Comparisons of Different Clustering Algorithms for Privacy of Online Social Media Network

Abstract

Online Social Networks (OSN) connect billions of users with direct consequences to offline activities. As of late, OSNs have seen critical development and accepted much consideration in exploration. These Networks have consistently been a significant part of daily life; however, since an ever-increasing number of individuals are associated with the internet, their online partners satisfy an undeniably significant job. Users use online applications where they end up sharing lots of information including personal data. Hence this data including huge information is money-making for data owners and they share data with third parties like advertisers who are ethical users. This data also consists of personal data. As data can be used for unethical purposes by unauthorized users which leads to different attacks and illegal use of data and personal information. It is fundamental to first anonymize users’ data before imparting it to any of the third parties like advisers. Anonymization preserves data privacy. However, anonymization prompts data loss, which by implication influences the data utility. Balancing data privacy and utility of information is an open research issue. Different clustering algorithms can be applied for anonymizing social network data. Comparison of clustering algorithms leads to the best algorithm to provide k-anonymity to social media networks.