Integrating Sentiment Analysis in Predictive Models: A Comparative Study on Game Popularity on Steam
Graduation Year
2025
Date of Submission
11-2024
Document Type
Open Access Senior Thesis
Degree Name
Bachelor of Arts
Department
Mathematics
Reader 1
Mark Huber
Terms of Use & License Information
Rights Information
© 2024 Khaleefa AlHemeiri
Abstract
Over the past decades, the gaming industry has managed to evolve into a multi-billion-dollar enterprise. Gaming platforms such as Steam foster unprecedented amounts of engagement among players worldwide daily. In this thesis, we investigate the effect of incorporating sentiment-driven metrics, specifically YouTube view counts and positive reviews, into predictive models for game popularity. In addition, by comparing our linear regression sentiment-based approach to the Bayesian hierarchical folded normal model used by De Luisa et al. (2021), we can understand the many differences, strengths, and limitations of each methodology. In our thesis, we focus on three games. Each is of varying popularity and is also the same three games used by De Luisa et al.: Dark Souls 3, Subnautica, and Redirection. For more popular games, sentiment-driven variables demonstrated moderate correlations with player counts, suggesting their limited, but existing utility in gauging game popularity. However, our linear regression model performed significantly more poorly for lesser-known games, as data sparsity undermined its predictive reliability. On the other hand, the Bayesian model developed by De Luisa et al. continuously produced reliable predictions for games of all popularity levels, albeit with diminishing accuracy for exceptionally popular games. These findings underline the importance of data availability and methodological adaptability in predictive modeling for games. While sentiment analysis does offer valuable insights for high-engagement games, integrating it with more flexible and adaptable techniques, such as Bayesian modeling, may provide a more balanced framework for predicting player engagement across diverse gaming contexts.
Recommended Citation
AlHemeiri, Khaleefa, "Integrating Sentiment Analysis in Predictive Models: A Comparative Study on Game Popularity on Steam" (2025). CMC Senior Theses. 3723.
https://scholarship.claremont.edu/cmc_theses/3723
Included in
Applied Statistics Commons, Numerical Analysis and Computation Commons, Other Mathematics Commons, Statistical Models Commons