Perbandingan Particle Swarm Optimization Dan Naive Bayes Terhadap Sentimen Instagram Pada Pemilihan Presiden

Abstract

This research aims to compare the performance of two analysis techniques, namely Particle Swarm Optimization (PSO) and Naive Bayes, in evaluating public sentiment on Instagram after the 2024 presidential election. PSO is used to optimize important features in sentiment analysis, while Naive Bayes is used as a comparison for assess PSO performance. Data was obtained from user uploads, comments and interactions on Instagram regarding the presidential election. After data collection and pre-processing, significant features are extracted and both methods are applied for sentiment analysis. Evaluation was carried out to compare the Akurasi and performance of the two methods. The research results show that PSO has the potential to improve the Akurasi of sentiment analysis compared to Naive Bayes. This research provides in-depth insight into the advantages and disadvantages of each method in the context of sentiment analysis on the Instagram platform after the presidential election, as well as an important contribution to data-based political decision making.