Public Sentiment Analysis to Support Indonesian Government Induction Stove Program

Abstract

This research focuses on a comprehensive examination of public sentiment surrounding PT PLN's pilot project to convert LPG stoves to induction stoves. By conducting sentiment analysis, the study aims to understand public perspectives and opinions, identify improvement areas, and enhance the quality of future projects. The research framework includes data collection, preprocessing, and analysis, using four algorithms for sentiment classification: Naïve Bayes classifier, logistic regression, support vector machine, and K-nearest neighbor. The accuracy of these algorithms varied, with logistic regression achieving the highest accuracy at 70%. This study's preliminary results indicate public sentiment toward the PLN induction stove project, with 50% positive, 26% negative, and 24% neutral sentiments. Word cloud visualization was utilized to highlight significant words based on frequency. The research emphasizes leveraging sentiment analysis to drive positive changes and align projects with community expectations. Further research can explore factors influencing sentiment, strategies to address concerns, and the long-term impact of incorporating public sentiment in decision-making processes.