Hiding Sensitive Medical Data Using Simple and Pre-Large Rain Optimization Algorithm through Data Removal for E-Health System

Privacy has become a significant factor of e-Health system in the area of data mining termed as Privacy preserving data mining (PPDM) as it can uncover underlying rules and hide sensitive data for data sanitization. Various algorithms and heuristics have been studied to hide sensitive data using transaction removal. However, they are facing challenges to attain the reasonable side effects. Thus, rain optimization algorithm (ROA) based sensitive data hiding techniques is proposed in this paper. Using this algorithm, suitable transactions to be removed are selected. Besides, in this work, ROA based two frameworks are designed for data sanitization that are simple ROA to remove transaction (sROA2RT) and pre-large ROA to remove transaction (pROA2RT). In this algorithm, fitness is evaluated based on four side effects such as hiding failure, artificial cost, missing cost and dissimilarity of database. The proposed frameworks are evaluated using three e-Health datasets. Compared to previous frameworks, the proposed frameworks attain reasonable side effects.