A Multidimensional Anonymization method to protect online released user’s data from different types of attacks and validation of released large dataset in public domain

Abstract

When  users are filling their information online to the different agencies most of the users need to submit their own personal and sensitive information which need to keep confidential, user’s need to submit their personal details to medical agencies, insurance policies, voting information, registration of vehicle etc. these agencies will generally release this confidential information submitted online by different users for earning money or for research use or any other official purpose, even these organizations try to keep confidentiality of user’s data by removing user’s Name,UID No.,PAN No. etc. Which can easily identify individual still by liking other publically datasets with users submitted data one can easily obtained individual users sensitive information which must be keep confidential, Here I am proposing methodology to prevent Linking, Background knowledge and Homogeneity attacks on user’s data and perform validation of publically released large dataset by applying Various anonymization methods on publically available large datasets.