Preservation of Privacy using Multidimensional K-Anonymity Method for Non-Relational Data
Abhijit J. Patankar1, Kotrappa Sirbi2, Kshama V. Kulhalli3
1Mr. Abhijit J. Patankar, Research Scholar, Vishvesvaraya Technological University, Belagavi (Karnataka), India.
2Dr. Kotrappa Sirbi, Professor, Department of CSE, KLE’s Dr. M.S.S.C.E.T, Belgaum (Karnataka), India.
3Dr. Kshama V. Kulhalli, Principal, D. Y. Patil C.E.T, Kolhapur (Maharashtra), India.
Manuscript received on 20 September 2019 | Revised Manuscript received on 06 October 2019 | Manuscript Published on 11 October 2019 | PP: 544-547 | Volume-8 Issue-2S10 September 2019 | Retrieval Number: B10960982S1019/2019©BEIESP | DOI: 10.35940/ijrte.B1096.0982S1019
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Mining of huge data having complexity is a challenging issue also maintaining Privacy of data is also equally important ,sometimes there is a need to release data for use of researchers or for the purpose of gaining knowledge or earn money this release of data includes releas e of all attributes of personal data. when this type of data like Insurance record data, Medical diagnosis data, funding scheme data is release even if we remove sensitive attribute like Name for hiding personal details still data re-identification is possible by linking public data like voters data with these released data and by linking the quasi identifiers we are able to get sensitive information about person like critical disease, financial position etc. by applying k–Anonymization using multiple dimensions of attributes we are able to hide these sensitive attributes by generalising and suppressing the Quasi identifiers so that when linking with public database is done no records are re-identified, also we obtained results for quality measures for anonymisation and observed that the value of k once we start increase after some threshold anonymity starts decreasing so there is a need to choose proper value of k on non-relational data.
Keywords: re-identification.
Scope of the Article: Data Base Management System