Privacy-Preserving Big Data Analytics Using Homomorphic Encryption: A Comprehensive Evaluation in Healthcare Applications

This study addresses the critical challenge of preserving data privacy while conducting analytics on sensitive information using homomorphic encryption. Homomorphic encryption enables the analysis of encrypted data without compromising confidentiality, which is crucial in domains like healthcare and finance. This research evaluates the efficacy of homomorphic encryption in safeguarding data privacy while facilitating meaningful analysis. Through rigorous experimentation, this research assessed the performance of various machine learning models, including Logistic Regression and Naive Bayes, on both encrypted and unencrypted datasets. The findings indicate that while encryption introduces a slight reduction in accuracy, the overall performance remains high, demonstrating the viability of homomorphic encryption for privacy-preserving data analytics. This research provides valuable insights and strategies for balancing data privacy and analytical accuracy, paving the way for enhanced data privacy practices in real-world applications.