Challenges in Online Social Networks (OSNs) Deal with Machine Learning Algorithms

Abstract Climate change has become a rapid debate among the people because of the drastic challenges faced by the entire world. The Online Social Networking (OSN) site bestowed the medium of discussion where people share their opinions and concerns. In this research, the primary dataset is extracted by using the keyword #climatechange from the renowned OSN site X formerly known as Twitter for sentiment analysis. The objective of the study is to explore the topic of discussion conferred in the considered climate change dataset. This task is achieved with the help of a manually designed program in which three factors of climate change are inspected and analyzed. The TextBlob tool is employed for the annotation of the deemed factors. The results are interpreted with the help of three supervised machine learning classifiers namely Logistic Regression, Naïve Bayes, and Support Vector Machine. These three techniques are implemented to compare and contrast the results based on four parameters precision, recall, f1-score, and accuracy of the model. The Naïve Bayes classifier shows a significant performance among all the other classifiers.