Content Based Player and Game Interaction Model for Game Recommendation in the Cold Start setting

Game recommendation is an important application of recommender systems. Recommendations are made possible by data sets of historical player and game interactions, and sometimes the data sets include features that describe games or players. Collaborative filtering has been found to be the most accurate predictor of past interactions. However, it can only be applied to predict new interactions for those games and players where a significant number of past interactions are present. In other words, predictions for completely new games and players is not possible. In this paper, we use a survey data set of game likes to present content based interaction models that generalize into new games, new players, and both new games and players simultaneously. We find that the models outperform collaborative filtering in these tasks, which makes them useful for real world game recommendation. The content models also provide interpretations of why certain games are liked by certain players for game analytics purposes.

[1]  Juho Hamari,et al.  Five-Factor Inventory of Intrinsic Motivations to Gameplay (IMG) , 2019, HICSS.

[2]  Michael Mateas,et al.  People Tend to Like Related Games , 2015, FDG.

[3]  Yehuda Koren,et al.  On the Difficulty of Evaluating Baselines: A Study on Recommender Systems , 2019, ArXiv.

[4]  Nick Yee,et al.  Motivations for Play in Online Games , 2006, Cyberpsychology Behav. Soc. Netw..

[5]  Steven L. Lytinen,et al.  Using Game Reviews to Recommend Games , 2014 .

[6]  Gabriel Jacobs,et al.  Segmentation of the games market using multivariate analysis , 2005 .

[7]  Alan Said,et al.  Comparative recommender system evaluation: benchmarking recommendation frameworks , 2014, RecSys '14.

[8]  Yifan Hu,et al.  Collaborative Filtering for Implicit Feedback Datasets , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[9]  Les Nelson,et al.  Online gaming motivations scale: development and validation , 2012, CHI.

[10]  Vicente Julián,et al.  A CBR-Based Game Recommender for Rehabilitation Videogames in Social Networks , 2014, IDEAL.

[11]  Giuseppe Manai,et al.  HybridRank: A Hybrid Content-Based Approach To Mobile Game Recommendations , 2014, CBRecSys@RecSys.

[12]  Christian Bauckhage,et al.  Archetypal Game Recommender Systems , 2014, LWA.

[13]  Jonathan L. Herlocker,et al.  Evaluating collaborative filtering recommender systems , 2004, TOIS.

[14]  Ulrich Paquet,et al.  The Xbox recommender system , 2012, RecSys.

[15]  Tapio Pahikkala,et al.  Toward more realistic drug^target interaction predictions , 2014 .

[16]  Rita Orji,et al.  Towards a Trait Model of Video Game Preferences , 2018, Int. J. Hum. Comput. Interact..

[17]  Lior Rokach,et al.  Introduction to Recommender Systems Handbook , 2011, Recommender Systems Handbook.

[18]  Tapio Pahikkala,et al.  Fast Kronecker Product Kernel Methods via Generalized Vec Trick , 2016, IEEE Transactions on Neural Networks and Learning Systems.

[19]  K. P. Kallio,et al.  At Least Nine Ways to Play: Approaching Gamer Mentalities , 2011, Games Cult..

[20]  Dietmar Jannach,et al.  Offline performance vs. subjective quality experience: a case study in video game recommendation , 2017, SAC.

[21]  Jouni Smed,et al.  Validating gameplay activity inventory (GAIN) for modeling player profiles , 2018, User Modeling and User-Adapted Interaction.

[22]  Roberto Turrin,et al.  Performance of recommender algorithms on top-n recommendation tasks , 2010, RecSys '10.

[23]  Darryl Charles,et al.  Behavlets: a method for practical player modelling using psychology-based player traits and domain specific features , 2016, User Modeling and User-Adapted Interaction.