Bit Manipulation Based Data Transformation Approach for Privacy Preservation in Data Mining

Organizations makes use of a variety of data mining techniques to discover hidden patterns in large databases. The data mining process can disclose some confidential information. Privacy is becoming an increasingly important concern in many data mining applications. Privacy-preserving techniques provide a new approach to solve this problem. It drives home legitimate data mining results without exposing the original data values, ensuring both privacy and accuracy. In this research, we present a unique Binary Bit manipulation strategy that focuses on manipulating integers in a binary manner to boost data utility while maintaining anonymity. Following binary encoding, integers are subjected to specific bit manipulation techniques. We include noise into binary bit manipulation, a novel method targeted at enhancing data privacy. The performance of the proposed approach has been evaluated using the K-means clustering algorithm. The proposed Binary Bit Manipulation techniques have been assessed for misclassification error rate against the Heart-disease dataset from the UCI repository.