Exploring the Influence of Contexts for Mobile Recommendation
Jun Zeng, Feng Li, Yinghua Li, Junhao Wen, Yingbo Wu · International Journal of Web Services Research · 2017
With the rapid development of mobile internet, it is difficult to obtain high-quality recommendation in such a complicated mobile environment, just depending on traditional user-item binary information. How to use multiple contexts to generate satisfying recommendation has been a hot topic in some fields like e-commerce, tourism and news. Context aware recommender system (CARS) imports contexts into recommender to generate ubiquitous and personalized recommendation. In this paper, the basic information of CARS, such as the definition of context, the process of CARS and evaluation are introduced carefully. In order to explore whether contexts have a great influence on recommendation or not, the authors conduct experiments on real datasets. Experimental results show recommender that incorporates contexts significantly improves performance over the traditional recommender. Finally, State of the art about CARS is detailed.