Europe represents the largest international market for online gambling, with a Gross Gaming Revenue (GGR) that reached €24.9 billion in 2020. To wit, online gambling is a major industry in every European country, generating billions of Euros in revenue for commercial actors and governments alike. Despite such evidently beneficial effects, online gambling is ultimately a vast social experiment with potentially disastrous social and personal consequences that could result in an overall deterioration of social and familial relationships.

In 2012, the European Commission released a statement highlighting the need for regulatory policies to aid in the detection of pathological gambling behaviours, citing “a responsibility to protect those citizens and families who suffer from a gambling addiction.” In Portugal, the Gambling Inspection and Regulation Service is “responsible for the control and regulation of gambling activities in casinos and bingo halls, as well as online gambling and betting.” This authority receives, daily, all data related to online gambling activities (the amount of money spent, the deposits made, the number of discrete login events, the number of minutes spent online, the hour of each access event, and the type of gambling activity selected (i.e., slot machines, poker, soccer, or other sporting events)) pursued by every user on every online platform with services that are accessible to Portuguese citizens. Despite this prodigious collection of data, the authority lacked appropriate tools for identifying gambling addicts and acknowledged a profound scarcity of actionable data regarding the actual scope of gambling addiction and a consequent lack of expertise about how best to deal with this problem.

To tackle this problem, Mauro Castelli and his team from NOVA IMS developed and implemented an AI (Artificial Intelligence)-based system that capitalizes on the vast amount of data collected every day by the online gambling providers operating in Portugal and, from its analysis, models the behaviours associated with addicted gamblers and allows for their early detection. The team proposed a system based on state-of-the-art machine learning methods including, but not limited to, the use of evolutionary-based algorithms and deep neural networks. The proposed system provided a transformative effect on the operations of the Portuguese gambling authority. In particular, the system allowed to compare and characterize the behaviours of online gamblers and to dynamically adapt its output based on the behavioural pattern observed every day for all the users in Portugal.

To provide an initial idea concerning the impact of the project, after the training phase of the machine learning algorithms, using historical data from 2019, we were able to identify approximately 1500 users corresponding to pathological players.

The Portuguese authority is currently using the proposed system, having today constant information concerning online gamblers, receiving automatic alerts when a user is moving toward a “dangerous” behaviour (i.e., indicating a possible addiction), and taking actions to protect citizens at risk of developing serious addiction by designing prevention actions.

As a result of this project, the team produced 3 PhD theses, 4 master theses, 32 journal publications, and 22 conference publications. Remarkably, despite being developed and tested in Portugal, it would be possible to extend the project adoption to all the EU countries where the same regulations concerning the control of gambling activities exist.