The research of personalized recommendation system based on "HowNet"

"HowNet"^[1] is a commonsense knowledge database. The description object of "HowNet" is the concept represented by Chinese and English words. It is used to reveal the relationship between the concepts and the relationship between attributes of concepts. It has a rich of lexical semantic knowledge and world knowledge, and those knowledge is the basic resource of Chinese information processing. This paper adopts the words semantic similarity calculation based on "HowNet" to expand the user interest model and document feature vector space model in the personalized recommendation system which we build. In the end, we verify the effectiveness of the proposed method by comparison experiment.