ANALISA SENTIMEN UU PERAMPASAN ASET DENGAN ALGORITMA MULTINOMINAL NAIVE BAYES

This study aims to analyze public sentiment towards the Asset Forfeiture Law using the Multinomial Naïve Bayes (MNB) algorithm. This Law has become a trending topic, generating various reactions on social media, particularly on the Twitter/X platform. Sentiment analysis is necessary to determine the public's opinion tendency, whether positive, negative, or neutral, regarding the policy. The research process begins with data collection through a scraping technique on Twitter, followed by data preprocessing, sentiment labeling, and the training and testing of the classification model. Previous studies indicate that the MNB algorithm is capable of achieving high levels of accuracy in classifying various text topics. The results of this study show that the MNB algorithm successfully classified public opinion regarding policy issues with an accuracy of 88%. This research is expected to provide a tangible contribution to systematically understanding the public's perception of the Asset Forfeiture Law, based on data. Furthermore, the analysis results can be used as input for the evaluation process of future public policies.