Title: Privacy-preserving machine learning and robust cryptography in TensorFlow for empowering secure IoT analytics

Authors: Kritika Purohit; Surendra Yadav

Addresses: Career Point University, Kota (Rajasthan), India ' Vivekananda Global University Jaipur (Rajasthan), India

Abstract: Integrating privacy-preserving machine learning with robust cryptographic techniques, particularly leveraging the TensorFlow platform, holds significant promise in extracting valuable insights from encrypted data while addressing privacy concerns. This collaboration bridges disciplines like encryption, machine learning, distributed systems, and high-performance computing. Machine learning's user-friendly interface enables the application of cutting-edge cryptographic methods, even for non-experts, ensuring security across the entirety of TensorFlow applications, from input data to models and code. This study introduces an innovative cryptographic framework within TensorFlow, demonstrating enhanced reliability and up to 90% improved efficiency over conventional methods. These findings underscore its potential to revolutionise privacy-conscious data utilisation, especially in the context of internet of things (IoT) applications.

Keywords: privacy-preserving; machine learning; robust cryptography; TensorFlow; Secure IoT; analytics; internet of things; data privacy.

DOI: 10.1504/IJCVR.2026.155542

International Journal of Computational Vision and Robotics, 2026 Vol.17 No.2, pp.199 - 215

Received: 26 Aug 2023
Accepted: 15 Nov 2023
Published online: 05 Aug 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article