An Extensive Study on Data Anonymization Algorithms Based on K-Anonymity

International Conference on Materials, Alloys and Experimental Mechanics 2017

材料科学;金属学;机械制造

Simi, M.S.^1 ; Nayaki, K Sankara^2 ; Elayidom, M. Sudheep^3

Dept of Computer Science, Adishankara Institute of Engineering and Technology, Kalady, Kerala

683574, India^1

Dept of Information Technology, Adishankara Institute of Engineering and Technology, Kalady, Kerala

683574, India^2

Dept of Computer Science, School of Engineering, CUSAT, Kochi, Kerala

22, India^3

关键词: Generalization;    K-Anonymity;    Microdata;    Quasi identifiers;    Suppression;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/225/1/012279/pdf
DOI  :  10.1088/1757-899X/225/1/012279

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

For business and research oriented works engaging Data Analysis and Cloud services needing qualitative data, many organizations release huge microdata. It excludes an individual's explicit identity marks like name, address and comprises of specific information like DOB, Pin-code, sex, marital status, which can be combined with other public data to recognize a person. This implication attack can be manipulated to acquire any sensitive information from social network platform, thereby putting the privacy of a person in grave danger. To prevent such attacks by modifying microdata, K-anonymization is used. With potentially increasing data, the effective method to anonymize it stands challenging. After series of trails and systematic comparison, in this paper, we propose three best algorithms along with its efficiency and effectiveness. Studies help researchers to identify the relationship between the values of k, degree of anonymization, choosing a quasi-identifier and focus on execution time.