Sentiment Analysis of Textual Data using various ML Techniques: A comparative study

Abstract

Sentiment analysis is a type of opinion mining which is used to determine the person’s opinion,
feelings, thoughts and judgement expressed on the text. Sentiment analysis main concern deals
with classifying what the person expresses in its text and analyzing these text helps to know whether
the person is angry, sad, happy etc. So, in this paper we have classified the text in 3 parts positive,
negative and neutral. The positive text means the person is happy, or supporting a good cause etc.,
negative text means either the person is angry, sad, upset etc. and neutral text deals with the person
giving facts or information about something. This paper deals with how we have used three models
for classification of text naïve bayes, random forest and support vector machine.