Authors:
Danilo Verhaert
;
Majid Nateghizad
and
Zekeriya Erkin
Affiliation:
Cyber Security Group, Department of Intelligent Systems, Delft University of Technology and The Netherlands
Keyword(s):
Recommender System, Privacy-preserving, Homomorphic Encryption, Multi-party Computation, Comparison Protocol.
Related
Ontology
Subjects/Areas/Topics:
Applied Cryptography
;
Cryptographic Techniques and Key Management
;
Data and Application Security and Privacy
;
Data Engineering
;
Databases and Data Security
;
Information and Systems Security
;
Privacy
;
Privacy Enhancing Technologies
Abstract:
The significant growth of medical data has necessitated the development of secure health-care recommender systems to assist people with their health-being effectively. Unfortunately, there is still a considerable gap between the performance of secure recommender systems and normal versions. In this work, we develop a privacy-preserving health-care recommendation algorithm to reduce that gap. The main strength of our contribution lies in providing a highly efficient solution, while the sensitive medical data are kept confidential. Our studies show that the runtime of our protocol is 81,5% faster than the existing implementation for small bit-lengths, and even more so for large bit-lengths.