Sentiment Analysis of Twitter Use on Policy Institution Services using Naïve Bayes Classifier Method

Abstract

Twitter is one of the social media used to respond to various services of public service institutions, including the police. The research aims to determine the community's assessment of the services and performance of police institutions delivered via Twitter. This study uses the Naïve Bayes Classifier algorithm to classify topics and public sentiment towards tweets from police agencies. The results obtained were 181 positive tweets, 322 negative tweets, and 33 neutral tweets. Sentiment analysis showed 55% responded positively to police activities, 19.1% responded positively to public comments, and 91.8% responded positively to social services. It can be concluded that most people support police activities and services, but most people are still dissatisfied with police performance.