Performance Analysis of Different Models for Twitter Sentiment

Abstract

Sentiment analysis is the classification of emotions (such as positive negative and neutral) in the textual data. It helps companies to know the sentiments of customers regarding their products and services that they provide. Today many customers express their reviews on particular product on social media sites. If we analyze those reviews using sentiment analysis, we could know whether customers are happy or not with certain products. Twitter represents the largest and most dynamic datasets for data mining and sentiment analysis. Therefore, Twitter Sentiment Analysis plays an important role in the research area with significant applications in industry and academics. The purpose of this paper is to provide an optimal algorithm for Twitter sentiment analysis by comparing the accuracy of various machine learning models. In this context, nine well-known learning-based classifiers have been evaluated based on confusion matrices.