TITLE: Learning Taxonomic Relation by Casebased Reasoning

PRESENTER: Prof. Ken Satoh

AFFILIATION: Hokkaido University, Japan

DATE: Thursday, 14 December 2000 TIME: 12:00 - 13:00

PLACE: Meeting Room 1 Ground Floor Samuels Building

* All Welcome *

Enter through gate 11 in Botany Street for parking

For more information on the location see URL http://www.cse.unsw.edu.au/~aishare/finding/

ABSTRACT:

In this paper, we propose a learning method of minimal casebase to represent taxonomic relation in a tree-structured concept hierarchy. We firstly propose case-based taxonomic reasoning and show an upper bound of necessary positive cases and negative cases to represent a relation. Then, we give an learning method of a minimal casebase with sampling and membership queries. We analyze this learning method by sample complexity and query complexity in the framework of PAC learning.

BIOGRAPHY OF SPEAKER:

Ken Satoh graduated from University of Tokyo in 1981 and joined Fujitsu Laboratories Ltd. He was sent to Institute for New Generation Computer Technology (ICOT) in 1987 and returned to Fujitsu Laboratories in 1992. Since 1995, he has been an associate professor of Hokkaido University. He is interested in logical foundations of AI.

Seminar Convenor:

Bernhard Hengst
Tel: +61-2-9385-3988,
E-mail: bernhardh@cse.unsw.edu.au
School of Computer Science & Engineering, UNSW.