Research on Recommendation Algorithms in Web of Things System

With the development of Web2.0, human beings are no longer simple units in the Internet world. Web of Things (WoT) system makes devices and humans connect to each other, and gradually becomes the leading force of information creation, transmission and acquisition. However, the widely network applications and information circulation in recent years have resulted in information overload, which means the browsers cannot correctly get their needed information due to the complexity and university of the information. Recommendation system is a new way for information acquisition, which studies the user preference model, detects their potential demand and recommend their interested information. It is a promising way to solve the information overload problem. This paper firstly introduces the necessity of recommendation mechanism in WoT system. And then presents the traditional recommendation algorithms: content-based recommendation, collaborative filtering recommendation and the hybrid recommendation. Finally we propose an item-based recommendation algorithm in WoT system to recommend devices based on the item evaluation.