<![CDATA[Objective: This study aims to analyze and classify public sentiment toward prominent individuals on social media, focusing on distinguishing between acceptable opinion and hate speech. Method: A dataset comprising 700 tweets about public figures was collected and processed using the Support Vector Machine (SVM) classification algorithm, with Term Frequency–Inverse Document Frequency (TF-IDF) employed for feature weighting. The approach was chosen to evaluate the effectiveness of SVM in handling textual sentiment data. Results: The implementation of the SVM model demonstrated reliable performance in categorizing tweets according to sentiment polarity and identifying instances of hate speech. The TF-IDF weighting method enhanced the accuracy of feature representation within the model. Novelty: This study contributes to computational linguistics and social media analytics by providing an empirical evaluation of SVM’s capability in sentiment classification related to public figures, offering potential applications for automated monitoring of online discourse and hate speech detection.]]>
Copyrights © 2025
