Secondary navigation

Pioneering research to detect early warning signs of harmful online gambling

Pioneering research to detect early warning signs of harmful online gambling

Published:10 August 2017

Online gambling image courtesy of John Seb Barber via Flickr Creative Commons.

CQUniversity has been commissioned to analyse wagering industry data to develop a list of behavioural indicators and a predictive algorithm that can detect early warning signs of harmful betting.

The pioneering research has been jointly funded for $300 000 (ex-GST) by the Commonwealth Department of Social Services and the Victorian Responsible Gambling Foundation.

Lead investigator, Professor Nerilee Hing said that "developing this predictive model to ultimately protect consumers who bet online will be the first research of its kind in Australia”.

She explained that the predictive algorithm can be embedded into wagering operators’ websites to automatically detect customers at risk.

The research team, also comprising CQUniversity’s Associate Professor Matthew Browne, Professor Matthew Rockloff, Professor Brijesh Verma and Dr Alex Russell, will work with online wagering data to develop this tool that allows the wagering provider to intervene and protect customers at risk of harm.

It will enable online gambling operators to identify customers displaying signs of harm, and can also be embedded into online wagering systems to generate dynamic messages tailored to individual customers based on their betting patterns.

Associate Professor Browne explained, “the developed algorithm will utilise statistical machine learning applied to actual wagering transactions data".

"Unlike some commercially available systems developed overseas, our model will be independently developed from industry and non-proprietary – making it more transparent. We will be testing the predictions of our model using established measures of gambling harm and problem gambling severity.”

Professor Verma said “we will develop novel feature selection and ensemble learning techniques and incorporate them into our predictive model which will help us in achieving highly accurate predictions and also in identifying the most significant indicators from real-world wagering industry data".

"Enhancing consumer protection in online wagering is important because rates of harmful gambling are much higher amongst online gamblers than amongst those using other platforms," he said.

"Through using historical data, this research will attempt to identify patterns of behaviour that occur prior to significant gambling harms. Such a tool would allow wagering operators and customers to take appropriate action before experiencing severe gambling harm."

The total funding is $300 000 with the Commonwealth contributing $200 000 to the research and the Victorian Responsible Gambling Foundation contributing $100 000 to the project.

The project will be managed by the Victorian Responsible Gambling Foundation as part of its strategic research program that focuses on reducing harm from gambling.