PRIVACY RISK IN RECOMMENDED SYSTEM

Abstract

In many on-line applications, the range of content that is offered to users is so wide that a requirement for automatic recommender systems arises. Such systems will give a personalized selection of relevant things to users. In practice, this may facilitate people realize fun movies, boost sales through targeted advertisements, or facilitate social network users meet new friends. To produce correct personalized recommendations, recommender systems depend on detailed personal information on the preferences of users. Ratings, consumption histories and personal profiles are examples. Recommender systems are useful, but the privacy risks associated in aggregation and process personal information are typically underestimated or neglected. Many users are not sufficiently aware if and the way a lot of their information is collected, if such information is sold-out to third parties or how securely it is saved and for how long. This paper aims to provide insight into privacy in recommender systems. First, we shall discuss different varieties of existing recommender systems. Second, an overview of the data that is employed in recommender systems is given. Third, I analyze the associated risks to information privacy. Finally, relevant research areas for privacy-protection techniques and their relevancy to recommender systems are mentioned.