A Review of Privacy Protection Data Utility Mining Using Perturbation
D. S. Eunice Little Dani, R. Shalinib · 2024
The method of addressing the automated finding of information inside the database is known as data mining. The capacity to keep and record personal data on service users and people has rapidly increased as a result of recent breakthroughs in hardware technology. The danger of disclosing data to third parties during data mining operations has grown due to data sets that contain sensitive personal information. Data mining that protects privacy strives to maintain the value of the data without violating respect for the privacy of private or confidential data. An extension of maintaining privacy when mining data that takes care of both quantity and utility is referred to as maintaining your privacy Utility mining. Because perceptive data must be kept private for analytical purposes, privacy preservation in data mining with high-utility sets has grown in importance and difficulty. Several industries, including a basket of goods, trading, analysis of click-streams, medical surveys, and computational biology, generate a large number of data using utility-oriented patterns and analytics by taking into account precise information. Perturbation is a technique that adheres to the data owner’s privacy regulations while changing the database’s contents within certain restrictions. In this research, the Frequency Count supply an excellent method for privacy preservation that is utilized for fast perturbation and provides a simple technique that we employ to conceal all highly sensitive item sets. This also cleanses the transaction database and employs a unique perturbation technique, distinct structures for item set tables, and a sorted list of sensitive items. Our paper is focused on the effectiveness shown by contrasting the Fast Perturbation Using the Frequency Count method with the Using Tree Structures and Tables for the Rapid Perturbation approach, it can be extended and implemented using different criteria.