Balancing quality and efficiency in private clustering with affinity propagation

Abstract

In many machine learning applications, training data consists of sensitive information from multiple sources. Privacy-preserving machine learning using secure computation enables multiple parties to compute on their joint data without disclosing their inputs to each other. In this work, we focus on clustering, an unsupervised machine learning technique that partitions data into groups. Previous works on privacy-preserving clustering often leak information and focus on the k-means algorithm, which provides only limited clustering quality and flexibility. Additionally, the number of clusters k must be known in advance. We analyze several prominent clustering algorithms’ capabilities and their compatibility with secure computation techniques to create an efficient, fully privacy-preserving clustering implementation superior to k-means. We find affinity propagation to be the most promising candidate and securely implement it using various multi-party computation techniques. Privacy-preserving affinity propagation does not require any input parameters and consists of operations that are relatively efficient with secure computation. We consider passive security as well as active security with an honest and dishonest majority. We offer the first comparison of privacy-preserving clustering between these scenarios, enabling an understanding of the exact trade-offs between them. Based on the clustering quality and the computational and communication costs, privacy-preserving affinity propagation offers a good trade-off between quality and efficiency for practical privacy-preserving clustering.