Login Search
FMDB Transactions on Sustainable Health Science Letters Submit Paper
    Crossref logo

    Archiving

    Portico

    Fernando Martins De Bulhão is a member of

    Crossref logo

A Hybrid Graph-Based Deep Learning Model with Explainable AI for Early Identification of Online Gaming Disorder

Authors:
R. Swathi Priya, S. Silvia Priscila

Addresses:
Department of Computer Applications, Bharath Institute of Higher Education and Research, Chennai, Tamil Nadu, India. Department of Computer Science, Bharath Institute of Higher Education and Research, Chennai, Tamil Nadu, India.

Abstract:

Online Gaming Disorder (OGD) is a major mental health issue comprising the inability to control gaming and the progressive increase in gaming's importance compared to other activities. The research presented here suggests a novel Hybrid Graph-Based Deep Learning Model incorporating Explainable AI (XAI) for the early identification of OGD. The study is based on a special dataset of 129 data instances gathered using digital behavioural and psychometric measurements. The model accommodates non-linear relationships between gaming frequency, social isolation, and emotional regulation by capturing user interactions as graphs of complex relationships. The model allows for non-linear relationships among gaming frequency, social isolation, and emotional regulation by representing user interactions as intricate graphs. Researchers used the Python packages PyTorch Geometric for the implementation of GNN's, and SHAP for the XAI part. The hybrid architecture integrates Graph Convolutional Networks and Long Short-Term Memory (LSTM) to study structural social networks and temporal gameplay patterns. Results show that the model is highly accurate in diagnosis and provides readable feedback on specific risk factors. Since the high-risk classification is assigned based on XAI, the clinician can grasp the rationale behind the classification, thereby narrowing the gap between complicated black-box algorithms and clinical practice. This model provides a good way to support early intervention, which may help reduce the future psychological consequences of pathological gaming.

Keywords: Online Gaming Disorder (OGD); Explainable AI; Mental Health; Behavioural Analytics; Deep Learning (DL); PyTorch Geometric; Hybrid Architecture Integration.

Received on: 06/07/2025, Revised on: 11/09/2025, Accepted on: 16/11/2025, Published on: 18/08/2026

DOI: 10.69888/FTSHSL.2026.000731

FMDB Transactions on Sustainable Health Science Letters, 2026 Vol. 4 No. 3, Pages: 202-211

  • Views : 19
  • Downloads : 3
Download PDF