Development and 30-Month Revalidation of a Machine Learning System for Detecting Self-Reported Gambling Problems on an Online Gambling Platform
Bibliographic record
Abstract
Online gambling platforms are highly accessible, increasingly popular, and see relatively high rates of gambling-related harms. In response to these trends, we sought to develop machine learning models that detect at-risk online gamblers using transactional data collected over the course of their betting.\nUsers of a provincially-operated gambling website in Quebec, Canada were recruited in September 2019 (N = 9,145), and February 2022 (N = 11,258). Participants completed the Problem Gambling Severity Index (PGSI), and consented to release their online gambling data for the prior 12 months. After fitting two random forest classification models, our first (2019) and second (2022) validation studies correctly identified 81.94% and 81.88% of users at higher-risk for experiencing problems (PGSI ≥ 8). They further classified 72.20% and 73.94% of lower-risk (PGSI < 8) users on the site. Important features of harmful online gambling appear to include the variability of weekly betting amounts, and frequent cash deposits on the site.\nAlthough routine system evaluations remain necessary, these results indicate that a machine learning system can stably detect problem gambling risk over a 30-month period. They further allow researchers to estimate activity-specific harms, and discover behavioural factors related to gambling disorder using large, ecological datasets.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".