Understanding Microaggregation- A technique of Statistical Disclosure Control for Privacy Preserving and Data Publishing in Inter-Cloud

In today's world organizations outsource their data to cloud to enjoy number of advantages such as location independent data storage, ubiquitous data access and on demand high quality service. Data is collected by many of independent sources as micro data and published for decision making, predictions and to improve research. Outsourcing the data has an important side effect- the privacy of the individuals whose data are being collected and analyzed is at risk. It becomes the responsibility of the data curator i.e., the cloud service provider to preserve the privacy. Statistical Disclosure control or Statistical Disclosure Limitation seeks to alter the original microdata so that the utility of original microdata and modified microdata are at least same. Microaggregation is one of Statistical Disclosure Control methods used to protect the individual identification present in the microdata. This method works by partitioning the microdata into groups called cluster of k records and then replaces the record in each group with the centroid of the group. This paper defines microdata, provides the overview of various existing microaggregation methods for Statistical Disclosure Control. The paper also provides a novel model of microaggregation method in inter-cloud infrastructure.