Grant: Alliance Grants
University of Victoria
$25,000.00
Behavioural biomarkers of problematic gaming through data mining and semi-supervised machine learning
FY 2024-25
Summary
In 2023, 79% of the global online population play video games. Although playing videogames can lead to cognitive, social, and emotional benefits, it can also lead to obsessive and addictive playing, which is associated with a range of physical, psychological, social, and vocational harms. Knowing when gaming leads to benefits versus harms is not simple, as harmful outcomes are a complex relationship between the player and their context.
To find patterns of behaviour that represent harm, researchers have proposed machine learning (ML) models trained on the large amount of low-level data that is generated when players engage with games. Videogame distribution platforms collect massive amounts of behavioural data from players, which can be accessed through Application Programming Interfaces (APIs). However, the lack of publicly available large labeled datasets (i.e., game data that has fields corresponding to harm, such as obsessive passion, toxicity, or gaming addiction) prevents progress in building ML models of how gaming leads to benefits or harms. To address this gap, in this project, we propose to build semi-supervised computational models of behaviours that lead to harm, using both unsupervised and supervised ML algorithms of labeled and pseudo-labeled datasets.
In the proposed research, our objective is to employ advanced data mining and machine learning models on game data to identify behavioural patterns associated with harm. We will use a small labelled dataset to seed the gathering of a large pseudo-labeled dataset that will be used for semi-supervised machine learning approaches to detect clusters of players exhibiting different patterns, and associated these with harmful outcomes.
Program
Alliance Grants
Recipient
Location
British Columbia, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
597280-2024
Name nom
Mandryk, Regan RL
Department département
Computer Science
Organization id
7
Institution établissement
University of Victoria
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
25000
Program id
ALLRP
Program name en
Alliance Grants
Program name fr
Subventions Alliance
Group en
RESEARCH PARTNERSHIPS
Group fr
PARTENARIATS DE RECHERCHE
Committee code
96
Committee name en
RPP Internal Decision Cttee
Committee name fr
Cté de décision interne - PPR
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2700
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Information technology
Research subject fr
Technologie de l'information
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Behavioural biomarkers of problematic gaming through data mining and semi-supervised machine learning
Keywords
biomarkers; computational models; data mining; digital games; machine learning; phenotyping; problematic gaming; semi-supervised machine learning
Application summary
In 2023, 79% of the global online population play video games. Although playing videogames can lead to cognitive, social, and emotional benefits, it can also lead to obsessive and addictive playing, which is associated with a range of physical, psychological, social, and vocational harms. Knowing when gaming leads to benefits versus harms is not simple, as harmful outcomes are a complex relationship between the player and their context.
To find patterns of behaviour that represent harm, researchers have proposed machine learning (ML) models trained on the large amount of low-level data that is generated when players engage with games. Videogame distribution platforms collect massive amounts of behavioural data from players, which can be accessed through Application Programming Interfaces (APIs). However, the lack of publicly available large labeled datasets (i.e., game data that has fields corresponding to harm, such as obsessive passion, toxicity, or gaming addiction) prevents progress in building ML models of how gaming leads to benefits or harms. To address this gap, in this project, we propose to build semi-supervised computational models of behaviours that lead to harm, using both unsupervised and supervised ML algorithms of labeled and pseudo-labeled datasets.
In the proposed research, our objective is to employ advanced data mining and machine learning models on game data to identify behavioural patterns associated with harm. We will use a small labelled dataset to seed the gathering of a large pseudo-labeled dataset that will be used for semi-supervised machine learning approaches to detect clusters of players exhibiting different patterns, and associated these with harmful outcomes.