Sentiment Analysis and Communication Networks Towards Ridwan Kamil in #PilkadaJakarta2024 on Social Media X (Twitter)

Abstract

The development of social media has changed the landscape of political communication, especially in shaping public opinion towards political actors ahead of regional head elections. This study aims to analyze public sentiment and the structure of communication networks towards Ridwan Kamil in the context of the 2024 Jakarta Pilkada on social media X (Twitter). The methods used include sentiment classification with the Naïve Bayes Classifier (NBC) algorithm and network analysis using the Social Network Analysis (SNA) approach through the Gephi software. A total of 2,000 tweets were analyzed after undergoing pre-processing stages, including cleaning, case folding, tokenizing, stopword removal, and stemming using TF-IDF. The classification results show that 45.6% of tweets are positive, 31.8% are neutral, and 22.6% are negative, with a model accuracy of 62.90% and a margin of error of ±2.56%. In terms of communication structure, the network was found to be fragmented and influenced by informal nodes such as @BungkusTukang, @NalarPolitik_, and @irenejuliency, while the @ridwankamil account played a strategic role as a liaison between clusters. The communication patterns formed followed the configuration of a personal star ego network and a bridge-type one. These findings indicate that digital political communication is dominated not only by official actors but also by non-institutional actors who significantly influence public discourse