Within the DataCamo ( Data Camouflaging ) project, we investigate techniques that enable data mining on private data while respecting the privacy of data owners. In particular, we focus on distributed data mining systems where private data are either partitioned and stored in autonomous entities, or fully distributed in the sense that each data tuple is held by its individual owner. Our work is motivated by the challenges posed by the rapidly growing needs of sharing information through the Internet, whose success depends on protecting privacy in communications between business and customers, and between different businesses.