Researchers from ITMO University report that they have predicted personality traits, such as gender, using data from an online gaming platform. This is one of the first machine learning studies to be applied to large amounts of gaming data. The approach could improve systems that recommend games to users. It could also be used to detect gaming addiction .The results were presented at the AAAI conference.
Video games are firmly established in popular culture, and the number of online and offline products for gaming platforms is growing day by day. In turn, users generate increasing amounts of data, which can be used to develop gaming behavior models or to identify personal characteristics. This is useful, for example, for the early detection of gaming addiction, as well as for marketing research in the video game sector.
Until now, the majority of gaming research has been done manually, on small data sets. However, in order to draw statistically significant conclusions, it is necessary to analyze large amounts of data. Scientists from ITMO University and the National University of Singapore are now among the first to use machine learning for this purpose. Using data collected on the behavior of users of the gaming platform Steam and a specially developed and trained model, the scientists were able to predict the gender of the player from their behavior during the game.
The database for the analysis was collected from the service Player.me, which provides information about Steam accounts and social media. By comparing users' gaming data with posts on Twitter, Facebook and Instagram, the researchers discovered links between gaming behavior and personal characteristics. As a result, the model was based on characteristics such as time spent playing, achievements, preferred game genres, in-game payments, etc.
“The idea of our research is to use the data provided by games to study human behavior in real life. Social networks seem to be a good source for such information. However, people think about their behavior on social networks: they choose what to post and what not to post, while when playing games, they behave as in real life, without thinking about many things,” notes Ivan Samborskii, a graduate student at ITMO University.
According to the scientists, analyzing video game data can help discover the interests, location, gender, and demographics of users, as well as assess the time a person is willing to spend playing. The researchers will work to improve the resulting model, making predictions about users more accurate. They also plan to adopt a model for predicting gaming addiction.
