An Analysis of Different Machine Learning Algorithms for Personality Type Predictions Based on Social Media Posts

Ruiwen Yu

2023

Abstract

In the digital era, people have increasingly used social media as a platform to communicate with each other and express feelings. Researchers have vast data to predict users’ personalities, who play various practical roles in real life. This paper focuses on MBTI personality trait prediction based on users’ social media posts. Comparisons and Analyses are made between the three most common methods of personality prediction, including Naive Bayes (NB), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost), to find out the most effective one. Results show no evident correlation between the accuracy of the models and the training method; each model’s capability to predict varies. Models trained on XGBoost have the highest accuracy on average, although all three outcomes are incredibly close. One model trained on XGBoost with well-tuned hyperparameters obtained the best performance. Further analyses show other factors. For example, data distribution also influences the model’s performance in some ways.

Paper Citation

in Harvard Style

Yu R. (2023). An Analysis of Different Machine Learning Algorithms for Personality Type Predictions Based on Social Media Posts. In Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-705-4, SciTePress, pages 386-391. DOI: 10.5220/0012816400003885

in Bibtex Style

@conference{daml23,
author={Ruiwen Yu},
title={An Analysis of Different Machine Learning Algorithms for Personality Type Predictions Based on Social Media Posts},
booktitle={Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2023},
pages={386-391},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012816400003885},
isbn={978-989-758-705-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML
TI - An Analysis of Different Machine Learning Algorithms for Personality Type Predictions Based on Social Media Posts
SN - 978-989-758-705-4
AU - Yu R.
PY - 2023
SP - 386
EP - 391
DO - 10.5220/0012816400003885
PB - SciTePress