Sea Animal Image Classification Using Machine Learning Algorithms for Accurate and Scalable Prediction

Sea Animal Image Classification Using Machine Learning Algorithms for Accurate and Scalable Prediction | 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 Sea Animal Image Classification Using Machine Learning Algorithms for Accurate and Scalable Prediction Mohammad Subhi Al-Batah, Mowafaq Salem Alzboon This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7387846/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 10 Mar, 2026 Read the published version in Discover Artificial Intelligence → Version 1 posted 10 You are reading this latest preprint version Abstract This study presents a comparative analysis of machine learning algorithms for classifying sea animal images using Orange Data Mining. The dataset, consisting of labeled images categorized into three classes—squid, statfish, and whale—was sourced from an open-access repository and processed using supervised learning workflows. Various models, including Neural Networks, Support Vector Machines (SVM), Random Forests, k-Nearest Neighbors (kNN), Logistic Regression, Naïve Bayes, Gradient Boosting, and AdaBoost, were evaluated using 10-fold cross-validation. Performance was assessed across multiple metrics: Area Under the Curve (AUC), Classification Accuracy (CA), F1-score, Precision, Recall, and LogLoss. The Neural Network model yielded the best overall performance with an AUC of 0.990 and a classification accuracy of 93.2%. SVM and Logistic Regression closely followed, outperforming other traditional and ensemble methods. Confusion matrix analysis further supported these findings, demonstrating low misclassification rates for Neural Networks. ROC curve evaluations for individual classes confirmed the robustness of top-performing models. The findings validate the effectiveness of low-code platforms like Orange in streamlining image classification pipelines for ecological and biological image datasets. This study provides valuable insights for researchers aiming to deploy interpretable and scalable machine learning solutions in marine biology and related domains. Machine learning image classification sea animal dataset Orange Data Mining neural networks Figures Figure 1 Figure 2 Figure 3 1. Introduction The rapid advancement in computer vision and machine learning technologies has significantly transformed image classification tasks in various scientific domains, including marine biology. In underwater ecosystems, visual data is essential for species identification, biodiversity monitoring, and ecological studies (LeCun, Bengio, & Hinton, 2015 ). Manual classification of sea animal images is both labor-intensive and prone to inconsistency, especially when dealing with large-scale datasets or morphologically similar species (Tuia et al., 2016 ). Machine learning (ML), particularly supervised learning algorithms, offers promising solutions by automating classification based on learned patterns from labeled data. Traditional algorithms such as Support Vector Machines (SVM), Random Forests, and k-Nearest Neighbors (kNN), as well as ensemble models and deep neural networks, have demonstrated excellent performance in image-based tasks (Krizhevsky, Sutskever, & Hinton, 2012 ; Domingos, 2012 ). The effectiveness of these models depends on a variety of factors including dataset quality, feature representation, model architecture, and evaluation methodology (Goodfellow, Bengio, & Courville, 2016 ). Orange Data Mining, a low-code visual machine learning platform, has emerged as a practical tool for building data analysis workflows without requiring extensive programming skills (Demsar et al., 2013 ). With a drag-and-drop interface and integrated support for powerful algorithms, Orange facilitates rapid model development, evaluation, and visualization, making it particularly useful for educational and applied research contexts (Curk et al., 2005 ). This study aims to evaluate the performance of several machine learning algorithms within Orange for the task of sea animal image classification. The research contributes to the field in the following ways: It provides a systematic evaluation of nine popular classification algorithms using a real-world sea animal image dataset with 10-fold cross-validation. It compares the models using performance metrics including Area Under the Curve (AUC), Classification Accuracy (CA), F1-score, Precision, Recall, and LogLoss. It analyzes confusion matrices and Receiver Operating Characteristic (ROC) curves to visualize class-wise performance. It demonstrates the effectiveness and accessibility of Orange as a low-code platform for marine image classification tasks. By benchmarking different models and using Orange’s visual interface, this study aims to support researchers in choosing suitable machine learning techniques for ecological applications. 2. Related Work Image classification of marine species is a growing area of interest, particularly for ecological conservation, fish stock monitoring, and automated environmental assessments. The challenge lies in identifying species from images taken under diverse lighting, angles, and occlusion conditions—factors which demand robust classification models. In recent years, deep learning models such as Convolutional Neural Networks (CNNs) have shown superior performance in image classification tasks, including marine environments. For instance, Salman et al. ( 2019 ) achieved over 94% accuracy using CNNs to classify fish species in underwater imagery. Villon et al. ( 2018 ) compared deep CNN models with traditional feature-based methods like Histogram of Oriented Gradients (HOG) paired with Support Vector Machines (SVM), showing that deep learning yielded improved generalization, especially in noisy underwater conditions. However, CNNs are computationally expensive and require large datasets to perform optimally. For researchers dealing with limited labeled data or lacking GPU resources, classical machine learning models remain a strong alternative. Maji et al. ( 2019 ) demonstrated that Random Forests and SVMs could classify plankton species with accuracy comparable to CNNs when domain-specific features were extracted. Similarly, Sethi et al. ( 2020 ) highlighted that decision trees and ensemble models offered interpretable results, which is advantageous in ecological studies where model transparency is crucial. Ensemble methods such as AdaBoost, Gradient Boosting, and Random Forests have gained popularity for their robustness and versatility across domains. In the context of fish species classification, Tang et al. ( 2020 ) found that Gradient Boosting outperformed single classifiers in terms of both precision and recall. These ensemble techniques combine weak learners to form a strong predictor, reducing variance and mitigating overfitting. Despite the broad adoption of machine learning for biological classification, few studies have explored low-code environments like Orange Data Mining in this context. Orange offers drag-and-drop interfaces and integrates key algorithms such as neural networks, SVM, decision trees, and ensemble learners, allowing researchers without coding experience to develop full ML pipelines (Toplak et al., 2017 ). While prior applications of Orange focus on clinical and business datasets (Curk et al., 2005 ), its potential in marine image classification remains underutilized. This paper addresses this gap by comparing nine supervised learning models using Orange on a real-world sea animal image dataset. Unlike prior studies that emphasize a single architecture, this work presents a well-rounded benchmark across traditional, ensemble, and neural classifiers using six evaluation metrics, confusion matrices, and ROC curve analysis. 3. Methodology 3.1 Dataset Description The dataset used in this study is the Sea Animals Image Dataset obtained from Kaggle ( https://www.kaggle.com/datasets/vencerlanz09/sea-animals-image-dataste ). It comprises 5,290 images of marine animals categorized into three classes: Squid , Statfish , and Whale . The images are organized into class-labeled folders, which makes them suitable for supervised learning. Each image represents one of the animal types in various underwater settings, simulating real-world scenarios such as varying lighting conditions, orientations, and backgrounds. 3.2 Data Preprocessing The images were preprocessed using Orange’s built-in image embedding tool, which converts image data into feature vectors using a pre-trained deep learning model (Inception-v3). This transformation enables classical machine learning algorithms to process visual features effectively without manually extracting them. The resulting dataset was then normalized to ensure consistent scale across features and shuffled to prevent ordering bias. 3.3 Platform: Orange Data Mining Orange Data Mining is an open-source visual programming environment built in Python that enables users to design machine learning pipelines through an intuitive drag-and-drop interface. In this study, Orange version 3.34 was used. The workflow comprised the following components: Image Embedding: For feature extraction from images using Inception-v3. Data Table and Preprocessing Widgets: To clean and normalize data. Test & Score: For model evaluation using stratified 10-fold cross-validation. Classification Algorithms: Connected in parallel to allow comparative evaluation. Confusion Matrix and ROC Analysis: For detailed performance visualization. 3.4 Machine Learning Models Evaluated Nine classification models were tested using Orange's native implementations: Neural Network (Multilayer Perceptron) Support Vector Machine (SVM) k-Nearest Neighbors (kNN) Random Forest Decision Tree Naïve Bayes Logistic Regression AdaBoosting Gradient Boosting An additional constant classifier was included as a performance baseline. 3.5 Evaluation Protocol All models were evaluated using 10-fold cross-validation to ensure generalizability. Performance was assessed using the following metrics: Area Under the Curve (AUC): Measures the trade-off between true positive and false positive rates. Classification Accuracy (CA): The percentage of correct predictions. F1-Score: Harmonic mean of precision and recall. Precision: Ratio of true positives to predicted positives. Recall: Ratio of true positives to actual positives. Logarithmic Loss (LogLoss): Measures the confidence of predictions. These metrics provide a comprehensive view of each model’s predictive behavior, from raw accuracy to probabilistic reliability. 4. Results This section reports the empirical evaluation of ten supervised machine learning classifiers trained and tested on the Sea Animal Image Dataset using Orange Data Mining. Each model was assessed using 10-fold cross-validation, and results were computed as averages across all three classes: Squid , Statfish , and Whale . The evaluation focused on metrics including Classification Accuracy (CA), Area Under the Curve (AUC), F1-score, Precision, Recall, and Logarithmic Loss (LogLoss). 4.1 Performance Evaluation Metrics The average classification performance of each model is summarized in Table 1 . Among all models, the Neural Network achieved the highest results, with an AUC of 0.990 and a CA of 93.2%. It also had the lowest LogLoss (0.219), indicating high predictive confidence. SVM and Logistic Regression followed closely with CA values of 92.6% and AUCs of 0.988 and 0.985, respectively. In contrast, Naïve Bayes and Decision Tree models showed significantly higher LogLoss and lower accuracy, highlighting their limitations in handling complex image embeddings. Table 1 Test and Score Analyses Based on Orange Data Mining Using 10-Fold Cross Validation (Averaged Over All Classes) Model AUC CA F1 Precision Recall LogLoss Neural Network 0.990 0.932 0.932 0.932 0.932 0.219 SVM 0.988 0.926 0.925 0.926 0.926 0.189 Logistic Regression 0.985 0.926 0.926 0.926 0.926 0.390 Gradient Boosting 0.983 0.913 0.913 0.913 0.913 0.236 kNN 0.972 0.909 0.908 0.910 0.909 0.898 Random Forest 0.967 0.880 0.880 0.880 0.880 0.462 AdaBoosting 0.912 0.859 0.860 0.860 0.859 0.794 Tree 0.862 0.843 0.843 0.842 0.843 3.593 Naïve Bayes 0.932 0.826 0.824 0.823 0.826 4.479 Constant 0.500 0.369 0.199 0.136 0.369 1.096 4.2 Confusion Matrix Analysis Table 2 presents the confusion matrices for each model, outlining how accurately each class ( Squid , Statfish , Whale ) was predicted. The Neural Network model showed the lowest number of misclassifications across all classes, correctly classifying 1444 out of 1640 Squid , 1652 out of 1700 Statfish , and 1835 out of 1950 Whale samples. Table 2 Confusion Matrix Analyses for the Models: Actual and Predicted Counts Model Actual Class Predicted Squid Predicted Statfish Predicted Whale Neural Network Squid 1444 52 144 Statfish 37 1652 11 Whale 108 7 1835 SVM Squid 1135 142 363 Statfish 89 1598 13 Whale 263 49 1638 Logistic Regression Squid 1447 41 152 Statfish 46 1642 12 Whale 139 3 1808 Gradient Boosting Squid 1388 65 187 Statfish 54 1621 25 Whale 123 7 1820 Random Forest Squid 1337 76 227 Statfish 70 1603 27 Whale 216 12 1722 kNN Squid 1614 68 258 Statfish 42 1648 10 Whale 93 9 1848 AdaBoost Squid 1300 87 253 Statfish 105 1560 35 Whale 252 13 1685 Tree Squid 1230 100 310 Statfish 83 1596 21 Whale 291 24 1635 Naïve Bayes Squid 1135 142 363 Statfish 89 1598 13 Whale 263 49 1638 4.3 ROC Curve Analysis ROC curves were generated for each of the three target classes— Squid , Statfish , and Whale —to compare classifier discrimination capabilities. Figures 1 – 3 show that the Neural Network and SVM achieved the steepest ROC curves, indicating a stronger true positive rate with fewer false positives. This was especially notable for the Whale class, where the Neural Network model approached ideal ROC performance. All ROC figures compare the performance of the top four models: Neural Network, SVM, Gradient Boosting, and Random Forest. 5. Discussion and Comparison 5.1 Interpretation of Model Performance The experimental results confirmed the Neural Network model as the most effective classifier for the Sea Animal Image Dataset, achieving the highest AUC (0.990), accuracy (93.2%), and the lowest LogLoss (0.219). These outcomes reflect the model’s ability to effectively generalize across classes using rich image embeddings. Support Vector Machine (SVM) and Logistic Regression also showed competitive results, both achieving 92.6% accuracy. Their performance illustrates the strength of classical models when applied to high-quality, embedded feature representations. Ensemble methods such as Gradient Boosting and Random Forest offered slightly lower, yet robust, performance. In contrast, Naïve Bayes and Decision Tree models showed higher error rates, reflecting their limitations in complex visual domains. 5.2 Literature Comparison Table 3 presents a performance comparison with existing studies. While deep learning models in prior research slightly outperform the current approach in some cases, the low-code workflow used here achieved similar accuracy with significantly reduced implementation complexity. Table 3 Performance Comparison with Selected Literature Study / Model Accuracy (%) Dataset Description Notes This Study-Neural Network 93.2 5,290 images (3 classes) Visual ML with Inception-v3 embeddings Salman et al. ( 2019 ) – CNN 94.0 ~ 4,000 underwater fish images Deep learning, custom code Villon et al. ( 2018 ) – HOG + SVM 87.0 3,800 coral reef images Feature-engineered SVM Maji et al. ( 2019 ) – RF/SVM 90.0 2,000 plankton samples Classical ML with handcrafted features These results confirm that Orange, combined with embedded features, enables classical models to perform comparably with more complex deep learning pipelines, making it a practical solution for rapid deployment or educational use. 5.3 Strengths and Limitations Strengths : Comprehensive benchmarking of ten models using standardized metrics. Use of a low-code platform (Orange), supporting ease of experimentation. Class-specific insights via confusion matrices and ROC curves. Limitations : The dataset covered only three classes, limiting generalization to more complex tasks. Orange offers limited hyperparameter tuning compared to full-code frameworks. Future improvements could involve automated augmentation or model stacking techniques. 6. Conclusion This study conducted a comprehensive evaluation of ten machine learning algorithms for the classification of sea animal images using Orange Data Mining. Leveraging a real-world dataset comprising Squid , Statfish , and Whale images, the models were assessed through 10-fold cross-validation across six performance metrics. Among the tested classifiers, the Neural Network demonstrated superior performance, achieving an AUC of 0.990 and a classification accuracy of 93.2%. Traditional models such as SVM and Logistic Regression also performed competitively when coupled with Inception-v3 image embeddings. Confusion matrix and ROC analyses confirmed the consistency of top-performing models across all classes. The study further illustrated that accessible low-code platforms like Orange can effectively support high-quality image classification, making advanced machine learning tools usable by non-programmers and interdisciplinary researchers. This approach holds promise for educational use, ecological monitoring, and rapid prototyping in image-based tasks. Future work will focus on extending this research to larger and more diverse marine datasets, integrating advanced augmentation techniques, and exploring hybrid workflows that combine the usability of Orange with the fine-tuning flexibility of code-based environments. Declarations Funding This research received no external funding. Acknowledgment The author would like to thank the team at Jadara University and the scientific research deanship for their support and encouragement throughout this study. Appreciation is also extended to the Orange Data Mining community for providing an open and accessible platform that enabled this research. Conflict of Interest The author declares no conflict of interest. Consent for publication: Not applicable. Consent to participate: Not applicable. Ethics approval: Not applicable. Author Contribution Mohammad Subhi Al-Batah: Conceptualization, Methodology, Data Analysis, Writing – Original Draft Preparation, Supervision.Mowafaq Salem Alzboon: Literature Review, Data Curation, Visualization, Writing – Review & Editing.Both authors have read and approved the final manuscript. References Curk T, Demsar J, Xu Q, Leban G, Petrovic U, Bratko I, Zupan B. (2005). Microarray data mining with visual programming. Bioinformatics, 21(3), 396–8. https://doi.org/10.1093/bioinformatics/bti016 Demsar J, Curk T, Erjavec A, Gorup Č, Hočevar T, Milutinovič M, Zupan B. (2013). Orange: Data mining toolbox in Python. Journal of Machine Learning Research, 14, 2349–53. Domingos P. (2012). A few useful things to know about machine learning. Communications of the ACM, 55(10), 78–87. https://doi.org/10.1145/2347736.2347755 Goodfellow I, Bengio Y, Courville A. (2016). Deep learning. MIT Press. Krizhevsky A, Sutskever I, Hinton GE. (2012). ImageNet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, 25, 1097–105. LeCun Y, Bengio Y, Hinton G. (2015). Deep learning. Nature, 521(7553), 436–44. https://doi.org/10.1038/nature14539 Maji S, Pusdekar R, Shukla A, Bandyopadhyay S. (2019). Comparative study of machine learning algorithms for marine plankton classification. Procedia Computer Science, 167, 2151–60. https://doi.org/10.1016/j.procs.2020.03.266 Salman A, Jalal A, Shamsher R, Kim K. (2019). Fish species classification in underwater environments using deep learning. Sensors, 19(3), 545. https://doi.org/10.3390/s19030545 Sethi SS, Ewers RM, Jones NS, Orme CD, Picinali L. (2020). Robust, real-time and autonomous monitoring of ecosystems with an open, low-cost, networked device. Methods in Ecology and Evolution, 11(7), 885–95. https://doi.org/10.1111/2041-210X.13361 Tang Y, Hu X, Wang Y, Zhang X. (2020). Application of ensemble learning in fish species recognition. Applied Artificial Intelligence, 34(5), 412–29. https://doi.org/10.1080/08839514.2020.1762963 Toplak M, Curk T, Demšar J, Zupan B. (2017). Orange: From experimental machine learning to interactive data analysis. KDD Tutorials. https://doi.org/10.1145/3097983.3097996 Tuia D, Volpi M, Copa L, Kanevski M, Munoz-Mari J. (2016). A survey of active learning algorithms for supervised remote sensing image classification. IEEE Journal of Selected Topics in Signal Processing, 10(2), 606–17. https://doi.org/10.1109/JSTSP.2015.2509170 Villon S, Chaumont M, Subsol G, Claverie T, Villéger S. (2018). Coral reef fish detection and recognition in underwater videos by supervised machine learning: Comparison between Deep Learning and HOG + SVM methods. Ecological Informatics, 48, 121–30. https://doi.org/10.1016/j.ecoinf.2018.10.005 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 10 Mar, 2026 Read the published version in Discover Artificial Intelligence → Version 1 posted Editorial decision: Revision requested 24 Nov, 2025 Reviews received at journal 19 Nov, 2025 Reviewers agreed at journal 12 Nov, 2025 Reviewers agreed at journal 20 Oct, 2025 Reviews received at journal 28 Sep, 2025 Reviewers agreed at journal 26 Sep, 2025 Reviewers invited by journal 18 Sep, 2025 Editor assigned by journal 30 Aug, 2025 Submission checks completed at journal 30 Aug, 2025 First submitted to journal 16 Aug, 2025 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-7387846","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":521715268,"identity":"701ffe37-903f-4790-83dd-18b3bb0e72a2","order_by":0,"name":"Mohammad Subhi Al-Batah","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+0lEQVRIiWNgGAWjYDACdh4QeUCOsYHBACJygAHGwgGYeUCKDhiTriWxgQFJC17A38x78POHmjvpze2HN36uqGGQ47uRwFDwAY8WicN8yRIHjj3LbexJK5Y8c4zBWBKoxXAGPmsO8xhIHGA7nNvYkGMg2cDGkLgBqMWYB48O+cM8xj8O/Ducztj/xvhnwz+GeoJaDA7zmEkcbDucwDgjx0yysY0hwYCQFsPDfGkWZ/sOGzbOeFZm2dgnYTjzzMMGvH6RO957+EbFt8Pyhv3Jm282fLOR5zuefMwAX4ghrGsAUxJAzNiGPyphQB6JzfyAKC2jYBSMglEwUgAApCZWVMhy9IUAAAAASUVORK5CYII=","orcid":"","institution":"Jadara University","correspondingAuthor":true,"prefix":"","firstName":"Mohammad","middleName":"Subhi","lastName":"Al-Batah","suffix":""},{"id":521715269,"identity":"da3d22c9-eec9-4bf4-b1c7-46607ad3f9f9","order_by":1,"name":"Mowafaq Salem Alzboon","email":"","orcid":"","institution":"Jadara University","correspondingAuthor":false,"prefix":"","firstName":"Mowafaq","middleName":"Salem","lastName":"Alzboon","suffix":""}],"badges":[],"createdAt":"2025-08-16 13:53:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7387846/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7387846/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s44163-026-01005-9","type":"published","date":"2026-03-10T15:59:09+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":92475846,"identity":"fd217bca-2be8-4ce8-acb4-e820ad81a488","added_by":"auto","created_at":"2025-09-30 07:18:08","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":410625,"visible":true,"origin":"","legend":"","description":"","filename":"SeaAnimalImageClassificationUsingMachineLearningAlgorithmsforAccurateandScalablePrediction.docx","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/4b42cc2ce65df04a7033da1b.docx"},{"id":92475848,"identity":"04857045-c374-4939-a950-034401a9c5db","added_by":"auto","created_at":"2025-09-30 07:18:08","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4549,"visible":true,"origin":"","legend":"","description":"","filename":"b34770d9d83e4f9092b29aff0b022684.json","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/37c0f2169f7c72275c0636b1.json"},{"id":92478587,"identity":"76660cbd-54f1-4560-abc9-2f07aa7c3cfd","added_by":"auto","created_at":"2025-09-30 07:34:08","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":66214,"visible":true,"origin":"","legend":"","description":"","filename":"b34770d9d83e4f9092b29aff0b0226841enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/5438b44b8deb7a5128f55c88.xml"},{"id":92475844,"identity":"4cf35b19-fc55-43f8-80a5-48bf5d0c9fb3","added_by":"auto","created_at":"2025-09-30 07:18:08","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":62808,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/b3613dbbc8398a3354228ac1.jpeg"},{"id":92475851,"identity":"0c43f328-33ba-47e2-b37b-23ce1cef21c0","added_by":"auto","created_at":"2025-09-30 07:18:08","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10434,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/0572c1b8b65e7a3d6e20809a.png"},{"id":92475857,"identity":"9bfa3e63-36b6-472f-b57c-2cdc850de9cf","added_by":"auto","created_at":"2025-09-30 07:18:08","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34156,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/a8243ed49e2bd4ad2679ee9b.png"},{"id":92478027,"identity":"c4dab434-dac9-4fe1-9d2a-f1507a8ddca4","added_by":"auto","created_at":"2025-09-30 07:26:08","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28213,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/7765be443eb883043e44f320.png"},{"id":92478588,"identity":"c05264bd-af42-4e34-ab5b-f671df8adc1d","added_by":"auto","created_at":"2025-09-30 07:34:08","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31821,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/1698d0b9727111d71c0c2938.png"},{"id":92475855,"identity":"2ec9dfeb-1dec-4161-83b0-1aee8b29c03f","added_by":"auto","created_at":"2025-09-30 07:18:08","extension":"xml","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":65704,"visible":true,"origin":"","legend":"","description":"","filename":"b34770d9d83e4f9092b29aff0b0226841structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/675aa42121c2faa14af93d7d.xml"},{"id":92478025,"identity":"c734913a-acf1-4536-837d-69c75ec05305","added_by":"auto","created_at":"2025-09-30 07:26:08","extension":"html","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":70975,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/9cef9e986b932fe2128dc163.html"},{"id":92475856,"identity":"d51f1edb-335b-41f4-ae76-1305f80b6930","added_by":"auto","created_at":"2025-09-30 07:18:08","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":132619,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/447ae4bcae5817f474833e20.png"},{"id":92478023,"identity":"90d3aaff-5002-4135-80af-b5025b6c70ce","added_by":"auto","created_at":"2025-09-30 07:26:08","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":106456,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/625b1072db0ed7dca2bdba30.png"},{"id":92475849,"identity":"8e8206f6-b1fc-4205-a4cc-25e95dcd6d78","added_by":"auto","created_at":"2025-09-30 07:18:08","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":122868,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/2ef9a9cc71bb640427403c17.png"},{"id":104739399,"identity":"b8a60235-a823-405d-aaab-6641677b3339","added_by":"auto","created_at":"2026-03-16 16:05:48","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1129476,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7387846/v1/b810befd-9a46-4481-bc1f-a21e5db460c5.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Sea Animal Image Classification Using Machine Learning Algorithms for Accurate and Scalable Prediction","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe rapid advancement in computer vision and machine learning technologies has significantly transformed image classification tasks in various scientific domains, including marine biology. In underwater ecosystems, visual data is essential for species identification, biodiversity monitoring, and ecological studies (LeCun, Bengio, \u0026amp; Hinton, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Manual classification of sea animal images is both labor-intensive and prone to inconsistency, especially when dealing with large-scale datasets or morphologically similar species (Tuia et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eMachine learning (ML), particularly supervised learning algorithms, offers promising solutions by automating classification based on learned patterns from labeled data. Traditional algorithms such as Support Vector Machines (SVM), Random Forests, and k-Nearest Neighbors (kNN), as well as ensemble models and deep neural networks, have demonstrated excellent performance in image-based tasks (Krizhevsky, Sutskever, \u0026amp; Hinton, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Domingos, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). The effectiveness of these models depends on a variety of factors including dataset quality, feature representation, model architecture, and evaluation methodology (Goodfellow, Bengio, \u0026amp; Courville, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eOrange Data Mining, a low-code visual machine learning platform, has emerged as a practical tool for building data analysis workflows without requiring extensive programming skills (Demsar et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). With a drag-and-drop interface and integrated support for powerful algorithms, Orange facilitates rapid model development, evaluation, and visualization, making it particularly useful for educational and applied research contexts (Curk et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThis study aims to evaluate the performance of several machine learning algorithms within Orange for the task of sea animal image classification. The research contributes to the field in the following ways:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eIt provides a systematic evaluation of nine popular classification algorithms using a real-world sea animal image dataset with 10-fold cross-validation.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eIt compares the models using performance metrics including Area Under the Curve (AUC), Classification Accuracy (CA), F1-score, Precision, Recall, and LogLoss.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eIt analyzes confusion matrices and Receiver Operating Characteristic (ROC) curves to visualize class-wise performance.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eIt demonstrates the effectiveness and accessibility of Orange as a low-code platform for marine image classification tasks.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eBy benchmarking different models and using Orange\u0026rsquo;s visual interface, this study aims to support researchers in choosing suitable machine learning techniques for ecological applications.\u003c/p\u003e"},{"header":"2. Related Work","content":"\u003cp\u003eImage classification of marine species is a growing area of interest, particularly for ecological conservation, fish stock monitoring, and automated environmental assessments. The challenge lies in identifying species from images taken under diverse lighting, angles, and occlusion conditions\u0026mdash;factors which demand robust classification models.\u003c/p\u003e\u003cp\u003eIn recent years, deep learning models such as Convolutional Neural Networks (CNNs) have shown superior performance in image classification tasks, including marine environments. For instance, Salman et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) achieved over 94% accuracy using CNNs to classify fish species in underwater imagery. Villon et al. (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) compared deep CNN models with traditional feature-based methods like Histogram of Oriented Gradients (HOG) paired with Support Vector Machines (SVM), showing that deep learning yielded improved generalization, especially in noisy underwater conditions.\u003c/p\u003e\u003cp\u003eHowever, CNNs are computationally expensive and require large datasets to perform optimally. For researchers dealing with limited labeled data or lacking GPU resources, classical machine learning models remain a strong alternative. Maji et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) demonstrated that Random Forests and SVMs could classify plankton species with accuracy comparable to CNNs when domain-specific features were extracted. Similarly, Sethi et al. (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) highlighted that decision trees and ensemble models offered interpretable results, which is advantageous in ecological studies where model transparency is crucial.\u003c/p\u003e\u003cp\u003eEnsemble methods such as AdaBoost, Gradient Boosting, and Random Forests have gained popularity for their robustness and versatility across domains. In the context of fish species classification, Tang et al. (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) found that Gradient Boosting outperformed single classifiers in terms of both precision and recall. These ensemble techniques combine weak learners to form a strong predictor, reducing variance and mitigating overfitting.\u003c/p\u003e\u003cp\u003eDespite the broad adoption of machine learning for biological classification, few studies have explored low-code environments like Orange Data Mining in this context. Orange offers drag-and-drop interfaces and integrates key algorithms such as neural networks, SVM, decision trees, and ensemble learners, allowing researchers without coding experience to develop full ML pipelines (Toplak et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). While prior applications of Orange focus on clinical and business datasets (Curk et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), its potential in marine image classification remains underutilized.\u003c/p\u003e\u003cp\u003eThis paper addresses this gap by comparing nine supervised learning models using Orange on a real-world sea animal image dataset. Unlike prior studies that emphasize a single architecture, this work presents a well-rounded benchmark across traditional, ensemble, and neural classifiers using six evaluation metrics, confusion matrices, and ROC curve analysis.\u003c/p\u003e"},{"header":"3. Methodology","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e3.1 Dataset Description\u003c/h2\u003e\u003cp\u003eThe dataset used in this study is the Sea Animals Image Dataset obtained from Kaggle (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.kaggle.com/datasets/vencerlanz09/sea-animals-image-dataste\u003c/span\u003e\u003cspan address=\"https://www.kaggle.com/datasets/vencerlanz09/sea-animals-image-dataste\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e).\u003c/span\u003e It comprises 5,290 images of marine animals categorized into three classes: \u003cb\u003eSquid\u003c/b\u003e, \u003cb\u003eStatfish\u003c/b\u003e, and \u003cb\u003eWhale\u003c/b\u003e. The images are organized into class-labeled folders, which makes them suitable for supervised learning. Each image represents one of the animal types in various underwater settings, simulating real-world scenarios such as varying lighting conditions, orientations, and backgrounds.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Data Preprocessing\u003c/h2\u003e\u003cp\u003eThe images were preprocessed using Orange\u0026rsquo;s built-in image embedding tool, which converts image data into feature vectors using a pre-trained deep learning model (Inception-v3). This transformation enables classical machine learning algorithms to process visual features effectively without manually extracting them. The resulting dataset was then normalized to ensure consistent scale across features and shuffled to prevent ordering bias.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Platform: Orange Data Mining\u003c/h2\u003e\u003cp\u003eOrange Data Mining is an open-source visual programming environment built in Python that enables users to design machine learning pipelines through an intuitive drag-and-drop interface. In this study, Orange version 3.34 was used. The workflow comprised the following components:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eImage Embedding: For feature extraction from images using Inception-v3.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eData Table and Preprocessing Widgets: To clean and normalize data.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eTest \u0026amp; Score: For model evaluation using stratified 10-fold cross-validation.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eClassification Algorithms: Connected in parallel to allow comparative evaluation.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eConfusion Matrix and ROC Analysis: For detailed performance visualization.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Machine Learning Models Evaluated\u003c/h2\u003e\u003cp\u003eNine classification models were tested using Orange's native implementations:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eNeural Network (Multilayer Perceptron)\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eSupport Vector Machine (SVM)\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003ek-Nearest Neighbors (kNN)\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eRandom Forest\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eDecision Tree\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eNa\u0026iuml;ve Bayes\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eLogistic Regression\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eAdaBoosting\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eGradient Boosting\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eAn additional constant classifier was included as a performance baseline.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3.5 Evaluation Protocol\u003c/h2\u003e\u003cp\u003eAll models were evaluated using \u003cb\u003e10-fold cross-validation\u003c/b\u003e to ensure generalizability. Performance was assessed using the following metrics:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eArea Under the Curve (AUC): Measures the trade-off between true positive and false positive rates.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eClassification Accuracy (CA): The percentage of correct predictions.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eF1-Score: Harmonic mean of precision and recall.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003ePrecision: Ratio of true positives to predicted positives.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eRecall: Ratio of true positives to actual positives.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eLogarithmic Loss (LogLoss): Measures the confidence of predictions.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eThese metrics provide a comprehensive view of each model\u0026rsquo;s predictive behavior, from raw accuracy to probabilistic reliability.\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Results","content":"\u003cp\u003eThis section reports the empirical evaluation of ten supervised machine learning classifiers trained and tested on the Sea Animal Image Dataset using Orange Data Mining. Each model was assessed using 10-fold cross-validation, and results were computed as averages across all three classes: \u003cem\u003eSquid\u003c/em\u003e, \u003cem\u003eStatfish\u003c/em\u003e, and \u003cem\u003eWhale\u003c/em\u003e. The evaluation focused on metrics including Classification Accuracy (CA), Area Under the Curve (AUC), F1-score, Precision, Recall, and Logarithmic Loss (LogLoss).\u003c/p\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e4.1 Performance Evaluation Metrics\u003c/h2\u003e\u003cp\u003eThe average classification performance of each model is summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Among all models, the Neural Network achieved the highest results, with an AUC of 0.990 and a CA of 93.2%. It also had the lowest LogLoss (0.219), indicating high predictive confidence. SVM and Logistic Regression followed closely with CA values of 92.6% and AUCs of 0.988 and 0.985, respectively. In contrast, Na\u0026iuml;ve Bayes and Decision Tree models showed significantly higher LogLoss and lower accuracy, highlighting their limitations in handling complex image embeddings.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTest and Score Analyses Based on Orange Data Mining Using 10-Fold Cross Validation (Averaged Over All Classes)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eModel\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAUC\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCA\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eF1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePrecision\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eRecall\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eLogLoss\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNeural Network\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.990\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.932\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.932\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.932\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.932\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.219\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSVM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.988\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.926\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.925\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.926\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.926\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.189\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLogistic Regression\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.985\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.926\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.926\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.926\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.926\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.390\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGradient Boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.983\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.913\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.913\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.913\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.913\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.236\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ekNN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.972\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.909\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.908\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.910\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.909\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.898\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRandom Forest\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.967\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.880\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.880\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.880\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.880\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.462\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAdaBoosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.912\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.859\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.860\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.860\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.859\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.794\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTree\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.862\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.843\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.843\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.842\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.843\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e3.593\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNa\u0026iuml;ve Bayes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.932\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.826\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.824\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.823\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.826\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e4.479\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConstant\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.500\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.369\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.199\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.136\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.369\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e1.096\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e4.2 Confusion Matrix Analysis\u003c/h2\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents the confusion matrices for each model, outlining how accurately each class (\u003cem\u003eSquid\u003c/em\u003e, \u003cem\u003eStatfish\u003c/em\u003e, \u003cem\u003eWhale\u003c/em\u003e) was predicted. The Neural Network model showed the lowest number of misclassifications across all classes, correctly classifying 1444 out of 1640 \u003cem\u003eSquid\u003c/em\u003e, 1652 out of 1700 \u003cem\u003eStatfish\u003c/em\u003e, and 1835 out of 1950 \u003cem\u003eWhale\u003c/em\u003e samples.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eConfusion Matrix Analyses for the Models: Actual and Predicted Counts\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eModel\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eActual Class\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePredicted Squid\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePredicted Statfish\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePredicted Whale\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eNeural Network\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSquid\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1444\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e144\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStatfish\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1652\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e11\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWhale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e108\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1835\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eSVM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSquid\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1135\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e142\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e363\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStatfish\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1598\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWhale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e263\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1638\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eLogistic Regression\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSquid\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1447\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e152\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStatfish\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1642\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWhale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e139\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1808\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eGradient Boosting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSquid\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1388\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e187\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStatfish\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1621\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e25\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWhale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e123\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1820\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eRandom Forest\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSquid\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1337\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e227\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStatfish\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1603\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWhale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e216\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1722\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003ekNN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSquid\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1614\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e68\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e258\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStatfish\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1648\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWhale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e93\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1848\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eAdaBoost\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSquid\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1300\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e253\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStatfish\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e105\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1560\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e35\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWhale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e252\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1685\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eTree\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSquid\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1230\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e100\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e310\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStatfish\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e83\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1596\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWhale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e291\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1635\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eNa\u0026iuml;ve Bayes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSquid\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1135\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e142\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e363\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eStatfish\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1598\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWhale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e263\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1638\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e4.3 ROC Curve Analysis\u003c/h2\u003e\u003cp\u003eROC curves were generated for each of the three target classes\u0026mdash;\u003cem\u003eSquid\u003c/em\u003e, \u003cem\u003eStatfish\u003c/em\u003e, and \u003cem\u003eWhale\u003c/em\u003e\u0026mdash;to compare classifier discrimination capabilities. Figures\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e show that the Neural Network and SVM achieved the steepest ROC curves, indicating a stronger true positive rate with fewer false positives. This was especially notable for the \u003cem\u003eWhale\u003c/em\u003e class, where the Neural Network model approached ideal ROC performance.\u003c/p\u003e\u003cp\u003eAll ROC figures compare the performance of the top four models: Neural Network, SVM, Gradient Boosting, and Random Forest.\u003c/p\u003e\u003c/div\u003e"},{"header":"5. Discussion and Comparison","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e5.1 Interpretation of Model Performance\u003c/h2\u003e\u003cp\u003eThe experimental results confirmed the \u003cb\u003eNeural Network\u003c/b\u003e model as the most effective classifier for the Sea Animal Image Dataset, achieving the highest AUC (0.990), accuracy (93.2%), and the lowest LogLoss (0.219). These outcomes reflect the model\u0026rsquo;s ability to effectively generalize across classes using rich image embeddings.\u003c/p\u003e\u003cp\u003eSupport Vector Machine (SVM) and Logistic Regression also showed competitive results, both achieving 92.6% accuracy. Their performance illustrates the strength of classical models when applied to high-quality, embedded feature representations. Ensemble methods such as Gradient Boosting and Random Forest offered slightly lower, yet robust, performance. In contrast, Na\u0026iuml;ve Bayes and Decision Tree models showed higher error rates, reflecting their limitations in complex visual domains.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e5.2 Literature Comparison\u003c/h2\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e presents a performance comparison with existing studies. While deep learning models in prior research slightly outperform the current approach in some cases, the low-code workflow used here achieved similar accuracy with significantly reduced implementation complexity.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePerformance Comparison with Selected Literature\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eStudy / Model\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAccuracy (%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDataset Description\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNotes\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eThis Study-Neural Network\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e93.2\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5,290 images (3 classes)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eVisual ML with Inception-v3 embeddings\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSalman et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) \u0026ndash; CNN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e94.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e~\u0026thinsp;4,000 underwater fish images\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDeep learning, custom code\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVillon et al. (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) \u0026ndash; HOG\u0026thinsp;+\u0026thinsp;SVM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e87.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3,800 coral reef images\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eFeature-engineered SVM\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMaji et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) \u0026ndash; RF/SVM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e90.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2,000 plankton samples\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eClassical ML with handcrafted features\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThese results confirm that Orange, combined with embedded features, enables classical models to perform comparably with more complex deep learning pipelines, making it a practical solution for rapid deployment or educational use.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e5.3 Strengths and Limitations\u003c/h2\u003e\u003cp\u003e\u003cb\u003eStrengths\u003c/b\u003e:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eComprehensive benchmarking of ten models using standardized metrics.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eUse of a low-code platform (Orange), supporting ease of experimentation.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eClass-specific insights via confusion matrices and ROC curves.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eLimitations\u003c/b\u003e:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eThe dataset covered only three classes, limiting generalization to more complex tasks.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eOrange offers limited hyperparameter tuning compared to full-code frameworks.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eFuture improvements could involve automated augmentation or model stacking techniques.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003eThis study conducted a comprehensive evaluation of ten machine learning algorithms for the classification of sea animal images using Orange Data Mining. Leveraging a real-world dataset comprising \u003cem\u003eSquid\u003c/em\u003e, \u003cem\u003eStatfish\u003c/em\u003e, and \u003cem\u003eWhale\u003c/em\u003e images, the models were assessed through 10-fold cross-validation across six performance metrics.\u003c/p\u003e\u003cp\u003eAmong the tested classifiers, the Neural Network demonstrated superior performance, achieving an AUC of 0.990 and a classification accuracy of 93.2%. Traditional models such as SVM and Logistic Regression also performed competitively when coupled with Inception-v3 image embeddings. Confusion matrix and ROC analyses confirmed the consistency of top-performing models across all classes.\u003c/p\u003e\u003cp\u003eThe study further illustrated that accessible low-code platforms like Orange can effectively support high-quality image classification, making advanced machine learning tools usable by non-programmers and interdisciplinary researchers. This approach holds promise for educational use, ecological monitoring, and rapid prototyping in image-based tasks.\u003c/p\u003e\u003cp\u003eFuture work will focus on extending this research to larger and more diverse marine datasets, integrating advanced augmentation techniques, and exploring hybrid workflows that combine the usability of Orange with the fine-tuning flexibility of code-based environments.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research received no external funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author would like to thank the team at Jadara University and the scientific research deanship for their support and encouragement throughout this study. Appreciation is also extended to the Orange Data Mining community for providing an open and accessible platform that enabled this research.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author declares no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication: Not applicable.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate: Not applicable.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval: Not applicable.\u003c/strong\u003e\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eMohammad Subhi Al-Batah: Conceptualization, Methodology, Data Analysis, Writing \u0026ndash; Original Draft Preparation, Supervision.Mowafaq Salem Alzboon: Literature Review, Data Curation, Visualization, Writing \u0026ndash; Review \u0026amp; Editing.Both authors have read and approved the final manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eCurk T, Demsar J, Xu Q, Leban G, Petrovic U, Bratko I, Zupan B. (2005). Microarray data mining with visual programming. Bioinformatics, 21(3), 396\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/bioinformatics/bti016\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/bti016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDemsar J, Curk T, Erjavec A, Gorup Č, Hočevar T, Milutinovič M, Zupan B. (2013). Orange: Data mining toolbox in Python. Journal of Machine Learning Research, 14, 2349\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDomingos P. (2012). A few useful things to know about machine learning. Communications of the ACM, 55(10), 78\u0026ndash;87. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/2347736.2347755\u003c/span\u003e\u003cspan address=\"10.1145/2347736.2347755\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGoodfellow I, Bengio Y, Courville A. (2016). Deep learning. MIT Press.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKrizhevsky A, Sutskever I, Hinton GE. (2012). ImageNet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, 25, 1097\u0026ndash;105.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeCun Y, Bengio Y, Hinton G. (2015). Deep learning. Nature, 521(7553), 436\u0026ndash;44. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature14539\u003c/span\u003e\u003cspan address=\"10.1038/nature14539\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMaji S, Pusdekar R, Shukla A, Bandyopadhyay S. (2019). Comparative study of machine learning algorithms for marine plankton classification. Procedia Computer Science, 167, 2151\u0026ndash;60. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.procs.2020.03.266\u003c/span\u003e\u003cspan address=\"10.1016/j.procs.2020.03.266\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSalman A, Jalal A, Shamsher R, Kim K. (2019). Fish species classification in underwater environments using deep learning. Sensors, 19(3), 545. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/s19030545\u003c/span\u003e\u003cspan address=\"10.3390/s19030545\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSethi SS, Ewers RM, Jones NS, Orme CD, Picinali L. (2020). Robust, real-time and autonomous monitoring of ecosystems with an open, low-cost, networked device. Methods in Ecology and Evolution, 11(7), 885\u0026ndash;95. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/2041-210X.13361\u003c/span\u003e\u003cspan address=\"10.1111/2041-210X.13361\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTang Y, Hu X, Wang Y, Zhang X. (2020). Application of ensemble learning in fish species recognition. Applied Artificial Intelligence, 34(5), 412\u0026ndash;29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/08839514.2020.1762963\u003c/span\u003e\u003cspan address=\"10.1080/08839514.2020.1762963\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eToplak M, Curk T, Demšar J, Zupan B. (2017). Orange: From experimental machine learning to interactive data analysis. KDD Tutorials. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/3097983.3097996\u003c/span\u003e\u003cspan address=\"10.1145/3097983.3097996\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTuia D, Volpi M, Copa L, Kanevski M, Munoz-Mari J. (2016). A survey of active learning algorithms for supervised remote sensing image classification. IEEE Journal of Selected Topics in Signal Processing, 10(2), 606\u0026ndash;17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/JSTSP.2015.2509170\u003c/span\u003e\u003cspan address=\"10.1109/JSTSP.2015.2509170\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVillon S, Chaumont M, Subsol G, Claverie T, Vill\u0026eacute;ger S. (2018). Coral reef fish detection and recognition in underwater videos by supervised machine learning: Comparison between Deep Learning and HOG\u0026thinsp;+\u0026thinsp;SVM methods. Ecological Informatics, 48, 121\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecoinf.2018.10.005\u003c/span\u003e\u003cspan address=\"10.1016/j.ecoinf.2018.10.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"discover-artificial-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"diai","sideBox":"Learn more about [Discover Artificial Intelligence](https://www.springer.com/44163)","snPcode":"","submissionUrl":"","title":"Discover Artificial Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Machine learning, image classification, sea animal dataset, Orange Data Mining, neural networks","lastPublishedDoi":"10.21203/rs.3.rs-7387846/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7387846/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study presents a comparative analysis of machine learning algorithms for classifying sea animal images using Orange Data Mining. The dataset, consisting of labeled images categorized into three classes\u0026mdash;squid, statfish, and whale\u0026mdash;was sourced from an open-access repository and processed using supervised learning workflows. Various models, including Neural Networks, Support Vector Machines (SVM), Random Forests, k-Nearest Neighbors (kNN), Logistic Regression, Na\u0026iuml;ve Bayes, Gradient Boosting, and AdaBoost, were evaluated using 10-fold cross-validation. Performance was assessed across multiple metrics: Area Under the Curve (AUC), Classification Accuracy (CA), F1-score, Precision, Recall, and LogLoss. The Neural Network model yielded the best overall performance with an AUC of 0.990 and a classification accuracy of 93.2%. SVM and Logistic Regression closely followed, outperforming other traditional and ensemble methods. Confusion matrix analysis further supported these findings, demonstrating low misclassification rates for Neural Networks. ROC curve evaluations for individual classes confirmed the robustness of top-performing models. The findings validate the effectiveness of low-code platforms like Orange in streamlining image classification pipelines for ecological and biological image datasets. This study provides valuable insights for researchers aiming to deploy interpretable and scalable machine learning solutions in marine biology and related domains.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e","manuscriptTitle":"Sea Animal Image Classification Using Machine Learning Algorithms for Accurate and Scalable Prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-30 07:18:03","doi":"10.21203/rs.3.rs-7387846/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-24T13:11:32+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-19T19:03:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"233810970487964486176813843818541723964","date":"2025-11-12T11:28:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"107855148965136096467272843859405461653","date":"2025-10-20T14:41:10+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-28T04:25:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"37515371776744091270638290934375128869","date":"2025-09-26T05:32:29+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-18T09:15:15+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-30T13:38:22+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-30T13:36:34+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Artificial Intelligence","date":"2025-08-16T13:38:00+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-artificial-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"diai","sideBox":"Learn more about [Discover Artificial Intelligence](https://www.springer.com/44163)","snPcode":"","submissionUrl":"","title":"Discover Artificial Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5a4a0ec1-6ed4-4cb7-827b-37882425a3e1","owner":[],"postedDate":"September 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-16T16:02:35+00:00","versionOfRecord":{"articleIdentity":"rs-7387846","link":"https://doi.org/10.1007/s44163-026-01005-9","journal":{"identity":"discover-artificial-intelligence","isVorOnly":false,"title":"Discover Artificial Intelligence"},"publishedOn":"2026-03-10 15:59:09","publishedOnDateReadable":"March 10th, 2026"},"versionCreatedAt":"2025-09-30 07:18:03","video":"","vorDoi":"10.1007/s44163-026-01005-9","vorDoiUrl":"https://doi.org/10.1007/s44163-026-01005-9","workflowStages":[]},"version":"v1","identity":"rs-7387846","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7387846","identity":"rs-7387846","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}