The rapid expansion of online gambling has increased the demand for automated methods to identify problematic player behavior. While psychological research provides clinical criteria for addiction, the computational operationalization of these concepts into platform-level markers remains underutilized. This short paper presents a Systematic Literature Review (SLR) of addiction markers extracted from player tracking data in online betting and casino platforms. By analyzing 9 empirical studies selected from an initial pool of 141, we identified 22 distinct markers, ranging from traditional monetary indicators to platform-interaction signals such as canceled withdrawals and responsible gambling tool settings. Our analysis reveals a methodological evolution: while statistical models remain prevalent, recent studies increasingly leverage machine learning (e.g., Random Forest, Gradient Boosting) to predict high-risk user trajectories and behavioral transitions. The review highlights the need for more standardized definitions and evaluation practices to support the development of data-driven approaches for early detection of gambling-related harm.
Kouyoumdjian et al. (2026) studied this question.