MANIFOLD ACCESSIBLE SEGMENTATION BASED NARRATIVE DESIGN IN CLOUD COMPUTING FOR DATA PUBLISHING

N Jayachandra, H Hemachitra, & Angayarkanni
2016 unpublished
Distributing Data is a simple and cost-effective way for data sharing, but the confidentiality threat is a key anxiety in data publishing. One of the problems in such practices is how to trade-off between data utility and privacy protection. The problem seriously depreciates when the published data are used to do cluster analysis. The paper establishes a narrative progression known to be "Manifold Accessible Segmentation (MAS)". The technique utilizes to accomplish conserves enhanced data
more » ... enhanced data effectiveness than simplification and can be used for relationship confession safeguard. The method allocates the approval of assorted data utility metrics for dissimilar information removal tasks. Another important benefit of segmentation is that it can hold high-dimensional data we design a new method integrating sampling and generalization to implement the model. Our workload experiments confirm that MAS preserves better utility than generalization. Our experiments also demonstrate that MAS can be used to prevent membership disclosure.
fatcat:wbcjf27elvc7dbifjdgafrj6d4