Fine-grained k-anonymity for privacy preserving in cloud

摘要

Data sensitive information is a crucial concern of every individual. Hospitals lag their trust in privacy to take up the newest technologies of cloud like Information-as-a-service, storage-as-a-service, to deploy their patient's data for better health management. Intensive study is being undertaken to run-over the shortcomings of data privacy for the published information as well as the publisher, One amongst the methods is privacy by statistics using data mining techniques such as k-anonymity. The fundamental technique of k-anonymity is to anonymize sensitive information of an individual person published that could not be determined from at least (k - 1) instances. The best way to attain k-anonymity is by grouping similar records into a cluster by choosing the best seed value to balance utility and privacy in the published data. This paper proposes a Fine-grained k-anonymity algorithm which uses a systematic procedure of seed selection. The proposed method exhibits a minimum information loss than existing clustering algorithms.