Privacy-Preserving Data Mining Techniques for Big Data Analytics in Healthcare Using Differential Privacy

Main Article Content

Vijay Kumar Meena

Abstract

The increasing digitization of healthcare systems has generated massive datasets containing sensitive patient information. While data mining and analytics can extract valuable insights for clinical decision-making, privacy concerns and regulatory requirements such as HIPAA and GDPR restrict access to raw health data. Differential privacy (DP) has emerged as a rigorous framework for preserving privacy while enabling statistical analysis and machine learning on sensitive datasets. This paper examines the application of differential privacy in healthcare big data analytics, focusing on association rule mining, predictive modeling, and statistical aggregation. We present a privacy-preserving framework for healthcare data mining, incorporating Laplace and Gaussian noise mechanisms, and evaluate performance on public health datasets. Experimental results demonstrate that differential privacy maintains high utility with minimal information leakage, achieving comparable accuracy to non-private methods while protecting patient confidentiality. Case studies on disease pattern analysis and treatment outcome prediction highlight practical applications.

Article Details

Section
Articles