Fuzzy taxonomy, quantitative database and mining generalized association rules

Wang Shitong, Fu Lai Korris Chung, Shen Hongbin

Research output: Journal article publicationJournal articleAcademic researchpeer-review

15 Citations (Scopus)

Abstract

Mining association rules from databases is still a hot topic in data mining community in recent years. Due to the existence of multiple levels of abstraction (i.e, taxonomic structures) among the attributes of the databases, several algorithms were proposed to mine generalized Boolean association rules upon all levels of presumed crisp taxonomic structures. However, fuzzy taxonomic structures may be more suitable in many practical applications. In [9], the authors proposed an approach to mine generalized Boolean association rules with such fuzzy taxonomic structures. The main contribution of this paper is to extend their idea to mine generalized association rules from quantitative databases with fuzzy taxonomic structures. A new fuzzy taxonomic quantitative database model is presented, and the experimental results on realistic databases are demonstrated to validate this new model.
Original languageEnglish
Pages (from-to)207-217
Number of pages11
JournalIntelligent Data Analysis
Volume9
Issue number2
Publication statusPublished - 1 Dec 2005

Keywords

  • association rules
  • data mining
  • fuzzy taxonomic quantitative database

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this