Performance Analysis of Genetic Algorithm for Mining Association Rules
Association rule (AR) mining is a data mining task that attempts to discover interesting patterns or relationships between data in large databases. Genetic algorithm (GA) based on evolution principles has found its strong base in mining ARs. This paper analyzes the performance of GA in Mining ARs effectively based on the variations and modification in GA parameters. The recent works in the past seven years for mining association rules using genetic algorithm is considered for the analysis. Genetic algorithm has proved to generate more accurate results when compared to other formal methods available. The fitness function, crossover rate, and mutation rate parameters are proven to be the primary parameters involved in implementation of genetic algorithm. Variations and modifications introduced in primary GA parameters are found to have greater impact in increasing the accuracy of the system moderately. The speedup of the system is found to increase when the selection and fitness function are altered.
Keywords: Association rule, Genetic Algorithm, GA parameters, Accuracy, Speedup
Download Full-Text
ABOUT THE AUTHORS
K.Indira
Received her M.E. degree in 2005 from Department of Computer Science and Engineering, FEAT, Annamalai University, Chidambaram. She had been working as the Head of the Department of Computer Science for the past 12 years in Theivanai Ammal College for Women, Tamil Nadu, India from 1998 to 2007 and E.S. College of Engineering and Technology, Affiliated to Anna University , Chennai, India. Currently she is working towards her Ph.D degree in Evolutionary Algorithms applied for data Mining. Her areas of interest are Data Mining, Artificial Intelligence and Evolutionary Computing.
S. Kanmani
Received her B.E and M.E in Computer Science and Engineering from Bharathiyar University and Ph.D in Anna University, Chennai. She had been the faculty of Department of Computer Science and Engineering, Pondicherry Engineering College from 1992 onwards. Presently she is Professor in the Department of Information Technology, Pondicherry Engineering College. Her research interests are Software Engineering, Software testing, Object oriented system, and Data Mining. She is Member of Computer Society of India, ISTE and Institute of Engineers, India. She has published about 50 papers in various international conferences and journals.
K.Indira
Received her M.E. degree in 2005 from Department of Computer Science and Engineering, FEAT, Annamalai University, Chidambaram. She had been working as the Head of the Department of Computer Science for the past 12 years in Theivanai Ammal College for Women, Tamil Nadu, India from 1998 to 2007 and E.S. College of Engineering and Technology, Affiliated to Anna University , Chennai, India. Currently she is working towards her Ph.D degree in Evolutionary Algorithms applied for data Mining. Her areas of interest are Data Mining, Artificial Intelligence and Evolutionary Computing.
S. Kanmani
Received her B.E and M.E in Computer Science and Engineering from Bharathiyar University and Ph.D in Anna University, Chennai. She had been the faculty of Department of Computer Science and Engineering, Pondicherry Engineering College from 1992 onwards. Presently she is Professor in the Department of Information Technology, Pondicherry Engineering College. Her research interests are Software Engineering, Software testing, Object oriented system, and Data Mining. She is Member of Computer Society of India, ISTE and Institute of Engineers, India. She has published about 50 papers in various international conferences and journals.
