Automatic identification of noun phrases based on statistics and rules

By analyzing the characteristics of Chinese noun phrase, propose an automatic identification method of Chinese noun phrase based on statistics and rules. The method includes efficient binary statistical model, mutual information, and rules for knowledge established from a large number of corpus. Then we analyze the distribution of noun phrase in detail. The results show that the method is fit for the automatic recognition of noun phrase, and the F1 value is 85.49%.