Twitter Sentiment Analysis: A Novel Machine Learning Approach

Abstract

Sentiment Analysis is a technique used to classify the polarity of emotions (positive or negative) on a given text. Sentiment analysis is widely used in data mining [1] and information processing. Several studies have been done in the past to track the activities of the user on a social media platform like Twitter to gain general perception or response of a person toward a given entity such as products, services, organizations, individuals, political parties, etc. In this report, we have used two different machine learning techniques, Naïve Bayes and Support Vector Machine to do comprehension research on sentiment analysis.