Privacy preservation in digital platforms through anonymization technique

Anonymization is a helpful privacy protection approach that is applied in a range of technological domains, namely data gathering, cloud storage, and big data, to secure extremely sensitive data from access by third parties. As both the quantity and the quality of data produced in the modern world continue to expand, the need to safeguard it against any and all hazards is growing more pressing. The primary objective of this study is to offer a concise introduction to various methods of data privacy protection and differentiated privacy protection. A new k-anonymous solution that is distinct from the standard k-anonymous approach has indeed been proposed in order to solve the concern regarding the protection of personal information. In this article, a novel clustering technique is proposed as a means of achieving k-anonymity through increased efficiency. The vast majority of clustering methods require additional processing in order to analyses data. Nevertheless, the software will generate an improved clusters array in the event that the found initial centres are consistent with the data arrangement. Our study has developed a method, which is centered on the Dissimilarity Tree, for locating a more reliable initial centroid, as well as a cluster that is marginally more accurate while requiring a smaller amount of processing time and NCP. According to the graphical representation of the results, the overall information loss caused by the anonymized dataset is around 20% less on average than that caused by other procedures. In addition to this, it is successful with numerical as well as category attributes.