An improved graph factorization machine based on solving unbalanced game perception
2024

Improved Graph Factorization Machine for Mobile Game User Experience

Sample size: 789 publication Evidence: high

Author Information

Author(s): Xie Xiaoxia, Jia Yuan, Ma Tiande

Hypothesis

Can a new method called Balanced Graph Factorization Machine (BGFM) improve the prediction of mobile game user experience by addressing data imbalance and feature interactions?

Conclusion

The proposed BGFM model significantly outperforms existing algorithms in predicting mobile game user experience.

Supporting Evidence

  • The BGFM model outperformed eight state-of-the-art models in AUC, precision, recall, and F-measure.
  • Data balancing using Borderline-SMOTE improved the model's performance.
  • The BGFM effectively combines the strengths of factorization machines and graph neural networks.

Takeaway

This study created a new way to understand how players feel about mobile games by using a special model that looks at many different factors and balances the data.

Methodology

The study used Borderline-SMOTE for data balancing and developed the BGFM model to analyze feature interactions in a graph structure.

Potential Biases

Potential bias due to the reliance on a specific dataset from a single game.

Limitations

The study may not account for all possible user and environmental factors affecting game perception.

Participant Demographics

Data collected from 789 games, including user data such as player ID, age, gender, and skill level.

Digital Object Identifier (DOI)

10.3389/fnbot.2024.1481297

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication