Analisis Sentimen Data Twitter Topik Ekonomi Dan Industri Dengan Metode Naive Bayes Dan Random Forest

Abstract

Twitter has become a valuable source of information, and sentiment analysis can provide insights into public views and attitudes towards economic and industrial issues. This research aims to develop and compare the performance of two widely used classification methods, Naive Bayes and Random Forest, for sentiment analysis on Twitter data related to the economy and industry. By addressing the existing knowledge gap in sentiment analysis using Naive Bayes and Random Forest, this study provides a clear framework that empowers companies to efficiently process and leverage Twitter data, yielding valuable decision-making insights in the realm of economy and industry. A total of 11,833 data were divided into 70% training data and 30% testing data then classified using Naive Bayes, and Random Forest algorithms. The calculation results show positive sentiment of 28,52%, negative sentiment of 31,44%, and neutral sentiment of 40,04%. The comparison of the two algorithms obtained using Naïve Bayes gets the highest accuracy of 71,89%.