Abstract

How to keep up with the tendency of the literature and grasp the key-points of them from the holistic perspective rapidly is a new challenge both for the literature research and text mining. Most of current theories and tools are directed at finding one paper or a small amount of sample, not gaining a rapid understanding of the hot-keywords of all the papers about one given theme or topic. This paper presents an effort to integrate statistics, text mining, complex networks and visualization to analyze all the papers of one theme-complex networks. we extracted all the 5944 papers about complex networks on Web of Science (http://apps.webofknowledge.com/) from 1990 to 2013. Based on the two-mode affiliation network theory, a new frontier of complex network, we took the keywords of the papers as nodes, took the co-occurrence relationships as the edges and the times of the co-occurrence at the same time as the weight to construct the keywords’ co-occurrence equivalence networks in different year. Then we put forword the integrated way to analyze the evolution and the stability of the keywords’ co-occurrence equivalence networks, and analyze the topological features of the networks about complex networks to find out the hot-keywords and its’ trendency in different time.This paper procvide a useful tool and process to realize rapid understanding of the trend and the hotwords of a large amount of literature successfully.

Share

COinS