Thesis Topic Details

Topic ID:
Privacy-Preserving Recommendations Systems: Are users' choices biased by the recommendations systems themselves? The chicken or the Egg problem!
Guillaume Jourjon
Research Area:
Privacy, Recommender systems
Associated Staff
Aruna Seneviratne
Topic Details
R & D
Group Suitable:
The growth of Internet commerce has stimulated the use of Recommender Systems (RS), and in particular collaborative Filtering algorithms (CF). However, CFs are usually based on matching users according to their tastes. Although this connection has been used to recommend items, it has been overlooked as a potential source of user matching and possible information leakage, even when users are privacy-aware and hide some of their sensitive information.
We aim to exploit the collaborative filtering techniques as a powerful way to extract similarities between users' profiles. First, we will apply CF to detect users' interests, and hence select appropriate items to advertise on users profiles. In other words, online advertising can be targeted based on a user's interests.
Secondly, we will show how these techniques, when used for a malicious purpose, can be effective to profile and track users and to perform large scale efficient spam or social engineering campaigns. These two applications are on the way to our main contribution aiming to provide a privacy-preserving search and recommendations.
The student will work closely with researchers in the NICTA Networks Research Group. The project will be primarily based at the NICTA ATP laboratory, where the work environment is a group comprised of junior and senior research staff, and PhD students. IT support and resources will be provided by NICTA.
Past Student Reports
No Reports Available. Contact the supervisor for more information.

Check out all available reports in the CSE Thesis Report Library.

NOTE: only current CSE students can login to view and select reports to download.