Title: A front knowledge mapping analysis of international statistical research based on DCMM
Authors: Xing Wang - Renmin University of China (China) [presenting]
Abstract: Based on 31,681 academic papers from 22 authoritative journals of Statistics from 2008 to 2018 collected by Web of Science, a keyword co-occurrence network is built to explore the frontier knowledge map. The keywords structure and interaction mechanism of international Statistics are explored to reveal the connection mechanism of micro-key concepts in different fields within the discipline. By using the Degree-Corrected Mixed-Membership (DCMM) model and its mix-SCORE algorithm, the probability of keywords belonging to the different frontier subjects is estimated. Keywords network topology structure of keyword co-occurrence network is studied. With the bibliometrics experience, the fundamental law of knowledge development of Statistics is revealed, including research fields of Statistics, research hotspots of famous international universities, and the hotspots' growth path. The results illustrate that the influence of the keyword co-occurrence network structure is more significant than that of the keyword itself. The connection mechanism and inheritance mechanism of different universities have a substantial impact. The keyword co-occurrence networks of well-known international universities have various preferential connection mechanisms and diversified academic ecology. The keyword co-occurrence tends to be connected with newer hotspots nodes.