Forecasting the 2016 US Presidential Elections Using Sentiment Analysis

Abstract

The aim of this paper is to make a zealous effort towards true prediction of the 2016 US Presidential Elections. We propose a novel technique to predict the outcome of US presidential elections using sentiment analysis. For this data was collected from a famous social networking website (SNW) Twitter in form of tweets within a period starting from September 1, 2016 to October 31, 2016. To accomplish this mammoth task of prediction, we build a model in WEKA 3.8 using support vector machine which is a supervised machine learning algorithm. Our results showed that Donald Trump was likely to emerge winner of 2016 US Presidential Elections.