Association Rules Optimization using Particle Swarm Optimization Algorithm with Mutation
Mayank Agrawal1, Manuj Mishra2, Shiv Pratap Singh Kushwah3
1Mayank Agrawal, Research Scholar, ITM Universe, Gwalior, India.
2Manuj Mishra, Asst. Prof., Department of CSE, ITM Universe, Gwalior, India.
3Shiv Pratap Singh Kushwah, Asst. Prof., Department of CSE, ITM Universe, Gwalior, India.
Manuscript received on February 14, 2015. | Revised Manuscript received on February 25, 2015. | Manuscript published on March 05, 2015. | PP: 141-144 | Volume-5 Issue-1, March 2015. | Retrieval Number: A2552035115/2015©BEIESP
Open Access | Ethics and Policies | Cite
©The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: In data mining, Association rule mining is one of the popular and simple method to find the frequent item sets from a large dataset. While generating frequent item sets from a large dataset using association rule mining, computer takes too much time. This can be improved by using particle swarm optimization algorithm (PSO). PSO algorithm is population based heuristic search technique used for solving different NP-complete problems. The basic drawback with PSO algorithm is getting trapped with local optima. So in this work, particle swarm optimization algorithm with mutation operator is used to generate high quality association rules for finding frequent item sets from large data sets. The mutation operator is used after the update phase of PSO algorithm in this work. In general the rule generated by association rule mining technique do not consider the negative occurrences of attributes in them, but by using PSO algorithm over these rules the system can predict the rules which contains negative attributes.
Keywords: Particle Swarm Optimization (PSO), Mutation, Association rule, Support, Confidence, Frequent item set, Data mining.