Privacy Preserving Random Decision Tree Classification Over Horizontally and Vertically Partitioned Data

This paper presents a framework for constructing Random Decision Trees on horizontally and vertically partitioned dataset. It is assumed that data is distributed between two parties, such that for general benefits both are willing to model RDT classifier on whole dataset, but for privacy concerns, they refuse to share the original datasets. To this end, we propose algorithms based on secure weighted average protocol to securely compute the desired criteria in constructing the classification scheme.