<![CDATA[The 2024 election has sparked an explosion of public opinion across various digital platforms, but the complexity and large volume of data make it difficult for policymakers to understand public sentiment in a timely manner. Therefore, an accurate and efficient sentiment analysis method is needed to automatically classify public opinion. This study aims to analyze and compare the performance of the Naïve Bayes algorithm and an optimized Support Vector Machine (SVM) in classifying post-election public sentiment. The research method includes collecting 10,000 text data entries from various data sources, conducting text preprocessing, extracting features using the TF-IDF method, applying both algorithms with parameter tuning, and generating their performance using accuracy, precision, recall, and F1 score metrics. The results show that the optimized SVM algorithm delivers superior performance, achieving 88.24% accuracy, compared to 82.35% for Naïve Bayes. These findings indicate that SVM is more effective in handling complex public opinion sentiment classification, thus serving as a valuable reference for post-election policymaking]]>
Copyrights © 2025
