Privacy preserving k-nearest neighbor classification over encrypted database in outsourced cloud environments

Abstract

To utilize the cost-saving advantages of the cloud computing paradigm, individuals and enterprises increasingly resort to outsource their databases and data operations to cloud servers. However such solutions come with the risk of violating the privacy of users. To protect privacy, the outsourced databases are usually encrypted, making it difficult to run queries and other data mining tasks without decrypting the data first. Conventional encryption methods are either incapable of supporting such operations or computationally expensive to do so. In this paper, we aim to efficiently support computations over encrypted cloud databases, particularly focusing on privacy preserving k-nearest neigh..