Integration of PSO and Clustering algorithms for privacy preserving data mining

Abstract

Privacy Preserving Data Mining (PPDM) currently has become an important research area. There are some issues and problems related to PPDM have been identified. Information loss occurs when the original of data is modified to keep the privacy of those data. Effects of PPDM also cause the level of data quality to become lower. The aim of this research is to minimize information loss and increase the accuracy of mining result while maintaining the privacy level of data. A randomization approach based on optimization and clustering algorithms is proposed in order to minimize the information loss and improve the accuracy of data clustering quality for PPDM results. There are three main objectives for this research which is to perform data pre-processing on data through the normalization process and k-Anonymity algorithm. The second objective is to minimize data loss and increase the accuracy of data mining result using Particle Swarm Optimization and clustering algorithms. The third objective is to evaluate and benchmark the performance measurement based privacy level and data quality of enhanced PPDM. Diabetes dataset is used in this research and all instances are a numerical value. The outcome of this research is the privacy level of the dataset was increased while the information loss is minimized. The experimental results also show that the accuracy of data clustering quality can be preserved with using PSO algorithm.