MULTI PARTY PRIVACY PRESERVING DECISION TREE FOR HORIZONTALLY PARTITIONED DATA

Abstract

Data mining is the extraction of hidden predictive information from large databases and also a powerful new technology with great potential to analyze important information in their data warehouses. In this research work, we discussed methods for distributed privacy-preserving mining, and the methods for handling horizontally partitioned data. The primary contribution of this work is to propose a multi party privacy preserving decision tree for horizontally partitioned data by using ID3 algorithm. This has particular relevance to privacy-sensitive searches, particularly top-k queries, and meshes well with privacy policies. There remain many open problems in developing secure solutions based on efficient non secure query processing algorithms