Masking Techniques for Confidential Data Protection in Privacy-Preserving Data Mining
S. Vijayarani, S. Sharmila, M. Lavanya · International Journal of Darshan Institute on Engineering Research and Emerging Technologies · 2023
Privacy-Preserving Data Mining (PPDM) develops algorithms for altering sensitive data.The private knowledge of a person, industry, or business organization remains private after the usage of data from the database.Data modification is one of the prominent privacy-preserving techniques used to alter confidential information available in the database and guarantees high privacy protection.In this paper, a new masking technique is proposed for hiding sensitive numerical attributes that are later analyzed using clustering algorithms, namely k-means, filtered clusters, and density-based clusters.The proposed technique is used to hide confidential numerical attributes.After modification, the proposed algorithm compares the original and the modified data and ensures that all the data items are altered or not.Experimental evaluation is illustrated using the employee dataset.The accuracy is calculated based on the comparison of the original and the modified data set in terms of data items found in the number of clusters.For every clustering technique, both the original and modified at a set are divided into two, three, four, and five, clusters.Based on the performance metrics, the k-means algorithm gives the best result compared to other algorithms.The results obtained from the proposed technique are compared with the existing approach.The experimental result indicates that the newly developed method is more efficient than the existing approaches.