Sentiment Analysis on Movie Review using Naïve Bayes

Sentiment analysis is needed because the analysis of the emotion is somehow complicated due to slang phrases, misspellings, short-forms, recurring characters, the use of dialects, and modern emoticons. The aim of this project is to develop a system to analyze the sentiment of movie reviews and visualize the result of sentiment analysis. The dataset was captured from IMDb movie reviews. This dataset contains 50,000 instances with two columns which are the reviews and their sentiments. The data set undergoes a cleaning process to extract meaningful data. Two types of features have been investigated, namely, the Bag of Words (BoW) and the TF-IDF modelling with Naïve Bayes classifiers. The result of accuracy is 89% which indicates that the Naïve Bayes can be used for sentiment analysis in any movie review task.