SENTIMENT ANALYSIS FOR MOVIE RECOMMENDATION SYSTEM

Abstract

This paper discusses the application of sentiment analysis to recommend movies. The means of identifying and interpreting the sentiments of online users across numerous and diverse sources utilise sentiment analysis, which can also be described as the process of extracting the emotional tone of online text. The research aims to conduct sentiment analysis of film reviews, identify existing research questions and gaps, and implement the optimal possible strategy for completing the given work. By analyzing textual data, businesses can predict the developing market trends, be more effective with their corporate strategies and decision making since they will be able to determine whether the emotional tone is neutral, positive or negative. Specifically, this research looks at the precision, and effectiveness of Naive Bayes classification models, including Bernoulli Naive Bayes, Multinomial Naive Bayes and Complement Naive Bayes in both predicting positive and negative movie reviews.