Political Candidate Popularity Computation Using Sentiment Analysis

Abstract

Candidate popularity is of prime importance especially during or before elections. Traditionally
journalists and employees of polling organizations visit the voters to get information about the candidates popularity.
Voters when asked about a particular candidate give elaborated answers which contain positive as well as negative
sentiments. Due to mixed nature of answers calculating popularity score becomes difficult in traditional approach.
Also the calculated popularity score has a very low accuracy .Time,workforce and finance required for this traditional
approach is very high whereas results are not satisfactory.
This project makes popularity computation simple and accurate. Twitter tweets are used to compute the
popularity of a leaderor party.Tweets are extracted from twitter app API which will be processed using sentiment
analysis and after data from large number of tweets are obtained a popularity score will be computed and displayed.
This will allow journalists and polling organization to take input from bigger voter set and cost for carrying
out these tasks will be reduced to a great extent. Users or voters will get the information about popularity in a concise
and simple form as it will be displayed in the form of bar graphs and pie charts. Thus easing the task for voters as well
as journalists.