The exponential growth of databases containing personal information has rendered the task of extracting high quality information from collections of such databases very important. This task is hindered by the security concerns that arise, due to the confidentiality of the data records, and the reluctance of the organizations to disclose their data. This paper proposes a clustering algorithmic scheme that ensures privacy and confidentiality of the data without compromising the effectiveness of the clustering algorithm nor imposing high communication costs.
展开▼
