With the arrival of the big data era, it has become a spot for research to evaluate currency of network information so far. This paper proposes a model of network information currency evaluation based on Web semantic extraction method taking Web news as object of study. The author elaborates the method, technology and main functions on every layer of the model in detail, which have been used or completed, and focus on how to extract semantic information efficiently from the contents of Web news, in order to explore a research method for network information currency evaluation. The experimental results show the validity of the model design that plays a very important role in leading network users pay attention to more valuable network information and helping Web site managers build a higher currency Web site.