An Algorithm for Mining Association Rules Based on Improved Genetic Algorithm and its Application
Authors:
- Hong GuoHGHong Guo
- Ya ZhouYZYa Zhou
- 1 October 2009
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 117-120
Abstract
Genetic algorithm is an important algorithm of association rule mining. However, there is some issues that genetic algorithm easy to lead prematuring convergence and into the plight of local optimum, or convergence too much time and consume a large amount of time to search. For resolving this issues, the paper improves the algorithm through adopting an adaptive mutation rate and improving the methods of individual choice, and the improved genetic algorithm that applies to the mining association rules. The simulating experiments show that the improved genetic algorithm reduces the cost of computing, and improve the efficiency of association rule mining.Keywords
This publication has 3 references indexed in Scilit:
- Special issue on granular computing and data miningInternational Journal of Intelligent Systems, 2004
- Mining frequent patterns without candidate generation2000
- An effective hash-based algorithm for mining association rules1995