Facilitating fuzzy association rules mining by using multi-objective genetic algorithms for automated clustering
- 23 April 2004
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
We propose an automated clustering method based on multiobjective genetic algorithms (GA); the aim of this method is to automatically cluster values of a given quantitative attribute to obtain large number of large itemsets in low duration (time). We compare the proposed multi-objective GA-based approach with CURE-based approach. In addition to the autonomous specification of fuzzy sets, experimental results showed that the proposed automated clustering exhibits good performance over CURE-based approach in terms of runtime as well as the number of large itemsets and interesting association rules.Keywords
This publication has 8 references indexed in Scilit:
- Clustering association rulesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Fuzzy summaries in database miningPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Efficient Automated Mining of Fuzzy Association RulesLecture Notes in Computer Science, 2002
- Cure: an efficient clustering algorithm for large databasesInformation Systems, 2001
- Mining association rules from quantitative dataIntelligent Data Analysis, 1999
- Mining fuzzy association rulesPublished by Association for Computing Machinery (ACM) ,1997
- Association rules over interval dataPublished by Association for Computing Machinery (ACM) ,1997
- Mining quantitative association rules in large relational tablesACM SIGMOD Record, 1996