M-Privacy for Collaborative Data Publishing By Using Heuristic Approach

According to the survey Maintaining and preserving privacy has become more significant problem. We consider the collaborative data publishing problem for anonymizing horizontally partitioned data at multiple data providers. We consider a new type of “insider attack” by colluding data providers who may use their own data records (a subset of the overall data) to infer the data records contributed by other data providers. For M-privacy several anonymization techniques have been used such as bucketization, generalization, perturbation which does not prevent privacy and fail to maintain a privacy constraint and results in loss of information. So we consider the collaborative data publishing for anonymizing horizontally partitioned data at multiple data provider. First, we provide the notion of m-privacy, which give guarantees that the anonymized data satisfies a given privacy constraint against any intruder attack. Then we present heuristic algorithms with effective pruning strategies and adaptive ordering techniques for efficiently checking m-privacy for a set of records i.e to breach the privacy. Here intruder are also detected which try to breach the privacy. This technique shows the better utility and efficiency than the previous techniques.We develop a truthful and efficient M-privacy for collaborative data publishing by using pruning strategy and providing them anonymized data in case of emergency.