Abstract
Devising methods to publish social network data in a form that affords utility without compromising privacy remains a longstanding challenge, while many existing methods based on k-anonymity algorithms on social networks may result in nontrivial utility loss without analyzing the social network topological structure and without considering the attributes of sparse distribution. Toward this objective, we explore the impact of the attributes of sparse distribution on data utility. Firstly, we propose a new utility metric that emphasizes network structure distortion and attribute value loss. Furthermore, we design and implement a differentiated k-anonymity l-diversity social network anonymity algorithm, which seeks to protect users’ privacy in social networks and increase the usability of the published anonymized data. Its key idea is that it divides a node into two child nodes and only anonymizes sensitive values to satisfy anonymity requirements. The evaluation results show that our method can effectively improve the data utility as compared to generalized anonymizing algorithms.
Highlights
Nowadays, partly driven by many Web 2.0 applications, more and more social network data are publicly available and analyzed in one way or another, as the social network data has significant application value for commercial and research purposes [1]
Importantly, privacy strategies ignore the fact that the influence of each individual is different in a social network
We proposed a proper utility model UL(G,G’), and designed a graph anonymity
Summary
Partly driven by many Web 2.0 applications, more and more social network data are publicly available and analyzed in one way or another, as the social network data has significant application value for commercial and research purposes [1]. The social network data often have privacy information of individuals. It has become a major concern to prevent individual privacy disclosure when publishing the social network data. The k-anonymity l-diversity models aim to sanitize the published graph, eventually leading to data usability reduction for published social network data. The tradeoff between the individual’s privacy security and data utility while publishing the social network data has become a major concern [2,3,4]. The social networks are modelled as graphs in which nodes and edges correspond to social entities and social links between them, respectively, while users’ attributes and graph structures are composed of the corresponding social network data [5]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
