PRIVACY PRESERVING SOCIAL MEDIA DATA PUBLISHING FOR PERSONALIZED RANKING BASED RECOMMENDATION

A personalised recommendation is essential to assisting users in finding relevant information. To mine user preferences, it frequently uses a sizable collection of user data, particularly information about users' online behaviour (such as tagging, rating, and check-ins on social media). While private information (such as gender) may frequently be deduced from user activity data, exposing such data exposes users to inference attacks. In this research, we introduced Priv Rank, a continuous privacy- preserving social media data publishing architecture that allows for tailored ranking-based recommendations while defending users against inference assaults. Its main concept is to continuously obfuscate user activity data in order to limit the privacy leakage of user-specified private data within a certain data distortion budget. This also limits the ranking loss brought on by the data obfuscation.process in order to preserve the utility of the data for enabling recommendations.