Sentimental Analysis of Twitter Data using Classifier Algorithms

Abstract

Microblogging has become a daily routine for most of the people in this world. With the help of Microblogging people get opinions about several things going on, not only around the nation but also worldwide. Twitter is one such online social networking website where people can post their views regarding something. It is a huge platform having over 316 Million users registered from all over the world. It enables users to send and read short messages with over 140 characters for compatibility with SMS messaging. A good sentimental analysis of data of this huge platform can lead to achieve many new applications like – Movie reviews, Product reviews, Spam detection, Knowing consumer needs, etc. In this paper, we have devised a new algorithm with which the above needs can be achieved. Our algorithm uses three specific techniques for sentimental analysis and can be called a hybrid algorithm – (1) Hash Tag Classification for topic modeling; (2) Naïve Bayes Classifier Algorithm for polarity classification; (3) Emoticon Analysis for Neutral polar data. These techniques individually have some limitations for sentimental analysis.