This study examines behavioral profiles of different character types in EverQuest II, a popular massively multiplayer online role-playing game (MMORPG) developed by Sony Online Entertainment. The study uses the game's player activity data to construct behavioral profiles of online game players for the applications of normal behavior recognition and anomaly detection. The behavioral profiles give insight into not only what players do in the game to level up but also how they perform on different types of task and how they work with other players. The behavioral profiles this study reports provide insight into online game player behavior, and it provides information useful for choosing a character type suitable for one's objectives in playing the game. In this study, we develop a framework for automatic behavior profiling. The proposed framework consists of the following key components; 1) segmentation analysis of historical player behaviors, 2) behavior profiling of input users (new players whose behaviors we want to label, and 3) recommendation of tasks based on input users' objectives.