Protecting Privacy When Sharing and Releasing Data with Multiple Records per Person.

Area of Science:

• Computer Science
• Data Privacy
• Information Security

Background:

• Existing data privacy methods often underestimate risks in datasets where individuals have multiple records.
• Current policies typically assume a one-to-one relationship between individuals and records, which is insufficient for complex datasets.

Purpose of the Study:

• To propose novel measures for assessing privacy disclosure risks in datasets with multiple records per person.
• To develop an efficient computational procedure for anonymizing such data while preserving utility.

Main Methods:

• Introduced two new privacy disclosure measures: one for individual-record risk and another for sensitive-attribute risk.
• Demonstrated that these measures generalize existing metrics like k-anonymity and l-diversity.
• Developed an integrated computational procedure for anonymization, incorporating privacy and data quality measures.

Main Results:

• The proposed measures provide a more accurate assessment of disclosure risks in multi-record scenarios.
• The developed anonymization procedure effectively protects privacy while maintaining data quality.
• Experimental results on real-world data confirm the superiority of the proposed approach over existing techniques.

Conclusions:

• The novel privacy disclosure measures are essential for accurately evaluating risks in multi-record datasets.
• The integrated anonymization approach offers a robust solution for secure data sharing in research and analytics.
• This work advances the field of data privacy by addressing the limitations of traditional methods in complex data environments.