Syarif Hidayatullah, Ulfa Siti Nuraini
In the era of fast-growing social media, trend analyses of political topics have become crucial. This article analyzes online conversations related to the 2024 President and Vice President of Indonesia, Prabowo Subianto and Gibran Rakabuming Raka, through Twitter. The clustering method is used to group tweets based on their content. The analysis stages included data collection, text preprocessing, word weighting, dimension reduction, and clustering. Various clustering algorithms were evaluated to identify the best cluster result based on the highest silhouette score. As a result, the Hierarchical Clustering algorithm with the complete method obtained the highest silhouette score of 0.756 with PCA of 2 factors produced 3 clusters. Furthermore, a word cloud was constructed to draw conclusions for each cluster. The results of the analysis show that the first cluster discussed the previous leadership period, the second cluster discussed the Village Development Index (IDM), and the third cluster focused on conversations about political parties. This study is expected to provide insights into the dynamics of public conversations and political messages expressed on social media. © 2025 Author(s).
Data Science Study Program, Universitas Negeri Surabaya, Surabaya, Indonesia
