Classification of Sentiment Analysis on Tweets using Machine Learning Techniques

Abstract

Growth in social media has huge of amount of data which includes reviews about products ,blogs which discuss on the peoples opinion .We can learn sentiment analysis in web mining, data mining ,it is an application of Natural Language Processing. Due to growth in social media all the fortune companies are working on Opinion mining. The basic goal of Sentiment analysis is to ensure the sentence either as positive emotion or negative emotion. Sentiment analysis extracts the sentiments in the form various discussions, forums, blogs. Importance of social media leads in growth of sentiment analysis. For an organization, it wants to know about the people’s opinions on products which it had been released and it conducts surveys of products and opinion polls. Consumers also used to make research on products and price of product by using sentiment analysis. Marketers used to make research about company and products by effective utilization of sentiment analysis. This thesis contributes to classification of tweets in to either positive or negative using Machine learning techniques such as Nave Bayes classifier, Multinomial Nave Bayes algorithm, SVM Classifier, and Decision Tree. Comparative tabulation of performance of above mentioned classifiers is created to critically analyze the sentiment of tweets.