Privacy-preserving data publishing based on sanitized probability matrix using transactional graph for improving the security in medical environment

Abstract

Privacy is an important factor that hospitals should preserve while publishing data that involve sensitive information of individuals. Research seeks to find solutions for releasing data to the public without infringing the confidentiality of personal information. Sanitizing data makes them safe for publishing while maintaining essential information. The current paper proposes a transactional graph-based approach using adaptive probability sanitization for performing sanitization in medical environment. The method first generates transactional graphs for each user transaction, and on this basis, estimates a convergence and deviation measure for each item. Based on these values, we compute a probability matrix. We generate the ontology on the basis of this matrix. Experimental results reveal that the proposed approach generates efficient sanitization and data publication results.