With more and more games being played on the mobile phone platform, there is a need to enhance mobile gamer modelling and to improve the understanding of their preferences towards different aspects of game performance. This paper is the first to conduct a large scale study of the US mobile gaming market to collect profiling data and quantify the user-game experience. Our approach utilizes the unsupervised clustering method K-prototypes to classify mobile gamers and successfully identifies five distinctive groups. Moreover, for each of the discovered groups, we quantify gamer’s preferences for six device performance factors generalized as: visual smoothness, image quality, battery life, temperature, loading time and touch latency. The results of gamer modelling and their weighted preference scores could contribute to commercial use cases such as mobile game benchmarking and marketing.
[1]
Florian Baumann,et al.
Hardcore Gamer Profiling: Results from an unsupervised learning approach to playing behavior on the Steam platform
,
2018,
KES.
[2]
Yang Liu,et al.
The Game Performance Index for Mobile Phones
,
2019,
ArXiv.
[3]
Zhexue Huang,et al.
CLUSTERING LARGE DATA SETS WITH MIXED NUMERIC AND CATEGORICAL VALUES
,
1997
.
[4]
Ana Fernández del Río,et al.
Profiling Players with Engagement Predictions
,
2019,
2019 IEEE Conference on Games (CoG).
[5]
Anders Drachen,et al.
Tweeting your Destiny: Profiling Users in the Twitter Landscape around an Online Game
,
2019,
2019 IEEE Conference on Games (CoG).