Cellular automata and Hilditch thinning for extraction of user paths in online games
Ruck Thawonmas, Masayasu Hirano, Masayoshi Kurashige · 2006
To keep online games interesting to their players, it is important to detect players' behaviors. In this paper, in order to understand players' movement in online games, we propose a method for extraction of players' paths from their trails. In the proposed method, locations in a given map that are frequently visited are first intensified by cellular automata, and paths are then derived by the Hilditch thinning algorithm. Players' trails from an experimental online game, where three typical game missions are available, are used for performance evaluation. For performance evaluation, the proposed method is compared with a method using the median filter and the Hilditch thinning algorithm, a typical recipe in the area of image processing. According to the comparison results, the proposed method significantly outperforms its counter part in all cases, except the case with limited movement patterns.