Privacy-Preserving Protocols for Eigenvector Computation

In this paper, we present a protocol for computing the dominant eigenvector of a collection of private data distributed across multiple parties, with the individual parties unwilling to share their data. Our proposed protocol is based on secure multiparty computation with a trusted third-party arbitrator who deals with data encrypted by the other parties using an additive homomorphic cryptosystem. We also augment the protocol with randomization to make it difficult, with a high probability, for any party to estimate properties of the data belonging to other parties from the intermediate steps. The previous approaches towards this problem were based on expensive QR decomposition of correlation matrices, we present an efficient algorithm using the power iteration method. We present an analysis of the correctness, security, efficiency of protocol and experiments over a prototype implementation.