Public Sentiment Analysis from Twitter (X) About Dissolving the House With Naive Bayes & SVM
Article Sidebar
Main Article Content
Abstract
Twitter (X) has become a major platform for the expression of public opinion, including political issues such as hashtags #BubarkanDPR reflecting public dissatisfaction with the legislature. This study aims to (1) analyze public sentiment on the issue of "Dissolve the DPR" and (2) compare the performance of the Naive Bayes (NB) and Support Vector Machine (SVM) classification methods in the context of Indonesian texts. This study used 1989 tweet data obtained through web scraping. Data through the preprocessing stage (cleaning, normalization, stopword removal, stemming) and automatic labeling (positive, negative, neutral) using the RoBERTa pre-trained model . Features were extracted using TF-IDF, then the data was divided into 80% of the training data and 20% of the test data to train the NB and SVM models. The results of the evaluation showed that the Support Vector Machine (SVM) provided superior performance with an accuracy of 79%, compared to Naive Bayes (NB) with an accuracy of 74%. SVM also showed a more balanced F1-Score for negative (82%) and neutral (80%) classes. Both models showed significant difficulty in classifying positive sentiment (F1-Score SVM 12%; NB 0%), which was identified as a result of severe data imbalance in the dataset. In conclusion, SVM is more effective than Naive Bayes for this classification task, and the analysis shows that public sentiment towards #BubarkanDPR issue is dominated by negative and neutral polarity.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
[1] P. R. Sari, D. R. Indah, E. Rasywir, M. A. Firdaus, and G. Athalina, “Comparison of Naive Bayes and SVM Algorithms for Sentiment Analysis of PUBG Mobile on Google Play Store,” System, vol. 13, no. 6, p. 2767, 2024, doi: 10.32520/stmsi.v13i6.4814.
[2] A. Kumar, “Systematic literature review of sentiment analysis on Twitter using soft computing techniques,” no. November 2018, 2019, doi: 10.1002/cpe.5107.
[3] I. Syapriani, D. Putriana, R. P. Aula, G. S. Sembiring, V. Aulia, and S. Khoiria, "Journal of Nusantara Research on Social Media User Communication Strategy in Managing Language, Ethics, and Social Resistance in the Comment Room of Digital Platform X (Twitter)," vol. 1, pp. 360–368, 2025.
[4] Z. Drus, H. Khalid, Z. Drus, and H. Khalid, “ScienceDirect ScienceDirect Sentiment Analysis in Social Media and Its Application : Systematic Sentiment Analysis in Social Media and Its Application : Systematic Literature Review Literature Review,” Computation. Sci., vol. 161, pp. 707–714, 2019, doi: 10.1016/j.procs.2019.11.174.
[5] A. P. Widyassari, D. Salsabilla, and M. A. Amrozi, "Analysis of public sentiment on twitter towards president Prabowo's inauguration using the Naïve Bayes algorithm," vol. 10, no. 1, pp. 13–24, 2025.
[6] C. Villavicencio, J. J. Macrohon, X. A. Inbaraj, J. H. Jeng, and J. G. Hsieh, “Twitter sentiment analysis towards covid-19 vaccines in the Philippines using naïve bayes,” Inf., vol. 12, no. 5, 2021, doi: 10.3390/info12050204.
[7] P. R. Sari et al., “Comparison of Naive Bayes and SVM Algorithms for Sentiment Analysis of PUBG Mobile on Google Play Store,” vol. 13, pp. 2767–2779, 2024.
[8] W. Ningsih, B. Alfianda, and D. Wulandari, "Comparison of Naive Bayes and SVM Algorithms in Twitter Sentiment Analysis on Electric Car Use in Indonesia," vol. 4, no. April, pp. 556–562, 2024.
[9] M. Wahyudi, M. S. Novelan, U. Development, and P. Budi, "Comparison Of K-Nn And Naive Bayes Performance In Information Technology Governance To Analyze The Level Of Satisfaction Of Shopee Application Users," vol. 4307, no. August, pp. 3424–3431, 2025.
[10] H. Juwiantho et al., “Sentiment Analysis Twitter Bahasa Indonesia Berbasis Word2vec Menggunakan Deep Convolutional Neural Network Sentiment Analysis Twitter Indonesian Language Word2vec-Based Using Deep Convolutional Neural Network,” vol. 7, no. 1, pp. 181–188, 2020, doi: 10.25126/jtiik.202071758.
[11] A. D. Adhi Putra, "Sentiment Analysis on User Reviews of Bibit and Bareksa Applications with KNN Algorithm," JATISI (Tek Journal. Inform. and Sist. Information), vol. 8, no. 2, pp. 636–646, 2021, doi: 10.35957/jatisi.v8i2.962.
[12] M. I. Fikri, T. S. Sabrila, and Y. Azhar, “Comparison of Naïve Bayes and Support Vector Machine Methods in Twitter Sentiment Analysis,” Smatika J., vol. 10, no. 02, pp. 71–76, 2020.
[13] H. Susana and N. Suarna, "application of the naive bayes method classification model to the use of internet access stmik ikmi cirebon jl perjuangan no 10b kesambi cirebon city 3) software engineering study program stmik ikmi Cirebon Jl Perjuangan No 10B Kesambi Cirebon City 4) Computerized Accounting Study Program STMIK IKMI Cirebon Jl Perjuangan No 10B Kesambi Cirebon City, ” J. Sist. Inf. and Technology. Information), vol. 4, no. 1, pp. 1–8, 2022.
[14] gigih ,putra Kawani, “Journal of Informatics, Information System, Software Engineering and Applications,” J. Informatics, Inf. Syst. Softw. Eng. Appl., vol. 1, no. 2, pp. 73–081, 2019.
[15] Wikipedia, "Corruption Eradication Commission of the Republic of Indonesia," Wikipedia, vol. 7, no. 1, p. 1, 2024.
[16] D. Alita, Y. Fernando, and H. Sulistiani, "Implementation of the Svm Multiclass Algorithm on Indonesian Public Opinion on Twitter," J. Compact Techno, vol. 14, no. 2, p. 86, 2020, doi: 10.33365/jtk.v14i2.792.
[17] A. S. Ritonga and E. S. Purwaningsih, "The Application of the Support Vector Machine (SVM) Method in the Quality Classification of Smaw (Shield Metal Arc Welding)," Ilm. Edutic, vol. 5, no. 1, pp. 17–25, 2018.