Prediction of the 2023 Turkish Presidential Election Results Using Social Media Data

TL;DR

This study predicts the 2023 Turkish presidential election results by combining social media interaction data with traditional polls, demonstrating that ARIMAX models provide the most accurate forecasts across different time frames.

Contribution

It introduces a volume-based social media analysis approach combined with traditional polling data for election prediction, highlighting the effectiveness of ARIMAX models.

Findings

ARIMAX outperforms other models in prediction accuracy

Social media interaction volume correlates with election outcomes

Model performance consistent across various time windows

Abstract

Social media platforms influence the way political campaigns are run and therefore they have become an increasingly important tool for politicians to directly interact with citizens. Previous elections in various countries have shown that social media data may significantly impact election results. In this study, we aim to predict the vote shares of parties participating in the 2023 elections in Turkey by combining social media data from various platforms together with traditional polling data. Our approach is a volume-based approach that considers the number of social media interactions rather than content. We compare several prediction models across varying time windows. Our results show that for all time windows, the ARIMAX model outperforms the other algorithms.