Research of news tagging based on word frequency statistics and user information

This paper aims to design a new method to find the statistics model of news with the user information. Especially, the paper focuses on the analysis of new tags. The traditional TF-IDF algorithm based on the word frequency is introduced and there is new parameter of weight to the tags adding to the TF-IDF model. Through the tag selection formula of news tag, we can analysis the user behavior from the tags library, which will help us to make the news more readable to the reviewers. We use recall rate and F1 score to measure the tag value. From the simulation results we can see that the news topic can be modeled using the statistical model, and the work is significant to improve the content of the news.