A data perturbation method to preserve privacy using fuzzy rules

Abstract

Data mining methods analyze the patterns found in data, irrespective of the confidential information of an individual. It has led to raise privacy concerns about confidential data. Different methods are inhibited in data mining to protect these data. Privacy preserving data mining plays a major role in protecting confidential data. The paper focuses on data perturbation method to preserve confidential data present in the real-world datasets. These identified confidential data are perturbed using fuzzy membership function (FMF) and obtains fuzzy data. The mining utility such as classification and clustering methods are used. The accuracy is determined and compared between an original data and fuzzy data. The results shown in the paper proves the proposed method is efficient in preserving confidential data.