IFAA: An Improved Field Anonymity Algorithm for Efficient Big Data Privacy Preservation

Privacy protection for depression patients in mobile internet environments is critical, as the screening and assessment data typically exhibit characteristics of being scattered, multi-field, and highly sensitive. Existing solutions often struggle to balance data utility with privacy security for such specific data types. In this paper, based on the design of a Privacy Classification Protection (PCP) framework, we propose the Improved Field Anonymity Algorithm (IFAA) to enhance privacy protection for these multi-field datasets. While the PCP framework accommodates various techniques like encryption and differential privacy, IFAA specifically addresses the scenario where traditional anonymization is cost-effective but insecure against inference attacks. Guided by a rigorous mathematical formalization of re-identification risk, IFAA utilizes reversible non-linear square mapping and field name encryption to dynamically adjust data precision. Experimental results demonstrate that compared to full AES encryption, IFAA reduces processing time by over 95% while maintaining high resistance against distribution-based inference attacks. This approach achieves an optimal balance between security and utility, providing a lightweight solution particularly suitable for the real-time processing of depression screening data.