Predicting future problem gamblers using Machine Learning Algorithms

Abstract: The work in this thesis attempts to identify potential gambling addicts inan online gambling website using machine learning models. Machine Learning can play a major role in predicting and identifying high-risk online players leading to self-exclusion and providing automated self- help tools to problem gamblers. It helps to predict problem gamblers based on past usage history Machine learning aids in the training of data from users who are problem gamblers by definition. This was accomplished with the help of supervisedlearning, specifically using a classification algorithm such as Support Vector Machine and Naive Bayes. The player tracking system then creates a prediction for all active users based on their behavioral patterns. The final results would include using existing player behavior data to add predictions to improve the model and make it self-learning to detect potential gamblers. The system then makes a prediction for all active users based on theirrecent usage history. The final result includes a system for analyzing thepotential gambling addicts compare to those non-addicts and gambling ofpotential problem gamblers who show gambling signs of gambling addiction in order to predict those in future gambling sites.