SSP Group Meeting
March 19th, Wednesday, 2-3pm
Department of Artificial Intelligence, 80 South Bridge, Room F13


 

Learning Method for Knowledge-Based Planning System

Yoshimoto Ikkai, ikkai@ise.eng.osaka-u.ac.jp
Department of Information Systems Engineering
Faculty of Engineering - Osaka University

Knowledge-based planning system is one of many useful approaches. The solutions generated by the system depends on a quality of knowledge, so Knowledge acquisition is the largest bottleneck of a knowledge-based planning system.

We propose an inductive learning method of knowledge for status selection method, which is one of knowledge-based planning systems. Through the application of the proposed method to a job shop problem, effectiveness and evaluation of the proposed learning method are shown.