TreeCluster: clustering results of keyword search over databases

  • Authors:

  • Zhaohui Peng;Jun Zhang;Shan Wang;Lu Qin

  • Affiliations:

  • School of Information, Renmin University of China, Beijing, P.R. China;School of Information, Renmin University of China, Beijing, P.R. China;School of Information, Renmin University of China, Beijing, P.R. China;School of Information, Renmin University of China, Beijing, P.R. China

  • Venue:
  • WAIM '06 Proceedings of the 7th international conference on Advances in Web-Age Information Management
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

A critical challenge in keyword search over relational data- bases (KSORD) is to improve its result presentation to facilitate users' quick browsing through search results. An effective method is to organize the results into clusters. However, traditional clustering method is not applicable to KSORD search results. In this paper, we propose a novel clustering method named TreeCluster. In the first step, we use labels to represent schema information of each result tree and reformulate the clustering problem as a problem of judging whether labeled trees are isomorphic. In the second step, we rank user keywords according to their frequencies in databases, and further partition the large clusters based on keyword nodes. Furthermore, we give each cluster a readable description, and present the description and each result graphically to help users understand the results more easily. Experimental results verify our method's effectiveness and efficiency.