Novelty and Its Evaluation in Recommender System

the growth of the Internet has made it much more difficult to effectively extract useful information from all the available online information. Recommender systems tend to suggest items which are both unexpected and useful to users. These items are not only profitable to the retailers but also surprisingly suitable to consumers' tastes. Due to the imbalance in observed data for popular and tail items, existing methods fail to give results with novelty. User satisfaction with recommender systems is related to how accurately the system recommends and how much it supports the user's decision making. Because being accurate is not enough, because the quality of recommendation is also important. The Recommended quality is defined to meet or exceed expectations of customer, and, the user will not only be satisfied with some of the Repeated and "not so interesting" recommendation. We studied the papers to understand how novelty can be introduced in recommendations to improve user satisfaction.