<![CDATA[One nation with a democratic political system is Indonesia. The public is able to express themselves freely. The public's use of social media is expanding quickly, particularly among users of platform ‘X’. The now trending tweets concern the 2024 presidential election. The reaction to the results of the 2024 presidential election has ranged from positive to negative to neutral. Large numbers of tweets can be used as a source of information to do their sentiments analysis. It is possible to know if people, in general, are satisfied or unsatisfied with the outcome of the presidential election thanks to the emotion categorization. This study aims to analyze public sentiment regarding the election result utilizing machine learning methods which will provide insights into public opinion that can be useful in political strategy as well as in public discourse assessment. In this paper, we will compare the Naïve Bayes Classifier (NBC) and the Support Vector Machine (SVM) algorithms for tweet classification of platform ‘X’ sentiment. This study presents the performed data analysis on 2193 data points (from platform X) that have been classified into neutral, positive, and negative categories using the Naive Bayes Classifier (NBC) and Support Vector Machine (SVM) techniques. Balancing SMOTE is used to address data imbalance, and TF-IDF is applied for feature extraction. Results depicts that Naïve Bayes Classifier (NBC) gives an accuracy of 62.41% whereas Support Vector Machine (SVM) gives 62.19% accuracy. This accuracy on these creations demonstrates how able models can be when classifying varying public sentiments between political events, highlighting the abilities, but also weaknesses of such efforts in sentiment classification. This paper contributes to the further development of sentiment analysis by providing an assessment of how effective these algorithms are, and by stressing the need for unbalance data treatment on research utilizing social media.]]>
