Anonymization Technique For Privacy Preservation In Social Networks
Kalpana Chavhan, Praveen S. Challagidad · 2021 5th International Conference on Electrical, Electronics, Communication, Computer Technologies and Optimization Techniques (ICEECCOT) · 2021
Anonymization is a valuable privacy protection strategy utilized to protect highly sensitive data from third-party access in a variety of technology fields, including data mining, cloud storage and big data. Data protection against all threats is becoming increasingly important as the importance and quantity of data in today’s world grows. The primary goal of this study is to present concise review of data anonymization and differential privacy approaches. In order to address problem of privacy protection, a new k-anonymous solution has been proposed which differs from the traditional k-anonymous solution. In this paper, new algorithm for attaining k-anonymity by more effective clustering is proposed. The majority of clustering algorithms need more computation to process data; however, if the detected initial centers that are consistent with the data setup, then it will obtain a better cluster array. Our research proposes a Dissimilarity Treebased method for identifying a better initial centroid and a somewhat more accurate cluster with less computation time, as well as NCP. According to graphical findings, the anonymised dataset’s total information loss is about 20% lower on average than that of other approaches. It’s also capable of handling categorical and numerical characteristics.