Electronic Theses and Dissertation

Media sosial x sebagai platform utama untuk membentuk opini publik ketika pemilihan presiden 2024 di indonesia. bertopic adalah teknik pemodelan topik yang menggabungkan representasi berbasis transformer dengan teknik clustering untuk menghasilkan topik yang lebih interpretatif dan relevan. penelitian ini bertujuan menganalisis perubahan opini publik dengan mengeksplorasi fluktuasi sentimen terhadap tiga pasangan calon presiden selama kampanye, serta mengidentifikasi topik utama dan hashtag populer pada periode anomali menggunakan pendekatan bertopic. dataset penelitian terdiri dari 9.817.355 baris yang mencakup interaksi di media sosial x. analisis sentimen dianalisis menggunakan model indonesia-roberta, sementara topic modelling dilakukan menggunakan bertopic. social network analysis (sna) diterapkan untuk memetakan pola hubungan antar-hashtag. hasil menunjukkan 44,47% tweet bersentimen netral, 29,23% negatif, dan 26,10% positif. lonjakan sentimen negatif signifikan terjadi pada 7-9 januari 2024, bertepatan dengan debat calon presiden ketiga. topik utama mencakup yaitu dukungan generasi muda terhadap pasangan calon 03, persepsi masyarakat terhadap janji politik para calon, lalu kritik terhadap pertahanan indonesia, serta dukungan dan ujaran kebencian terhadap pasangan calon 01. visualisasi hashtag menunjukkan dominasi #debatcapres, dengan degree centrality tertinggi (1909), closeness centrality (0,57), dan eigenvector centrality sempurna (1,0). penelitian ini memberikan informasi tentang dinamika opini publik selama kampanye politik dan pentingnya strategi komunikasi berbasis data untuk memengaruhi persepsi publik secara efektif. hasil ini berguna bagi tim kampanye untuk merancang strategi komunikasi yang lebih baik, media untuk menyajikan analisis mendalam, dan masyarakat dalam memahami isu-isu utama selama pemilu, yang dapat memengaruhi keputusan pemilih.

Social media X is the main platform to shape public opinion during the 2024 Presidential Election in Indonesia. BERTopic is a topic modeling technique that combines transformer-based representation with clustering techniques to generate more interpretative and relevant topics. This research aims to analyze changes in public opinion by exploring sentiment fluctuations towards three presidential candidate pairs during the campaign, as well as identifying key topics and popular hashtags in anomalous periods using the BERTopic approach. The research dataset consists of 9,817,355 rows covering interactions on X social media. Sentiment analysis was analyzed using the Indonesia-RoBERTa model, while topic modeling was performed using BERTopic. Social Network Analysis (SNA) was applied to map the relationship pattern between hashtags. Results showed 44.47% of tweets had neutral sentiment, 29.23% were negative, and 26.10% were positive. A significant spike in negative sentiment occurred on January 7-9, 2024, coinciding with the third presidential debate. The main topics included the young generation's support for candidate pair 03, public perception of the candidates' political promises, then criticism of Indonesia's defense, and support and hate speech against candidate pair 01. Hashtag visualization shows the dominance of #DebatCapres, with the highest degree centrality (1909), closeness centrality (0.57), and perfect eigenvector centrality (1.0). This research provides information on the dynamics of public opinion during political campaigns and the importance of data-driven communication strategies to effectively influence public perception. These results are useful for campaign teams to design better communication strategies, the media to present in-depth analysis, and the public in understanding key issues during elections, which can influence voter decisions.