Modeling Player Performance in Massively Multiplayer Online Role-Playing Games: The Effects of Diversity in Mentoring Network

Publication Type

Conference Proceeding Article

Publication Date



This study investigates and reports preliminary findings on player performance prediction approaches which model player's past performance and social diversity in mentoring network in Ever Quest II, a popular massively multiplayer online role-playing game (MMORPG) developed by Sony Online Entertainment. Our contributions include a better understanding of performance metrics used in the game and a foundation of recommendation systems for mentors and apprentices. We examined three different game servers from the Ever Quest II game logs. In all three servers, the results from our analyses suggest that increase in social diversity in terms of characters and classes encountered moderately negatively correlates with player performance. Based on this finding, we built predictive models to predict player's future performance based on past performance and social diversity in terms of mentoring activities. Our results indicate that 1) models employing past performance and social diversity perform better and 2) prediction for mentors is generally better than that for apprentices.


massively multiplayer online games, mentoring, player performance, video games


Databases and Information Systems | Numerical Analysis and Scientific Computing

Research Areas

Data Management and Analytics


2011 International Conference on Advances in Social Networks Analysis and Mining





City or Country

Kaohsiung, Taiwan

Additional URL