The gender of a player can be recognized by his behavior.

Scientists from the St. Petersburg National Research University of Information Technologies, Mechanics and Optics with the help of machine learning were able to predict the gender of the player based on data about users of the online gaming platform.

Video games have long been firmly established in modern life: the number of online and offline products for various platforms is growing every day. Every day their users generate more and more data that can be used to develop models of game behavior or determine the personal characteristics of players. This is useful, for example, for early detection of gaming addiction, as well as for marketing research in the gaming field.

Until now, all game data research has been done manually in small samples. However, in order to make statistically significant conclusions, it is necessary to analyze large arrays of game data. Scientists from ITMO University, together with colleagues from the University of Singapore, were among the first to use machine learning and using the collected set of data on the behavior of users of the Steam gaming platform, as well as a specially developed and trained model, were able to predict its gender based on player behavior.

The database for analysis was collected by researchers based on the Player service. me, which allows you to match the accounts of Steam users with their profiles on social networks: Twitter, Facebook and Instagram. Based on this comparison, researchers looked for connections between data on game behavior and demographic indicators, and as a result, the model was based on such signs as time spent on the game, game achievements achieved, preferred game genres, the availability of in-game payments and others.

According to scientists, the analysis of game data allows you to assess the interests, location and demographics of users, as well as take into account how much time a person is willing to spend on games. Researchers will work on improving the resulting model to improve the accuracy of predictions about users. It is also planned to adapt it to determine the game addiction.

popmech.ru