This research aims to classify simulated players in a multi-agent game for the best position and role they may play in according to their abilities. Three approaches were investigated for this purpose, C4.5 classification algorithm, backpropagation neural network and radial basis function network. This work depends on a video game that uses 28 attributes to distinguish every player from another. The applied techniques examine the abilities of the players and classify them in one of four major positions/roles. The three approaches were compared by applying them on a data set collected manually from the selected game. The results obtained show promising capability of classification based on agents attributes.
Key words: Artificial neural networks, decision trees, multi-agent game, classification.
Copyright © 2023 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0