DATA LEAKAGE PREVENTION IN AI-POWERED RECOMMENDATION SYSTEMS USING HOMOMORPHIC ENCRYPTION
Subjects/Theme:
Data Leakage Prevention, Homomorphic Encryption, AI Recommendation Systems, Privacy Preservation, Secure Computation, Machine Learning Security, Encrypted Data ProcessingDescription
Security and Privacy in AI Systems,
Edited By: Dr. Sunita Chaudhary, Dr. Joydeb Patra
ISBN (978-81-685212-9-2)
AI-powered recommendation systems are integral to modern digital platforms, including e-commerce, streaming services, and social media. These systems rely heavily on user data, making them vulnerable to data leakage and privacy breaches. Traditional encryption methods protect data at rest and in transit but fail to secure it during computation. Homomorphic Encryption (HE) emerges as a promising solution by enabling computations on encrypted data without decryption. This paper explores the integration of homomorphic encryption in AI-based recommendation systems to prevent data leakage. It analyzes the architecture, advantages, challenges, and performance trade-offs of HE-based models. Experimental results demonstrate that while HE introduces computational overhead, it significantly enhances privacy preservation. The study concludes with future directions for optimizing HE in large-scale recommendation systems.
