A Survey of Privacy-Preserving Methods Across Horizontally Partitioned Data

Abstract

Data mining can extract important knowledge from large datacollections, but sometimes these collections are split among various parties.Data warehousing, bringing data from multiple sources under a singleauthority, increases risk of privacy violations. Furthermore, privacyconcerns may prevent the parties from directly sharing even some meta-data.Distributed data mining and processing provide a means to address this issue,particularly if queries are processed in a way that avoids the disclosure ofany information beyond the final result. This chapter describes methods tomine horizontally partitioned data without violating privacy and discusseshow to use the data mining results in a privacy-preserving way. The methodsdescribed here incorporate cryptographic techniques to minimize theinformation shared, while adding as little as possible overhead to the miningand processing task.