Sentiment Analysis using SVM, Twin SVM, and Naïve Bayes

There is a variety of information available today on websites like Twitter, blogs, and e-commerce platforms, including evaluations and conversations about events. It takes time to manually analyze these opinions. Sentiment analysis and opinion mining use machine learning techniques to automatically analyze text and determine people’s attitudes towards a topic or product to address this. In one’s free time, watching films can be an engaging experience. There is a huge selection of films accessible to stream online or see in theatres in the modern digital age. Reading movie comments on IMDB, which are many and cover a variety of topics, including star ratings, has grown in popularity. In this work, the IMDB movie review dataset is subjected to sentiment analysis. The goal is to develop a machine learning model that can ascertain if a given scenario is generally thought to be good or negative. Since machines can only understand numbers and not text, the incoming data must first be preprocessed and converted into a numerical vector space. SVM, Twin SVM, and Nave Bayes are three machine learning algorithms that are used, and the outcomes are assessed and contrasted. In this analysis, Twin SVM stands out for performing better.