Document Details

Document Type : Thesis 
Document Title :
TOWARDS A MORE PRIVATE DATA MINING APPROACH FOR RECOMMENDER SYSTEMS
نحو المزيد من الخصوصية في طرق تنقيب البيانات المستخدمة في آنظمة التوصيات
 
Subject : Faculty of Computing and Information Technology 
Document Language : Arabic 
Abstract : With the use of new generations of Global Positioning System GPS-enabled smartphones, site-based recommendation systems appear on the market, giving users appropriate suggestions while moving places to visit based on user tastes. These systems are very popular, and their benefits are very interesting to users, but there are many privacy problems which have become a major concern in the research community. In location-based recommendation systems (LbRSs), the user is constrained to build queries that depend on the actual location to ask for the closer points of interest (POIs). An external attacker can analyze these queries or track the actual location of the LbRS user to reveal personal information. Consequently, ensuring high privacy protection (which is including location privacy and query privacy) is fundamental. In this thesis, we provide an extensive survey of location-based recommendation systems and analyze them from a user privacy view. We then propose a model that guarantees high privacy protection for LbRS users. The proposed model has three main components: The first component (selector_D) uses a new location privacy protection approach, namely, the smart dummy selection (SDS) approach. The SDS approach generates a strong dummy location that has high resistance versus a semantic location attack. The second component 〖(encryptor〗_ID) uses an encryption-based approach that guarantee a high level of query privacy versus a sampling query attack. The last component (constructor_Q) constructs the protected query that is sent to the LbRS server. Our proposed system is supported by a checkpoint technique to ensure a high availability quality attribute. Our proposed system yields competitive results compared to similar systems under various privacy and performance metrics. 
Supervisor : Dr. Nermin Hamza and ‏ 
Thesis Type : Master Thesis 
Publishing Year : 1441 AH
2019 AD 
Co-Supervisor : Dr. Reem Alotaibi 
Added Date : Wednesday, December 18, 2019 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
تهاني صالح النزاويAl-Nazzawi, Tahani SalehResearcherMaster 

Files

File NameTypeDescription
 45706.pdf pdf 

Back To Researches Page