Role of Mathematical Modelling and Learning Techniques for Privacy Preservation: A Systematic Review

Abstract

Recent advancements in disruptive technology have gathered extensive data from various sectors, including healthcare, transportation, retail, market prediction, surveillance, finance, and telecommunication. Quantifiable information has been obtained from the massive amount of shared consumer data to achieve valuable insights into each of these sectors. Moreover, augmented mobile spectrum usage has paved the way for tracing consumers' activities and interests via numerous working prototype systems and commerce apps. For protecting the sensitive information and maintaining its integrity of stored data, it has upheld the necessity to mathematical modeling paradigm design and learning frameworks for protection of user data where all the storage and operations are supported out without unveiling any details. For protecting these details containing one's confidential evidences, classical privacy-preserving based on other techniques and methods developed over the few decades. However, classical models/ techniques have severe problems with data preservation of individual information. This paper provides a comprehensive review of the existing mathematical modelling and learning techniques and framework for privacy preservation along with significant challenges of privacypreserving biometric schemes and highlight the future research pathways in preserving biometric schemes are discussed. © 2023. All Rights Reserved.