A Secure and Efficient kNN Classification Algorithm Using Encrypted Index Search and Yao’s Garbled Circuit over Encrypted Databases

• Published in: Future Data and Security Engineering pp. 21 - 38
• Main Authors: Kim, Hyeong-Jin, Shin, Jae-Hwan, Chang, Jae-Woo
• Format: Book Chapter
• Language: English
• Published: Cham Springer International Publishing 2018
• Series: Lecture Notes in Computer Science
• Subjects: Cloud computing; Data privacy; Database outsourcing; Hiding data access pattern; kNN classification algorithm; Query protection
• DOI: 10.1007/978-3-030-03192-3_3
• ISBN: 9783030031916; 3030031918
• ISSN: 0302-9743

Database outsourcing has been popular according to the development of cloud computing. Databases need to be encrypted before being outsourced to the cloud so that they can be protected from adversaries. However, the existing kNN classification scheme over encrypted databases in the cloud suffers from high computation overhead. So we proposed a secure and efficient kNN classification algorithm using encrypted index search and Yao’s garbled circuit over encrypted databases. Our algorithm not only preserves data privacy, query privacy, and data access pattern. We show that our algorithm achieves about 17x better performance on classification time than the existing scheme, while preserving high security level.