View Item 

Application of tree mining to matching of knowledge structures of decision tree type

Access Status
Open access
Authors
Hadzic, Fedja
Dillon, Tharam S.
Date
2007
Type
Book Chapter

Metadata
Show full item record
Citation
Hadzic, Fedja and Dillon, Tharam S. 2007. Application of tree mining to matching of knowledge structures of decision tree type, in Meersman, R. and Tari, Z. and Herrero, P. (ed), On the move to meaningful internet systems 2007: OTM 2007 workshops, pp. 1319-1328. Berlin / Heidelberg: Springer.
Source Title
On the move to meaningful internet systems 2007: OTM 2007 workshops
DOI
ISBN
Faculty
Curtin Business School
School of Information Systems
School
Centre for Extended Enterprises and Business Intelligence
Remarks

The original publication is available at http://www.springerlink.com

URI
Collection
Abstract

Abstract: Matching of knowledge structures is generally important for scientific knowledge management, e-commerce, enterprise application integration, etc. With the desire of knowledge sharing and reuse in these fields, matching commonly occurs among different organizations on the knowledge describing the same domain. In this paper we propose a knowledge matching method which makes use of our previously developed tree mining algorithms for extracting frequent subtrees from a tree structured database. Example decision trees obtained from real world domains are used for experimentation purposes whereby some important issues that arise when extracting shared knowledge through tree mining are discussed. The potential of applying tree mining algorithms for automatic discovery of common knowledge structures is demonstrated.