Towards Outsourced Privacy-preserving Multiparty DBSCAN

This fast abstract proposes a privacy-preserving quantum-secure protocol for DBSCAN clustering algorithm. It allows multiple parties to jointly compute clusters without revealing their individual datasets to each other. We show how to outsource the DBSCAN computation on encrypted data to a cloud service provider securely such that the parties do not need to perform the expensive clustering computation locally. We also discuss some future research directions at the end of the paper.