Definição de métricas para análise de dados em jogos do tipo MMORPG.

Abstract

The market for computer games is in a level never seen before, estimating to move US$99.6 billion in 2016. Among the various categories of games, MMORPGs (Massive Multiplayer Online Role-Playing Game) attract hundreds of millions of players, who interact through their characters simultaneously in the same universe. As a result of the large number of players, an unimaginable mass of data is generated, containing the actions of each player in the game. Due to the size of the data, and the number of distinct attributes recorded, extract useful information can be challenging. These characteristics of the logs of those games make data mining as a viable option to extract information, as this is already prepared to handle large volumes of data, and has its own algorithms for knowledge discovery in datasets. The proposed final paper aims to identify which characteristics in MMORPG’s game logs can be used for analysis and classification of player profiles. For this, a data mining algorithm will be used, checking which characteristics bring better results to players’ categorization. The paper also try to identify characteristics in common among different game logs, looking for maintain generic results.