Using Topic and Sample Weighting Clustering to Detect Hotspots and Their Trend in a Special Domain

摘要

Detecting hotspots and trend in a special domain based on the traditional method is costly, time-consuming and hysteretic. A method based on topic and sample weighting clustering is proposed to analysis and cluster the academic papers including the publication date. The PageRank value of each document is calculated according to the cited relationship among them, and it is used as the weight in the clustering. The topic and sample weighting clustering method can detect hotspots and trend of a particular domain. Experimental results show that the proposed method can detect hotspots and trend in a special domain quickly and accurately.