Privacy Preserving Combinatorial Function for Multi-Partitioned Data Sets

Abstract

preserve the electronic fitness records of data mining applications, Data perturbation is used. It is a form of privacy- preserving data mining. An improved amount and range of information stored in databases has direct to an enhancement in the desire for ranked and best match queries. Such queries are mainly applicable when dealing with privacy- sensitive information. To facilitate privacy preservation in data mining or machine learning algorithms over horizontally partitioned or vertically partitioned data, many protocols have been proposed using SMC and various secure building blocks. In this work, we plan to present an effective and efficient cluster based privacy preserving data perturbation technique to mine Multi-partitioned data sets that comprises of both vertical and horizontal data sets which is current demand of e- business and e-commerce data mining environment. To evaluate a trade of between data privacy and transparency of individual's data, data perturbation technique is presented with validation and authentication. In multi-partitioned data distribution, data perturbation raised ambiguity between vertical and horizontal partitions of the data. To overcome the ambiguity, we plan to introduce divisive k-neighbor clusters for multi-partitioned data sets. The performance is evaluated with bench data sets obtained from popular e-business / e- commerce sites. (Amazon, e-bay etc.,). Keywords-partitioned data sets, Privacy preservation, Data perturbation technique.