Similarity measure models and algorithms for hierarchical cases

Publication Type:
Journal Article
Citation:
Expert Systems with Applications, 2011, 38 (12), pp. 15049 - 15056
Issue Date:
2011-11-01
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2011001013.pdf190.2 kB
Adobe PDF
Many business situations such as events, products and services, are often described in a hierarchical structure. When we use case-based reasoning (CBR) techniques to support business decision-making, we require a hierarchical-CBR technique which can effectively compare and measure similarity between two hierarchical cases. This study first defines hierarchical case trees (HC-trees) and discusses related features. It then develops a similarity evaluation model which takes into account all the information on nodes' structures, concepts, weights, and values in order to comprehensively compare two hierarchical case trees. A similarity measure algorithm is proposed which includes a node concept correspondence degree computation algorithm and a maximum correspondence tree mapping construction algorithm, for HC-trees. We provide two illustrative examples to demonstrate the effectiveness of the proposed hierarchical case similarity evaluation model and algorithms, and possible applications in CBR systems. © 2011 Elsevier Ltd. All rights reserved.
Please use this identifier to cite or link to this item: