Find in Library

With the development of information technology and application of the Internet, People gradually entered the time of information overload from information scarcity. User satisfaction with recommender systems is related not only to how accurately the system recommends but also to how much it supports the user’s decision making. Novelty is one of the important metrics of customer satisfaction. There is an increasing realization in the Recommender Systems (RS) field that novelty is fundamental qualities of recommendation effectiveness and added-value. This paper combed research results about definition and algorithm of novel recommendation, and starting from the meaning of "novel", defined novelty of item in recommendation system. Experiment proved using the definition of novelty to recommend can effectively recognize the item that the user is familiar with and ensure certain accuracy.