Modelling temporal patterns of news report

We study the news system and explore news report mechanism based on the corpus of Sina news website. Instead of using complete content of news, we utilize short keywords to represent original news and mainly focus on the report time series of news. Empirical analysis shows that the distributions of time intervals follow power-law distribution with exponential cutoff on both single-class level and aggregate level. In addition, we analyze the vitality of keywords to explain why the distributions of different kinds of news show difference in the exponential cutoff tail. On the basis of these findings, we propose a hybrid mechanism queuing model to reveal the hidden principle of news reports. The simulation results can well fit the empirical data. The present study may shed some light on the mechanism of temporal patterns of news report and human selection behaviors.