A Review of Sentiment Analysis Techniques

Abstract

The innovation of Web 2.0 is increasing day by day and its latest hype is Microblogging. Short messages exchange platforms connect people worldwide in an unprecedented manner by publishing short text updates regarding various topics. People freely express their views and opinions on microblogging sites that give rise to multiple user sentiments. In order to precisely determine the mood and nature of user it is essential to analyze the sentiments based on contextual information. Further, sentiment analysis is required for different markets to consider their consumer views and critics to improve their products. We have analyzed different sentiment analysis techniques and algorithms in order to highlight their pros and cons. Along with their comparison on basis of Precision, Recall, F-measure, and Accuracy. This study helps to identify the major difficulties in the area of sentiment analysis that are still not resolved and being considered by the researchers.