Contrast Motif Discovery In Minecraft

Understanding event sequences is a vital side of recreation analytics, since it's relevant to many player modeling questions. This paper introduces a way for analyzing event sequences by detecting contrasting motifs; the aim is to discover subsequences which might be considerably more comparable to at least one set of sequences vs. other sets. In comparison with current methods, our approach is scalable and capable of dealing with long event sequences. We utilized our proposed sequence mining method to research player habits in Minecraft , a multiplayer online sport that supports many forms of player collaboration. As a sandbox game, it provides players with a large amount of flexibility in deciding how to complete tasks; this lack of objective-orientation makes the issue of analyzing Minecraft event sequences more difficult than event sequences from more structured games. Using our method, we had been able to find contrast motifs for a lot of player actions, despite variability in how totally different gamers accomplished the identical tasks. Moreover, we explored how the level of participant collaboration affects the distinction motifs. Although this paper focuses on functions inside Minecraft, our software, which we now have made publicly out there along with our dataset, can be used on any set of recreation event sequences.