A fuzzy data mining algorithm for finding sequential patterns

  • Authors:
  • Yi-Chung Hu;Ruey-Shun Chen;Gwo-Hshiung Tzeng;Jia-Hourng Shieh

  • Affiliations:
  • Institute of Information Management, National Chiao Tung University, Hsinchu 300, Taiwun, R.O.C.;Institute of Information Management, National Chiao Tung University, Hsinchu 300, Taiwun, R.O.C.;Institute of Management of Technology, National Chiao Tung University, Hsinchu 300, Taiwan, R.O.C.;Institute of Management of Technology, National Chiao Tung University, Hsinchu 300, Taiwan, R.O.C.

  • Venue:
  • International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Since fuzzy knowledge representation can facilitate interaction between an expert system and its users, the effective construction of a fuzzy knowledge base is important. Fuzzy sequential patterns described by natural language are one type of fuzzy knowledge representation, and can thus be helpful in building a prototype fuzzy knowledge base. We define that a fuzzy sequence is an ordered list of frequent fuzzy grids, and the length of a fuzzy sequence is the number of frequent fuzzy grids in the frequent fuzzy sequence. Frequent fuzzy grids and frequent fuzzy sequences can be determined by comparing individual fuzzy supports with the user-specified minimum fuzzy support. A fuzzy sequential pattern is just a frequent fuzzy sequence, but it is not contained in any other frequent fuzzy sequence. In this paper, an effective algorithm called the Fuzzy Grids Bused Sequential Patterns Mining Algorithm (FGBSPMA) is proposed to generate fuzzy sequential patterns. A numerical example is used to show an analysis of the user visit to websites, demonstrating the usefulness of the proposed algorithm.