Personalized suggestions are important to
help users find relevant information. It often depends on huge
collection of user data, especially users’ online activity (e.g.,
liking/commenting/sharing) on social media, thereto user
interests. Publishing such user activity makes inference
attacks easy on the users, as private data (e.g., contact
details) are often easily gathered from the users’ activity
data. during this module, we proposed PrivacyRank, an
adjustable and always protecting privacy on social media data
publishing framework , which protects users against frequent
attacks while giving personal ranking based
recommendations. Its main idea is to continuously blur user
activity data like user-specified private data is minimized
under a given data budget, which matches round the ranking
loss suffer from the knowledge blurring process
so on preserve the usage of the info for enabling
suggestions. a true world evaluation on both synthetic and
real-world datasets displays that our model can provide
effective and continuous protection against to the info given
by the user, while still conserving the usage of the blurred
data for private ranking based suggestion. Compared to other
approaches, Privacy Rank achieves both better privacy
protection and a far better usage altogether the rank based
suggestions use cases we tested.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.