A Sensitive Attribute based Clustering Method for kanonymization

TL;DR

This paper introduces a clustering-based anonymization method that groups data by sensitive attributes to enhance privacy while minimizing information loss in medical data sharing.

Contribution

It proposes a novel sensitive attribute-based clustering approach for k-anonymization that improves data utility and reduces information loss compared to existing methods.

Findings

Better information loss metrics

Reduced execution time

Effective privacy preservation

Abstract

In medical organizations large amount of personal data are collected and analyzed by the data miner or researcher, for further perusal. However, the data collected may contain sensitive information such as specific disease of a patient and should be kept confidential. Hence, the analysis of such data must ensure due checks that ensure protection against threats to the individual privacy. In this context, greater emphasis has now been given to the privacy preservation algorithms in data mining research. One of the approaches is anonymization approach that is able to protect private information; however, valuable information can be lost. Therefore, the main challenge is how to minimize the information loss during an anonymization process. The proposed method is grouping similar data together based on sensitive attribute and then anonymizes them. Our experimental results show the proposed method offers better outcomes with respect to information loss and execution time.