Abstract
In order to mine the research topics in massive articles, sort out the evolution context and correlation relationship of research topics, enhance the scientificity and vividness of evolution results, this paper proposes the concept of temporal influence factor as an important feature in keyword extraction. Using the time window method, the topic model is used to mine and identify topics and perform visual analysis. It is verified that keyword extraction with time series features can improve the effect of topic model. Through visualization, we can not only observe the overall trend of topic popularity, but also analyze the evolution of topic content in each time period and observe the trend of splitting and merging.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
