Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 1

MINING ASSOCIATION RULES USING CHAOTIC PARTICLE SWARM OPTIMIZATION

Submitted by Prashanth P Ramcharan Teja K Harish @ Sivasankar V Jeeva @ Rajasekar R

Under the guidance of Dr S. Kanmani

Abstract:
Association rule mining is a technique in data mining to analyze and present strong rules discovered in databases. Particle Swarm Optimization (PSO) is a swarm-based intelligence algorithm which simulates the movement of a bird flock or a fish school. It optimizes a problem with given candidate keys with regard to any measure of quality. Chaotic Particle Swarm optimization is a variation from the normal PSO which searches for solutions chaotically and extensively in the feasible region to update the promising solution.

The traditional rule mining algorithms like Apriori, DHP and DIC have low computational efficiency. Also they do not help in determining proper threshold values for support and confidence, which seriously affect the quality of the mining. An alternate method of Genetic Algorithm, based on evolutionary concepts, employs operators like crossover and mutation, thus increasing the complexity of the computation.

We would like to propose a novel algorithm of chaotic particle swarm optimization for association rule mining in order to improve computational efficiency as well as convergence of the results.The project is to be implemented using Java, and three different datasets namely Lenses, Haberman and Car evaluation are taken up for the study.

You might also like