Sentiment Analysis And Topic Modelling Of Candidate News In The 2024 General Election On Twitter Social Media Using Latent Dirichlet Allocation (LDA) Method

The use of Twitter as a platform to express public opinion regarding fuel subsidies in Indonesia. Through sentiment analysis using Support Vector Machine method and word-based lexicon, this study reveals that the majority of people are in favour of fuel price increase or subsidy policy change. The sentiment data obtained from this research, which includes positive, neutral and negative sentiments, provides a clear picture of the public's views on the issue. SVM classification method and validation with K-Fold Cross Validation were used to ensure the accuracy of sentiment analysis results obtained from Twitter data. This research is also expected to help society to gauge public opinion on news and candidates in elections. It helps understand how people respond to certain political issues and candidates and the results of sentiment analysis and topic modelling can provide a better understanding of the key issues that matter to voters. This can help candidates and political parties to craft more effective campaign messages and can also be used to detect hoaxes or false information that may spread on social media during elections. This is important for maintaining the integrity of the election.