A data mining approach in opponent modeling

A data mining approach in opponent modeling

In offline opponent modeling, large datasets can be utilized as training data to model the opponent. In the Coach competition of RoboCup Soccer, offline opponent modeling can be adopted to train the coach learn about the opponent's behavior patterns. Data-mining techniques, particularly decisio...

Full description

Saved in:
Bibliographic Details
Main Authors: Bulos, Remedios De Dios, Dulalia, Conirose, Go, Peggy Sharon L., Tan, Pamela Vianne C., Uy, Ma Zaide Ilene O.
Format: text
Published: Animo Repository 2005
Subjects:
Data mining
Soccer—Defense
Soccer—Training
Computer Sciences
Online Access:https://animorepository.dlsu.edu.ph/faculty_research/498
https://animorepository.dlsu.edu.ph/context/faculty_research/article/1497/type/native/viewcontent
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: De La Salle University
Description
Summary:In offline opponent modeling, large datasets can be utilized as training data to model the opponent. In the Coach competition of RoboCup Soccer, offline opponent modeling can be adopted to train the coach learn about the opponent's behavior patterns. Data-mining techniques, particularly decision-tree construction can be applied in identifying interesting behavior patterns of the opponent. This research explores the use of the decision-tree algorithm C4.5 to generate classification rules that will embody the offensive and defensive strategies (plans) of the coach against its opponent(s). To achieve this objective, the SimSoccer Coach system is built. © Springer-Verlag Berlin Heidelberg 2005.

Similar Items