Machine learning-based classification of gaming addiction risk levels using behavioral and mental health indicators

Abstract

In recent years, with the widespread adoption of digitalization, gaming addiction has emerged as a problem that increasingly affects individuals’ social relationships and academic lives. This study aims to predict gaming addiction risk levels using machine learning algorithms and to identify the variables influencing model performance. Within the scope of the research, Decision Trees, K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Extreme Gradient Boosting (XGBoost), and Multi-Layer Perceptron (MLP) models were comparatively analyzed. To address class imbalance in the dataset, the SMOTE method was applied, and a two-stage feature selection process was conducted to reduce model complexity and enhance discriminative power. In the first stage, feature importance scores were determined using the Random Forest (RF) algorithm, followed by the selection of the most suitable feature subset through Recursive Feature Elimination with Cross-Validation (RFECV). The findings indicate that the feature selection process positively affected model performance; however, the most stable and highest performance was achieved by the MLP model with an accuracy rate of 92.5%. It was observed that the early stopping strategy applied in the MLP architecture helped improve generalization performance by reducing the risk of overfitting. Overall, the findings indicate that machine learning methods can classify gaming addiction risk levels within the scope of the dataset used; however, external validation with different datasets and further expert-based evaluations are required before the model can be considered for clinical decision-support or early warning purposes.

Keywords

• Gaming Addiction
• Machine Learning
• Multi-Layer Perceptron (MLP)
• Feature Selection
• SMOTE
• Risk Classification

Supporting Institution

This research received no external funding.

Ethical Statement

The dataset used in this study was obtained from a publicly available and anonymous source (Kaggle: 'Gaming and Mental Health'). Since no human or animal subjects were directly involved in the research, ethical committee approval is not required.

Thanks

The authors would like to thank Kutahya Dumlupinar University, Department of Computer Engineering for the research facilities provided during this study.

Davranışsal ve ruhsal sağlık göstergeleri kullanılarak oyun bağımlılığı risk düzeyinin makine öğrenmesi tabanlı sınıflandırılması

Abstract

Son yıllarda dijitalleşmenin yaygınlaşmasıyla birlikte oyun bağımlılığı, bireylerin sosyal ilişkileri ve akademik yaşamı üzerinde giderek daha belirgin etkiler gösteren bir sorun haline gelmiştir. Bu çalışmada, makine öğrenmesi algoritmalarını kullanarak oyun bağımlılığı risk düzeylerinin tahmin edilmesi ve model performansını etkileyen değişkenlerin belirlenmesi amaçlanmıştır. Araştırma kapsamında Karar Ağaçları, K-En Yakın Komşu (KNN), Destek Vektör Makineleri (SVM), Aşırı Gradyan Artırma (XGBoost) ve Çok Katmanlı Yapay Sinir Ağları (MLP) modelleri karşılaştırmalı olarak analiz edilmiştir. Veri setindeki sınıfsal dengesizliği gidermek amacıyla SMOTE yöntemi uygulanmış, model karmaşıklığını azaltmak ve ayırt ediciliği artırmak için ise iki aşamalı bir özellik seçimi süreci gerçekleştirilmiştir. İlk aşamada Rastgele Orman (RF) algoritması ile değişkenlerin önem skorları belirlenmiş, ikinci aşamada ise Çapraz Doğrulama Destekli Yinelemeli Özellik Eleme (RFECV) yöntemi ile en uygun özellik alt kümesi seçilmiştir. Elde edilen bulgular, özellik seçimi sürecinin model performansını olumlu yönde etkilediğini göstermiştir, ancak en kararlı ve yüksek başarının %92,5 doğruluk oranı ile MLP mimarisi tarafından sergilendiğini ortaya koymuştur. MLP mimarisi özelinde devreye alınan erken durdurma (early stopping) stratejisinin aşırı öğrenme (overfitting) riskini azaltarak modelin genelleme performansına katkı sağladığı değerlendirilmektedir. Elde edilen bulgular, kullanılan veri seti kapsamında oyun bağımlılığı risk düzeylerinin makine öğrenmesi yöntemleriyle sınıflandırılabileceğini göstermektedir; ancak modelin klinik karar destek veya erken uyarı amacıyla kullanılabilmesi için farklı veri setleriyle dış doğrulama yapılmasına ve uzman değerlendirmesiyle desteklenen ek çalışmalara ihtiyaç duyulmaktadır.

Keywords

• Oyun Bağımlılığı
• Makine Öğrenmesi
• Çok Katmanlı Algılayıcı (MLP)
• Özellik Seçimi
• SMOTE
• Risk Sınıflandırması

Supporting Institution

Bu çalışma herhangi bir kurum veya kuruluş tarafından desteklenmemiştir.

Ethical Statement

Bu çalışmada kullanılan veriler Kaggle ('Gaming and Mental Health') üzerinden halka açık ve anonim olarak temin edilmiştir. Çalışma kapsamında herhangi bir canlı denek üzerinde deney veya anket yapılmadığı için etik kurul onayı gerekmemektedir.

Thanks

Yazarlar, bu çalışma süresince sağlanan araştırma olanakları dolayısıyla Kütahya Dumlupınar Üniversitesi Bilgisayar Mühendisliği Bölümü'ne teşekkür eder.

References

1. [1] Ş. Sağıroğlu, H. İ. Bülbül, A. Kılıç, M. Küçükali, Ş. Bayzan ve Y. Samur, Ed., Dijital Oyunlar- 1: Araçlar, Metodolojiler, Uygulamalar ve Öneriler, Ankara, Türkiye: Nobel Akademik Yayıncılık, 2022.
2. [2] K. Lindenberg, S. Kindt, and C. Szász-Janocha, “Effectiveness of cognitive behavioral therapy–based in-tervention in preventing gaming disorder and unspecified internet use disorder in adolescents: A cluster ran-domized clinical trial,” JAMA network open, vol. 5, no. 2, p. e2148995, 2022.
3. [3] J. Stevenson, R. Harding, S. Ofori, M. B. Wall, and R. Das, “Task-based functional connectivity in internet gaming disorder: A systematic review,” Progress in Neuro-Psychopharmacology and Biological Psychiatry, p. 111637, 2026.
4. [4] X. Zhou, J. Chen, J. Yao, D. T. Chui, and S.-K. Tang, “Internet gaming disorder, attention deficit hyperac-tivity disorder and learning in adults: a systematic review,” Frontiers in Psychiatry, vol. 16, p. 1735922, 2026.
5. [5] C. Marino, N. Canale, A. Vieno, G. Caselli, L. Scacchi, and M. M. Spada, “Social anxiety and Internet gam-ing disorder: The role of motives and metacognitions,” Journal of behavioral addictions, vol. 9, no. 3, pp. 617–628, 2020.
6. [6] E. B. Karnadi and S. Pangestu, “Does internet gaming disorder hinder academic performance?,” Jurnal Kajian Bimbingan dan Konseling, vol. 6, no. 1, p. 12, 2024.
7. [7] N. S. Hawi, M. Samaha, and M. D. Griffiths, “Internet gaming disorder in Lebanon: Relationships with age, sleep habits, and academic achievement,” Journal of behavioral addictions, vol. 7, no. 1, pp. 70–78, 2018.
8. [8] J. Han, Y. Seo, H. Hwang, S. M. Kim, and D. H. Han, “Efficacy of cognitive behavioural therapy for inter-net gaming disorder,” Clin Psychology and Psychoth, vol. 27, no. 2, pp. 203–213, Mar. 2020, doi: 10.1002/cpp.2419.

9. [9]M. P. Banerjee and N. Kumar, “An AI-Driven Intelligent System for Early Detection of Gaming Addiction and Harmful Digital Behaviour Among Indian Adolescents,” 2026, Accessed: Apr. 25, 2026. [Online]. Available: https://www.researchgate.net/profile/Pooja-Banerjee-7/publication/399842756_An_AI-Driv-en_Intelligent_System_for_Early_Detection_of_Gaming_Addiction_and_Harmful_Digital_Behaviour_Among_Indian_Adolescents/links/69a01b1b42f94d1212af160d/An-AI-Driven-Intelligent-System-for-Early-Detection-of-Gaming-Addiction-and-Harmful-Digital-Behaviour-Among-Indian-Adolescents.pdf
10. [10] American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, DSM-5-TR. American Psychiatric Association Publishing, 2022. doi: 10.1176/appi.books.9780890425787.
11. [11] H. M. Pontes and M. D. Griffiths, “Measuring DSM-5 internet gaming disorder: Development and valida-tion of a short psychometric scale,” Computers in human behavior, vol. 45, pp. 137–143, 2015.
12. [12] O. T. Arıcak, M. Dinç, M. Yay, ve M. D. Griffiths, “İnternet oyun oynama bozukluğu ölçeği kısa for-mu’nun (ioobö9-kf) türkçeye uyarlanması: Geçerlik ve güvenirlik çalışması”, Addicta: The Turkish Journal on Addictions, c. 5, sy. 4, ss. 615-636, 2018.
13. [13] W. A. Behaviours, “Gaming disorder.” WHO, 2020.
14. [14] J. B. Saunders, H.-J. Rumpf, N. Carragher, and V. Poznyak, “The Development of and Rationale for Gam-ing Disorder in ICD-11 and a Review of Available Assessment Tools,” Curr Addict Rep, vol. 12, no. 1, p. 18, Feb. 2025, doi: 10.1007/s40429-025-00631-0.
15. [15] D. B. Olawade, O. Z. Wada, A. Odetayo, A. C. David-Olawade, F. Asaolu, and J. Eberhardt, “Enhancing mental health with Artificial Intelligence: Current trends and future prospects,” Journal of medicine, surgery, and public health, vol. 3, p. 100099, 2024.
16. [16] E. K. Ozyirmidokuz, B. A. Celik, and E. A. Stoica, “# GAMEADDICTED: A Machine Learning Frame-work for Digital Game Addiction Detection, and Early Intervention,” IEEE Access, 2025, Accessed: Apr. 25, 2026. [Online]. Available: https://ieeexplore.ieee.org/abstract/document/11075684/
17. [17] Y. Gan et al., “Application of machine learning in predicting adolescent Internet behavioral addiction,” Frontiers in psychiatry, vol. 15, p. 1521051, 2025.
18. [18] F. Hasan, S. M. Rahman, M. S. Rahman, and R. Khatun, “Analyzing Game Addiction in Bangladesh Through Machine Learning Techniques,” in 2025 International Conference on Quantum Photonics, Artificial Intelligence, and Networking (QPAIN), IEEE, 2025, pp. 1–6. Accessed: Apr. 26, 2026. [Online]. Available: https://ieeexplore.ieee.org/abstract/document/11171979/
19. [19] R. S. Priya and S. S. Priscila, “Predicting Online Game Disorder in Youngsters Using K-Nearest Neigh-bors and Artificial Neural Networks,” in Multidisciplinary Advancements in Human-AI Augmentation, IGI Global Scientific Publishing, 2026, pp. 199–224. Accessed: Apr. 26, 2026. [Online]. Available: https://www.igi-global.com/chapter/predicting-online-game-disorder-in-youngsters-using-k-nearest-neighbors-and-artificial-neural-networks/389935
20. [20] N. N. Arun and W. Choothong, “Ensemble learning models for predicting the gaming addiction behav-iours of adolescents,” IJDMMM, vol. 17, no. 1, pp. 103–125, 2025, doi: 10.1504/IJDMMM.2025.144623.
21. [21] B. Jeong et al., “Multiple-kernel support vector machine for predicting internet gaming disorder using multimodal fusion of PET, EEG, and clinical features,” Frontiers in Neuroscience, vol. 16, p. 856510, 2022.
22. [22] A. M. A. A. Ali, “Explainability Driven Deep Learning Models for Intelligent IoT Based Diagnosis of In-ternet Gaming Disorder,” 2026, Accessed: Apr. 26, 2026. [Online]. Available: https://www.researchgate.net/profile/Ali-Mir-As-if/publication/400827910_Explainability_Driven_Deep_Learning_Models_for_Intelligent_IoT_Based_Diagnosis_of_Internet_Gaming_Disorder/links/69931a5612f837212a1b2e14/Explainability-Driven-Deep-Learning-Models-for-Intelligent-IoT-Based-Diagnosis-of-Internet-Gaming-Disorder.pdf
23. [23] S. Shahid, “Gaming and Mental Health,” Kaggle, 2026. [Online]. Available: https://www.kaggle.com/datasets/shaistashahid/gaming-and-mental-health (Accessed: Apr. 17, 2026).
24. [24] J. Tan, J. Yang, S. Wu, G. Chen, and J. Zhao, “A critical look at the current train/test split in machine learning,” Jun. 08, 2021, arXiv: arXiv:2106.04525. doi: 10.48550/arXiv.2106.04525.
25. [25] E. D. Öz, M. Küçük, R. Uyar ve D. Özdemir, Derin Öğrenme Teknikleri: Teoriden Uygulamaya, 1. bs., Ankara, Türkiye: Nobel Akademik Yayıncılık, 2025.
26. [26] A. J. Mohammed, M. Muhammed Hassan, and D. Hussein Kadir, “Improving classification performance for a novel imbalanced medical dataset using SMOTE method,” International Journal of Advanced Trends in Computer Science and Engineering, vol. 9, no. 3, pp. 3161–3172, 2020.
27. [27] A. M. Halim, M. Dwifebri, and F. Nhita, “Handling imbalanced data sets using SMOTE and ADASYN to improve classification performance of Ecoli data sets,” Building of informatics, technology and science (bits), vol. 5, no. 1, pp. 246–253, 2023.
28. [28] N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, “SMOTE: synthetic minority over-sampling technique,” Journal of artificial intelligence research, vol. 16, pp. 321–357, 2002.
29. [29] N. M. Khan, N. Madhav C, A. Negi, and I. S. Thaseen, “Analysis on Improving the Performance of Ma-chine Learning Models Using Feature Selection Technique,” in Intelligent Systems Design and Applications, vol. 941, A. Abraham, A. K. Cherukuri, P. Melin, and N. Gandhi, Eds., in Advances in Intelligent Systems and Computing, vol. 941. , Cham: Springer International Publishing, 2020, pp. 69–77. doi: 10.1007/978-3-030-16660-1_7.
30. [30] X. Cheng, “A comprehensive study of feature selection techniques in machine learning models,” Availa-ble at SSRN 5154947, 2024, Accessed: Apr. 26, 2026. [Online]. Available: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5154947
31. [31] J. M. N. Abad and A. Soleimani, “Novel feature selection algorithm for thermal prediction model,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 26, no. 10, pp. 1831–1844, 2018.
32. [32] D. Theng and K. K. Bhoyar, “Feature selection techniques for machine learning: a survey of more than two decades of research,” Knowl Inf Syst, vol. 66, no. 3, pp. 1575–1637, Mar. 2024, doi: 10.1007/s10115-023-02010-5.
33. [33] R. Iranzad and X. Liu, “A review of random forest-based feature selection methods for data science edu-cation and applications,” Int J Data Sci Anal, vol. 20, no. 2, pp. 197–211, Aug. 2025, doi: 10.1007/s41060-024-00509-w.
34. [34] H. Jeon and S. Oh, “Hybrid-recursive feature elimination for efficient feature selection,” Applied Scienc-es, vol. 10, no. 9, p. 3211, 2020.
35. [35] E. B. Hunt, Artificial intelligence. Academic Press, 2014. Accessed: Apr. 26, 2026. [Online]. Available: https://books.google.com/books?hl=tr&lr=&id=9y2jBQAAQBAJ&oi=fnd&pg=PP1&dq=%5B36%5D%09Hunt,+E.+B.+(2014).+Artificial+intelligence.+Academic+Press.&ots=uQYAXlc_xA&sig=RNxenbSjCeTUvr4krUSkUGvayNE
36. [36] S. S. Ray et al., “Leveraging ChatGPT and Bard: What does it convey for water treatment/desalination and harvesting sectors?,” Desalination, vol. 570, p. 117085, 2024.
37. [37] C. Janiesch, P. Zschech, and K. Heinrich, “Machine learning and deep learning,” Electron Markets, vol. 31, no. 3, pp. 685–695, Sep. 2021, doi: 10.1007/s12525-021-00475-2.
38. [38] P. C. Sen, M. Hajra, and M. Ghosh, “Supervised Classification Algorithms in Machine Learning: A Sur-vey and Review,” in Emerging Technology in Modelling and Graphics, vol. 937, J. K. Mandal and D. Bhattacharya, Eds., in Advances in Intelligent Systems and Computing, vol. 937. , Singapore: Springer Singa-pore, 2020, pp. 99–111. doi: 10.1007/978-981-13-7403-6_11.
39. [39] V. Sheth, U. Tripathi, and A. Sharma, “A comparative analysis of machine learning algorithms for classi-fication purpose,” Procedia Computer Science, vol. 215, pp. 422–431, 2022.
40. [40] F. Y. Osisanwo, J. E. Akinsola, O. Awodele, J. O. Hinmikaiye, O. Olakanmi, and J. Akinjobi, “Supervised machine learning algorithms: classification and comparison,” International Journal of Computer Trends and Technology (IJCTT), vol. 48, no. 3, pp. 128–138, 2017.
41. [41] M. Suyal and P. Goyal, “A review on analysis of k-nearest neighbor classification machine learning algo-rithms based on supervised learning,” International Journal of Engineering Trends and Technology, vol. 70, no. 7, pp. 43–48, 2022.
42. [42] A. F. Reza, R. Singh, R. K. Verma, A. Singh, Y.-H. Ahn, and S. S. Ray, “An integral and multidimensional review on multi-layer perceptron as an emerging tool in the field of water treatment and desalination process-es,” Desalination, vol. 586, p. 117849, 2024.