Privacy Preserving Data Mining With Abridge Time Using Vertical Partition Decision Tree
Sonal Yadav, Vivek Tiwari, Basant Tiwari · 2016
Security and privacy is an important concern during the analysis of the sensitive and private datasets. Data mining is used for the analysis of these datasets by classification, clustering, association rules etc. During the analysis of these datasets privacy should be maintained. Privacy preservation is a technique of analyzing the dataset with privacy such that external user can t access the conclusion of the analysis. Here in this paper a new decision tree based algorithm is proposed for the privacy of the datasets. The idea is to first divide the datasets vertically into a number of parties where each of the party calculates the gain of each of the attribute and send to the central authority where the gain from all the parties are compared to find the attribute with highest gain. The proposed technique implemented here provides reqired data and also provides less computational time and error rate.