Predicting online gambling self-exclusion: an analysis of the performance of supervised machine learning models

Percy, Chris; França, Manoel Vitor Macedo; Dragičević, Simo; Garcez, Artur S. d’Avila

doi:10.1080/14459795.2016.1151913

Cited by 57 publications

(55 citation statements)

References 31 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This provides face validity for Playscan's assessment. Previous studies ( Adami et al, 2013 , Braverman and Shaffer, 2012 , Dragicevic et al, 2011 , Haeusler, 2016 , Percy et al, 2016 , Philander, 2013 ) have examined different types of gambling patterns and how they relate to risk or self-exclusion. The studies validated that gambling data can be used to identify high risk behaviour and to identify self-excluders.…”

Section: Discussionmentioning

confidence: 99%

Experiences of Playscan: Interviews with users of a responsible gambling tool

Forsström

Jansson‐Fröjmark

Hesser

et al. 2017

Internet Interventions

View full text Add to dashboard Cite

Online gambling, encompassing a wide variety of activities and around-the-clock access, can be a potential risk factor for gamblers who tend to gamble excessively. Yet, the advent of online gambling has enabled responsible gambling (RG) features that may help individuals to limit their gambling behaviour. One of these features is RG tools that track gamblers' behaviour, performs risk assessments and provides advice to gamblers. This study investigated users' views and experiences of the RG tool Playscan from a qualitative perspective using a semi-structured interview. The tool performs a risk assessment on a three-step scale (low, medium and high risk). Users from every risk category were included. Twenty interviews were carried out and analysed using thematic analysis. Two main themes with associated sub-themes were identified: “Usage of Playscan and the gambling site” and “Experiences of Playscan”. Important experiences in the sub-themes were lack of feedback from the tool and confusion when signing up to use Playscan. These experiences counteracted positive attitudes that should have promoted usage of the tool. Providing more feedback directly to users is a suggested solution to increase usage of the RG tool.

show abstract

Section: Discussionmentioning

confidence: 99%

Experiences of Playscan: Interviews with users of a responsible gambling tool

Forsström

Jansson‐Fröjmark

Hesser

et al. 2017

Internet Interventions

View full text Add to dashboard Cite

show abstract

“…Notes: Limit, the chosen loss limit between January and March 2017; Number of playing days, number of active days between January and March 2017; red, Playscan status high risk between January and March 2017; Feedback, whether players received feedback that they had reached 80% of their loss limit between January and March 2017 technique. This algorithm performed best in two previous gambling-related studies that attempted to predict voluntary self-exclusion (i.e., Percy et al 2016;Philander 2014). Logistic regression was chosen because it is a classical and well-proven statistical method.…”

Section: Discussionmentioning

confidence: 99%

“…However, supervised techniques have been applied to predict problem gambling. For instance, Percy et al (2016) predicted self-exclusion with a sample of 845 online gamblers. They compared different statistical methods and found that the random forest technique performed best (see "Methods" section for a description of this technique and others).…”

Section: Limit-setting Tools In Gambling Environmentsmentioning

confidence: 99%

Predicting Limit-Setting Behavior of Gamblers Using Machine Learning Algorithms: A Real-World Study of Norwegian Gamblers Using Account Data

Auer¹,

Griffiths

2019

Int J Ment Health Addiction

View full text Add to dashboard Cite

Player protection and harm minimization have become increasingly important in the gambling industry along with the promotion of responsible gambling (RG). Among the most widespread RG tools that gaming operators provide are limit-setting tools that help players limit the amount of time and/or money they spend gambling. Research suggests that limit-setting significantly reduces the amount of money that players spend. If limitsetting is to be encouraged as a way of facilitating responsible gambling, it is important to know what variables are important in getting individuals to set and change limits in the first place. In the present study, 33 variables assessing the player behavior among Norsk Tipping clientele (N = 70,789) from January to March 2017 were computed. The 33 variables which reflect the players' behavior were then used to predict the likelihood of gamblers changing their monetary limit between April and June 2017. The 70,789 players were randomly split into a training dataset of 56,532 and an evaluation set of 14,157 players (corresponding to an 80/20 split). The results demonstrated that it is possible to predict future limit-setting based on player behavior. The random forest algorithm appeared to predict limit-changing behavior much better than the other algorithms. However, on the independent test data, the random forest algorithm's accuracy dropped significantly. The best performance on the test data along with a small decrease in accuracy in comparison to the training data was delivered by the gradient boost machine learning algorithm. The most important variables predicting future limit-setting using the gradient boost machine algorithm were players receiving feedback that they had reached 80% of their personal monthly global loss limit, personal monthly loss limit, the amount bet, theoretical loss, and whether the players had increased their limits in the past. With the help of predictive analytics, players with a high likelihood of changing their limits can be proactively approached.

show abstract

“…As ML-based diagnostic classification has been recently applied to other addictive behaviors and disorders ( 24 – 28 ), symptom-based categorization of IGD also appears to face a challenge of computation-based classification. Because anatomical abnormalities of the brain following IGD have been repeatedly reported in previous studies ( 5 – 7 , 9 ), we considered such neuroanatomical information from brain imaging data potential biomarkers for the diagnosis of IGD.…”

Section: Discussionmentioning

confidence: 99%

Discriminating Pathological and Non-pathological Internet Gamers Using Sparse Neuroanatomical Features

et al. 2018

View full text Add to dashboard Cite

Internet gaming disorder (IGD) is often diagnosed on the basis of nine underlying criteria from the latest version of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). Here, we examined whether such symptom-based categorization could be translated into computation-based classification. Structural MRI (sMRI) and diffusion-weighted MRI (dMRI) data were acquired in 38 gamers diagnosed with IGD, 68 normal gamers diagnosed as not having IGD, and 37 healthy non-gamers. We generated 108 features of gray matter (GM) and white matter (WM) structure from the MRI data. When regularized logistic regression was applied to the 108 neuroanatomical features to select important ones for the distinction between the groups, the disordered and normal gamers were represented in terms of 43 and 21 features, respectively, in relation to the healthy non-gamers, whereas the disordered gamers were represented in terms of 11 features in relation to the normal gamers. In support vector machines (SVM) using the sparse neuroanatomical features as predictors, the disordered and normal gamers were discriminated successfully, with accuracy exceeding 98%, from the healthy non-gamers, but the classification between the disordered and normal gamers was relatively challenging. These findings suggest that pathological and non-pathological gamers as categorized with the criteria from the DSM-5 could be represented by sparse neuroanatomical features, especially in the context of discriminating those from non-gaming healthy individuals.

show abstract

Predicting online gambling self-exclusion: an analysis of the performance of supervised machine learning models

Cited by 57 publications

References 31 publications

Experiences of Playscan: Interviews with users of a responsible gambling tool

Experiences of Playscan: Interviews with users of a responsible gambling tool

Predicting Limit-Setting Behavior of Gamblers Using Machine Learning Algorithms: A Real-World Study of Norwegian Gamblers Using Account Data

Discriminating Pathological and Non-pathological Internet Gamers Using Sparse Neuroanatomical Features

Contact Info

Product

Resources

About