An Enhanced Design for Anonymization in Social Networks

Objectives: Our primary objective is to safeguard the privacy of the users by anonymizing the sensitive data shared by the users in Social networking sites. Methods/Statistical Analysis: The user's friends are grouped dynamically into various categories such as best friends, normal friends and casual friends based on their closeness with the user. The most private sensitive data is solely disclosed to best friends, while the sensitive data is anonymized using generalization and revealed to the normal friends. The last category of friends is exposed to only the least private information using another level of generalization. Findings: Currently, Social networking sites have become the rapid and preferred way of communicating with each other to share information. On one hand the options and benefits expand constantly, while on the other data privacy risks and sensitivity issues accumulate, eventually the privacy of the user is at grave, our work addresses this issue. The proposed design is resilient to Sybil attacks, where it restricts the revelation of the sensitive data of the user by anonymizing the shared data among various categories of friends. Application/Improvement: The performance of the system is enhanced by restricting the access to the sensitive data among various categories of friends based on their closeness degree.