ONLINE SKILL LEVEL CLASSIFICATION OF REAL-TIME STRATEGY GAME PLAYERS

摘要

This paper introduces a novel approach for measuring a realtime strategy (RTS) game player's skill level while the game is played. The method scores a player based on the time distance between strategic decisions made by the player during a single game session. The results of this work demonstrate that skilled players score within a specific range regardless of the strategy executed. Our approach can classify a player as skilled or unskilled for the StarCraft: Brood War game and the Spring Engine Balanced Annihilation game.