2022). RoBERTuito: a pre-trained language model for social media text in Spanish. Proceedings of the Thirteenth Language Resources and Evaluation Conference.\u003c/li\u003e\n\u003cli\u003eP\u0026eacute;rez, J. M., Giudici, J. C., \u0026amp; Luque, F. (2021). pysentimiento: A Python toolkit for sentiment analysis and social NLP tasks. arXiv preprint arXiv:2106.09462 \u003c/li\u003e\n\u003cli\u003eRamos, J. (2003). Using tf-idf to determine word relevance in document queries. In Proceedings of the first instructional conference on machine learning (Vol. 242, pp. 133-142).\u003c/li\u003e\n\u003cli\u003eRussell, M. A. (2018). Mining the Social Web: Data Mining Facebook, Twitter, LinkedIn, Instagram, GitHub, and More (3rd ed.). O\u0026apos;Reilly Media, Inc.\u003c/li\u003e\n\u003cli\u003eSattarov, O., Jeon, H. S., Oh, R., \u0026amp; Lee, J. D. (2020). Forecasting bitcoin price fluctuation by Twitter sentiment analysis. \u003cem\u003e2020 International Conference on Information Science and Communications Technologies (ICISCT)\u003c/em\u003e. https://doi.org/10.1109/icisct50599.2020.9351527\u003c/li\u003e\n\u003cli\u003eSch\u0026ouml;lkopf, B., Platt, J. C., Shawe-Taylor, J., Smola, A. J., \u0026amp; Williamson, R. C. (2001). Estimating the dimension of a kernel subspace. In Proceedings of the Fourteenth Annual Conference on Computational Learning Theory (COLT\u0026apos;01), ACM.\u003c/li\u003e\n\u003cli\u003eSharma, N., Khosla, A., Kim, T., Gade, A., \u0026amp; Pagh, R. (2023). TF-IDF: A fundamental technique in natural language processing. GitHub Repository. https://github.com/GeneralMills/PyTrends.\u003c/li\u003e\n\u003cli\u003eSharma, P., \u0026amp; Sharma, D. (2022). Classification Reports: Essential Tools for Sentiment Analysis Model Evaluation. arXiv preprint arXiv:2208.03906.\u003c/li\u003e\n\u003cli\u003eSutskever, I., Vinyals, O., \u0026amp; Le, Q. V. (2014). Sequence to sequence learning with neural networks. In Advances in neural information processing systems (pp. 3104-3112).\u003c/li\u003e\n\u003cli\u003eValencia, F., G\u0026oacute;mez-Espinosa, A., \u0026amp; Vald\u0026eacute;s-Aguirre, B. (2019). Price movement prediction of cryptocurrencies using sentiment analysis and machine learning. \u003cem\u003eEntropy\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(6), 589. https://doi.org/10.3390/e21060589\u003c/li\u003e\n\u003cli\u003eWang, S., \u0026amp; Manning, C. D. (2012). Baselines and bigrams: Simple, good sentiment and topic classification. In Proceedings of the 50th annual meeting of the association for computational linguistics: Short papers-volume 2 (pp. 90-94).\u003c/li\u003e\n\u003cli\u003eWang, Y., Sun, Y., Liu, T., \u0026amp; Huang, X. (2016). A CNN-LSTM based model for text classification. arXiv preprint\u003c/li\u003e\n\u003cli\u003eWołk, K. (2020). Advanced social media sentiment analysis for short‐term cryptocurrency price prediction. \u003cem\u003eExpert Systems\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(2), e12493.\u003c/li\u003e\n\u003cli\u003eYahoo! (n.d.). \u003cem\u003eYahoo Finance - Stock Market Live, quotes, Business \u0026amp; Finance News\u003c/em\u003e. Yahoo! Finance. https://finance.yahoo.com/\u003c/li\u003e\n\u003cli\u003eYamashita, R., Nishida, Y., Kido, R., \u0026amp; Akita, K. (2018). Convolutional neural networks: an overview and applications in medical image analysis. In Medical Imaging Informatics (pp. 449-483). Springer, Cham.\u003c/li\u003e\n\u003cli\u003eZhang, X., Lai, L., Xu, C., \u0026amp; Liu, J. (2015). CNN-LSTM neural networks for sentence-level sentiment classification. arXiv preprint arXiv:1511.05352.\u003c/li\u003e\n\u003cli\u003eZhang, Y., Wallace, B., \u0026amp; Wang, D. (2016). Rationale-augmented convolutional neural networks for text classification. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing (pp. 795-804).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Sentiment analysis, Google Trends, Russian-Ukrainian War, Pysentimento, CNN-LSTM, SVM","lastPublishedDoi":"10.21203/rs.3.rs-3949248/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3949248/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSentiment analysis is a powerful tool for extracting valuable insights from social media data. In this paper, more than one million tweets spanning three months (March, June, and December 2022) regarding three cryptocurrencies: Bitcoin (BTC), Ethereum (ETH), and Binance Coin (BNB) during the Russian-Ukrainian War are considered. Two models, a convolutional neural network with long short-term memory (CNN-LSTM) and a support vector machine (SVM) with GloVe and TF-IDF features, are trained on a labeled dataset of more than fifty thousand tweets about Bitcoin labeled as (positive, negative, and neutral). A pretrained model (Pysentimento) for sentiment analysis is also employed to compare the performances of the three models. The models are tested on the labeled dataset and then evaluated on the unlabeled tweets, revealing that Pysentimento's level of accuracy outperforms the other two models. Google Trends, along with the opening and closing prices, and the volume of the three cryptocurrencies, in addition to the results of Pysentimento sentiment classification, are employed to apply the Pearson correlation coefficient and conduct price prediction analysis using the SARIMA model. It is found that Bitcoin may appeal to those seeking stability and a known record of accomplishment, while Binance Coin and Ethereum may attract investors looking for more diverse opportunities. Sentiment analysis using machine learning is found to provide invaluable information for cryptocurrency price forecasting and trading strategies, especially in the context of geopolitical events and market volatility.\u003c/p\u003e","manuscriptTitle":"Predicting Cryptocurrency Prices during Periods of Conflict: A Comparative Sentiment Analysis Approach Using SVM, CNN-LSTM, and PySentimento","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-02-14 07:28:44","doi":"10.21203/rs.3.rs-3949248/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e0f0c987-88e7-4d58-9440-41a5dee59496","owner":[],"postedDate":"February 14th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-02-20T06:22:18+00:00","versionOfRecord":[],"versionCreatedAt":"2024-02-14 07:28:44","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3949248","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3949248","identity":"rs-3949248","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}