Anonymization Based on Nested Clustering for Privacy Preservation in Data Mining

Privacy Preservation in data mining protects the data from revealing unauthorized extraction of information. Data Anonymization techniques implement this by modifying the data, so that the original values cannot be acquired easily. Perturbation techniques are variedly used which will greatly affect the quality of data, since there is a trade-off between privacy preservation and information loss which will subsequently affect the result of data mining. The method that is proposed in this paper is based on nested clustering of data and perturbation on each cluster. The size of clusters is kept optimal to reduce the information loss. The paper explains the methodology, implementation and results of nested clustering. Various metrics are also provided to explicate that this method overcomes the disadvantages of other perturbation methods.