| Autor: |
Bushra, S. Nikkath, A., Chandra Sekar |
| Předmět: |
|
| Zdroj: |
International Review on Computers & Software; Nov2013, Vol. 8 Issue 11, p2746-2754, 9p |
| Abstrakt: |
A massive computation power and storage capacity is offered by the cloud computing to the users to use the applications without infrastructure investment. For operational convenience and for economic advantages, a lot of privacy sensitive applications like health services are built on cloud. Typically, the data sets in these applications are anonymized to assert the privacy of data owners but the privacy requirements can be violated when new data join over time. In this paper, a set of records are given as input and some attributes are chosen as quasi-identifiers for data anonymization and the set of records are clustered based on k-means clustering technique. The k-anonymity constraint and information loss is checked for each cluster and the cluster is modified based on the k-anonymity constraint. Thereafter, new records are added after choosing the quasi-identifiers and after anonymize it. The new records get group with any cluster based on the k-anonymity constraint. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
