TOPOLOGICAL CLUES FOR PREDICTING OUTCOMES OF MULTIPLAYER ONLINE BATTLE ARENA GAMES

Dominik Deja; Mateusz Myślak

TOPOLOGICAL CLUES FOR PREDICTING OUTCOMES OF MULTIPLAYER ONLINE BATTLE ARENA GAMES

Abstract

With 27 million people playing League of Legends every day, e-sports became more and more important part of our everyday life. Rise of its popularity builds a demand for better understanding e-sports mechanics on a deeper level. In the article, we test a hypothesis that it is possible to predict an outcome of Multiplayer Online Battle Arena game based on topological clues only (such as area of polygon where vertices are players positions, and their dispersion). With accuracy, precision, and recall around 80%, we show that it is possible to construct a model which will be both: generic for MOBA games genre, and based on simple assumptions, which are hard to be perceived by the naked eye. Results can be used for better understanding how MOBA games works, as well as for developing more player-friendly games in terms of difficulty, and time one needs to learn how to play.

Authors (2)

Cite as

Full text

full text is not available in portal

full content of the article see on external site open in new tab

Keywords

ESPORTS, MOBA, PREDICTIVE ANALYTICS, DATA MINING

Details

Category:: Conference activity
Type:: publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
Title of issue:: Proceedings of the IADIS International Conference Interfaces and Human Computer Interaction strony 116 - 122
Language:: English
Publication year:: 2015
Bibliographic description:: Deja D., Myślak M.: TOPOLOGICAL CLUES FOR PREDICTING OUTCOMES OF MULTIPLAYER ONLINE BATTLE ARENA GAMES// Proceedings of the IADIS International Conference Interfaces and Human Computer Interaction/ ed. Katherine Blashki and Yingcai Xiao : Piet Kommers, Pedro Isaías and Heredina Fernandez Betancort, 2015, s.116-122
Verified by:: Gdańsk University of Technology