And the Winner is ...: Bayesian Twitter-based Prediction on 2016 U.S. Presidential Election

TL;DR

This paper presents a Naive-Bayesian model using Twitter data to predict the 2016 U.S. Presidential Election, achieving high accuracy and identifying key candidates.

Contribution

It introduces a simplified data preprocessing approach and demonstrates competitive predictive performance with Twitter-based election forecasting.

Findings

95.8% accuracy on cross-validation

Correctly predicted key nominees Ted Cruz and Bernie Sanders

Comparable results to existing methods

Abstract

This paper describes a Naive-Bayesian predictive model for 2016 U.S. Presidential Election based on Twitter data. We use 33,708 tweets gathered since December 16, 2015 until February 29, 2016. We introduce a simpler data preprocessing method to label the data and train the model. The model achieves 95.8% accuracy on 10-fold cross validation and predicts Ted Cruz and Bernie Sanders as Republican and Democratic nominee respectively. It achieves a comparable result to those in its competitor methods.