Outsourced privacy-preserving C4.5 decision tree algorithm over horizontally and vertically partitioned dataset among multiple parties

Abstract

Many companies want to share data for data-mining tasks. However, privacy and security concerns have become a bottleneck in the data-sharing field. The secure multiparty computation (SMC)-based privacy-preserving data mining has emerged as a solution to this problem. However, there is heavy computation cost at user side in traditional SMC solutions. This study introduces an outsourcing method to reduce the computation cost of the user side. We also preserve the privacy of the shared databy proposing an outsourced privacy-preserving C4.5 algorithm over horizontally and vertically partitioned data for multiple parties based on the outsourced privacy preserving weighted average protocol (OPPWAP) and outsourced secure set intersection protocol (OSSIP). Consequently, we have found that our method can achieve a result same the original C4.5 decision tree algorithm while preserving data privacy. Furthermore, we also implement the proposed protocols and the algorithms. It shows that a sublinear relationship exists between the computational cost of the user side and the number of participating parties.