Preserving Privacy in Data Mining using hybrid of Auto-Associative Neural Network and Particle Swarm Optimization: An application for bankruptcy prediction in banks

Data mining has emerged as a significant technology for gaining knowledge from vast quantities of business data, financial data, networked data and medical data. The goal of data mining is approaches are to develop generalized knowledge rather than identify specific information against specific individual. There has been growing concern that use of this technology is violating individual privacy. This is opening new challenges in the area of Privacy Preserving in Data Mining (PPDM). Privacy regulations and other privacy concerns may prevent data owners from sharing information for performing data analysis. To achieve a solution to this problem, data owners must design a strategy that meets privacy requirements and guarantees valid data mining results. This paper discusses the possibility of using neural network and Particle Swarm Optimization (PSO) algorithm for Preserving Privacy in Data Mining. The above task is carried on five benchmark data sets and four bankruptcy data sets. This paper presents methods where by the privacy and secrecy of a bank related sensitive data is taken care of on one hand and the resulting dataset is mined without a considerable loss in accuracy obtained in models. Multi Layer Perceptron, decision tree J48 and Logistic Regression are used as classifiers for illustration purpose. This paper gives the experimental results how the bankruptcy prediction can be done for various bank datasets while protecting the sensitive information of the banks. The results have been compared with the other methods used for Preserving Privacy in Data Mining namely random