Using machine learning to predict online gambling harm

View Abstract View Snapshot Back to Search Results

snapshot summaries

Author(s): Murch, W. Spencer ; Kairouz, Sylvia ; Dauphinais, Sophie ; Picard, Elyse ; Costes, Jean-Michel ; French, Martin

Journal: Addiction

Year Published: 2023

Date Added: May 18, 2023

Online gambling is becoming more and more popular. Research has shown that data collected by online gambling platforms can be used to give personalized assessment or referrals to treatment services for users who experience harm. The researchers used the Problem Gambling Severity Index (PGSI) and machine learning to classify people who are at higher risk of online problem gambling. Using data collected from people registered on lotoquebec.com, the researchers found that machine learning algorithms could be used with online gambling data to predict whether a person belongs to a PGSI higher-risk category.

Citation: Murch, W. S., Kairouz, S., Dauphinais, S., Picard, E., Costes, Jean-Michel, & French, M. (2023). Using machine learning to retrospectively predict self-reported gambling problems in Quebec. Addiction. Advance online publication. https://doi.org/10.1111/add.16179

Article DOI: https://doi.org/10.1111/add.16179

Keywords: behavioural addictions ; behavioural tracking ; machine learning ; online gambling ; problem gambling

Topics: Information for Operators ; Online Gambling

Conceptual Framework Factors: Exposure - Gambling Setting ; Types - Structural Characteristics ; Environment - Responsible Gambling ; Exposure - Accessibility ; Psychological Factors ; Resources - Harm Reduction, Prevention, and Protection ; Gambling Resources

Study Design: Observational: Cross-sectional

Geographic Coverage: Canada, Quebec

Study Population: Adults registered on the lotoquebec.com platform and had placed at least one bet in the last 12 months (n = 9145, with 88 participants excluded for not having placed a bet or made a deposit)

Sampling Procedure: Sent out invitation to all the people registered on lotoquebec.com and kept people who replied and people who fit their criteria.

Study Funding:

This study was funded by a Concordia University Horizon Postdoctoral Fellowship, the Mise sur toi Foundation, and the Fonds de recherche du Québec – Société et Culture.