Towards security and efficiency in distributed fitness recommendation with hashing

In the era of intelligent fitness technologies, the need for personalized training solutions that adapt to individual users’ physiological states, performance patterns, and exercise goals has grown considerably. This trend highlights the importance of advanced fitness recommendation systems capable of providing precise, adaptive, and data-driven training guidance. However, achieving such personalization requires continuous processing of sensitive physiological and behavioral data, which introduces critical security risks, including unauthorized access, data manipulation, and system vulnerabilities. To address these challenges, this paper presents a novel edge-enabled personalized fitness recommendation framework that integrates individualized training optimization with strong security guarantees. By leveraging the distributed and proximate processing capabilities of edge computing, the system minimizes communication exposure, reduces latency, and enables sensitive data to remain within secure edge environments or directly on the user’s device. This architectural design substantially mitigates security threats that commonly arise in centralized cloud-based processing. Comprehensive experimental evaluations demonstrate that the proposed framework not only delivers highly accurate and tailored fitness recommendations but also provides enhanced system robustness and data security compared with traditional cloud-centric approaches. The results indicate that edge computing offers a promising pathway toward secure, adaptive, and scalable personalized training technologies that support users’ performance improvement without compromising data integrity or system safety.