Sentiment Analysis (KEY IJP************568)

• Snehal Shah,Akshata Bhat,Sumitra Singh,Arya Chavan,Aryan Singh

Abstract

The process of locating and categorizing viewpointsor feelings represented in the source text is known as senti-ment analysis. Social media information like blog posts, statusupdates, and tweets generate a tonne of sentiment-rich data.Understanding the opinions of the masses can be enormouslyaided by sentiment analysis of this user-generated data. Twittersentiment analysis is more challenging than generic sentimentanalysis because of the prevalence of misspellings and slangterms. Twitter allows a character count of up to 140 characters.The knowledge base approach and Machine learning approachare the two strategies used for analyzing sentiments from the text.In this study, we try to analyze Twitter posts on electronic devices,like computers and smartphones, using the machine learningtechnique. By doing sentiment analysis in a specific domain, it ispossible to identify the effect of domain information in sentimentclassification. We introduce a new feature vector for extractingpeoples thoughts about products and classifying the tweets aspositive or negative.