Semantics Accommodated K-Anonymization (SAKA) Technique for Assorted Data

Abstract

Techniques approaching k-anonymity for shielding the mini data privacy over data mining. Mini gatherings and abstraction are two conventional techniques for deploying the k-anonymity method. However, the above two methodologies lead to major errors over the anonymity of combined small data gathering. To approach this problem, we propose an efficient and new anonymity method SAKA that accommodates extra semantics than abstraction and mini data gatherings that can handle combined small data combinations. SAKA is the abbreviation for Semantics Accommodated K-Anonymity technique implied for anonymization of assorted data. The concept of SAKA is to merge the mean assorted vector of arithmetic data with abstraction values of classified data as a grouping centroid. It uses its epitome of tuples along with its equivalent clusters. Here we propose efficient algorithms to anonymize assorted data. An empirical result proves that SAKA can anonymize the assorted mini data persuasively and the algorithm implemented will provide good rapport between the quality of data and effectiveness of the algorithm thus it correlates anonymization algorithms with anonymous data.