Research hotspot evolution behavior detection method based on KL distance similarity measure

Abstract

The invention discloses a research hotspot evolution behavior detection method based on KL distance similarity measurement, which combines publication subject and publication time series, and proposesa time series publication subject model TS-JTM to extract temporal hot spots from academic journals. On the basis of this, a model of research hotspot evolution of topic snapshot journals based on time series is established. At the same time, a detection method of topic evolution behavior in topic snapshots at adjacent times is proposed by using probability distribution KL distance similarity measurement method, in order to achieve fine-grained analysis of research hotspot evolution in journals.

Description

technical field [0001] The invention belongs to the technical field of document subject analysis and detection, and in particular relates to a method for detecting evolution behavior of research hotspots based on KL distance similarity measurement. Background technique [0002] With the continuous development of scientific research and exploration, research hotspots in the academic field have changed accordingly. Due to the mutual penetration between disciplines and the application of new technologies, academic research hotspots have evolved over time. In this process, some Old research questions will disappear, while new research questions will continue to emerge, and some research questions will fission or merge with other research questions over time. These behaviors lead to the evolution of disciplinary research hotspots. Therefore, it is necessary to analyze the evolution of research hotspots in the academic field and grasp the evolution trajectory of research hotspots ...

AI Technical Summary

Problems solved by technology

However, the evolution of topics over time is an important factor affecting topic extraction. The author topic model does not consider the time factor. When the author topic model is directly used in the corpus database of each time slice for topic extraction, it is independent in each time slice. Model parameters do not have time dependence, and do not take into account the influence of topics changing over time, which increases the uncertainty of topic words when assigning topics

Blei proposed the DTM model based on the LDA (Latent Dirichlet Allocation) model [2] , to achieve the extraction of time-series topics. However, the DTM model obtains the topics contained in each publication in the literature data set and its evolution over time by modeling the content of the data set, not the publications. needs

[0004] Therefore, there is a lack of an effective means in the prior art to solve the problem of behavior detection based on the time series of publications.

Images