Multi-objective genetic algorithm based approach for optimizing fuzzy sequential patterns

Author

Kaya, Mehmet ; Alhaji, R.

Author_Institution

Dept. of Comput. Eng., Firat Univ., Elazig, Turkey

fYear

2004

fDate

15-17 Nov. 2004

Firstpage

396

Lastpage

400

Abstract

This work introduces the optimized sequential pattern problem and presents a novel approach to find such patterns. All the methods described in the literature to optimize association rules employ a single objective measure, such as optimized confidence or optimized support. In this study, we propose a novel multiobjective genetic algorithm (GA) based optimization method for optimizing quantitative sequential patterns. The objective measures of are support, confidence and a parameter related to the total number of fuzzy sets in the sequence. Experimental results on a synthetic database demonstrate the effectiveness and applicability of the proposed method.

Keywords

data mining; fuzzy set theory; genetic algorithms; pattern classification; very large databases; association rule optimization; fuzzy sequential pattern optimization; fuzzy set sequence; multiobjective genetic algorithm; sequential pattern mining; synthetic database; Association rules; Computer science; Data mining; Data security; Databases; Fuzzy sets; Genetic algorithms; Optimization methods; Pattern analysis; Web mining;

fLanguage

English

Publisher

ieee

Conference_Titel

Tools with Artificial Intelligence, 2004. ICTAI 2004. 16th IEEE International Conference on

ISSN

1082-3409

Print_ISBN

0-7695-2236-X

Type

conf

DOI

10.1109/ICTAI.2004.91

Filename

1374214