Using machine learning to predict potential online gambling addicts.

Loading...
Thumbnail Image

URL

Journal Title

Journal ISSN

Volume Title

Perustieteiden korkeakoulu | Master's thesis

Date

2018-10-08

Department

Major/Subject

Computer Science

Mcode

SCI3042

Degree programme

Master’s Programme in Computer, Communication and Information Sciences

Language

en

Pages

45

Series

Abstract

Betting addicts on the gambling websites are difficult to identify because online gambling is by nature different from real gambling. This thesis attempts to identify potential gambling addicts in an online gambling website X using machine learning models. The models are based on user’s usage history on the website. The usage data is collected for each user from the site using JavaScript. The data is then analyzed and stored in a database. Machine learning models are then trained using Support Vector Machines with the data of users who are by definition problem gamblers. The system then makes a prediction for all active users based on their recent usage history. The final results include an automated system for daily learning and prediction of potential problem gamblers who show early signs of gambling addiction.

Description

Supervisor

Gionis, Aristides

Thesis advisor

Mantere, Markku

Keywords

betting addiction, machine learning, classification, predictive analysis

Other note

Citation