A Machine Learning Framework For Movies And Political Reviews Using Sentiment Analysis

Abstract

Various sentiments and opinions made across various social media sites provide us the ample opportunity to discuss and predict the result of many Movies in the recent past even before their release worldwide. Today, it is the big and prime tool for influencing prospective viewers. Sentiment analysis also regarded as opinion mining is presently the major research area that deals with the enormous data accumulated by generating and gathering by people and their day to day online activities. It is an advanced form of Natural Language Processing to predict the nature of various opinions of people and their mood based on their old posts, tweets, or other views shared across different social media. Sentiment analysis recognizes and describes various kinds of opinions in many aspects when it is aided with machine learning and another processing tool. Acquiring various Movies review and political review, its behavior, their analysis, and the prediction of different outcomes of Movies success and election results in the form of some opinion polls is one such area of opinion mining and sentiment analysis that we will discuss. Here, our prime concern will be on remarkable results observed in predicting the movies success and election result prediction across various parts of the india based on sentiment analysis by different researchers. newlineIn this work, we have shown the sentiment analysis on the reviews taken from IMDB on the behalf of movies released. People have given their reviews towards the Movies on IMDB. Our proposed sentiment analysis is based on Tf-IDF, NaiveBayes algorithm which shows improved classification than other traditional Natural Language Processing tools. In the next part of this work we have shown the sentiment analysis on the reviews taken from Twitter on the behalf of political parties like BJP, Congress and AAP. People have given their reviews towards the political party in India. Our proposed sentiment analysis is based on Valence Aware Dictionary and Sentiment Reasoner (VADER) which shows improved cla