Augmenting Wargame AI with Data Mining Technology

John Rushing, John Tiller, Steve Tanner, and Drew McDowell

Data mining methods can be used to augment a traditional wargame AI in a variety of ways. They can be used in situational analysis to identify high-level structures such as front lines, weakly defended targets, inadequate supply lines or other items of interest. They may also be used to model and predict the likely responses of players, and learn viable decision making strategies by example. These capabilities can both speed the development of an AI and improve its quality by reducing the amount of explicit programming of rules or scripts that would be required, by allowing the decision making to occur at a higher level, and by allowing the wargame AI to adapt over time and learn from previous experiences. Wargames have a long history of use as tools for training military personnel, and are likely to be used even more in the future. Modern computer wargames are sophisticated software tools that can provide detailed data about the actions taken by each player and the underlying situations within the game scenarios that lead up to these actions. Data mining tools can be used to aid in the analysis of this data in order to improve the feedback provided to the player, revealing general tendencies or weaknesses that could be exploited by an opponent.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.