The application of visualization techniques in recommendation systems

Abstract

The personalized recommendation system has been widely and maturely applied to various domains from social network to items recommendation such as videos, music, movies, books and online shopping. In the meantime, the information visualization technology based on big data has a substantial development. Considering the common based on big data, this paper discussed the connections of graph visualization and recommendation system from a new perspective. We propose three combinative points: Firstly, the Visual Analytics of social network can be combined with the recommendation computing; secondly, the graph drawing algorithms can be combined with the terminal presentation of recommended results. On the account of the evolutional nature of recommendation, the dynamic graphs would be the key area; thirdly, user interaction with the recommendation system can use the graph interaction strategies to achieve a more accurate and understandable result. In this paper, we discuss the visual techniques which can applied to recommendation system from a more general perspective. We also discuss some research challenges in the future.