Achieving Privacy Preservation in Data Mining Using Hybrid Transformation and Machine Learning Technique

Abstract

Extraction of meaningful patterns and knowledge from a vast number of datasets is known as data mining. Due to the availability of vast amounts of data and the need to turn that data into meaningful information, data mining has received a lot of attention in the IT sector in recent years. This crucial data can be applied to a number of fields, including fraud detection, market analysis, customer retention, manufacturing controls, and scientific research. We need privacy-preserving procedures when this data is moved from one location to another since different sorts of hackers or attackers may leak our private information to the public. In our study, we use a hybrid transformation strategy to achieve high degree privacy preservation. The transformation approach allows us to alter the given data objects' position, size, shape, and orientation. We employ the k means clustering technique to carry out machine learning operations. We use a dataset (the patient dataset) for experimental purposes, and WEKA is used for all operations. Comparing our work to the prior work, it provides the maximum level of anonymity.