Detecting Hot Topics And Research Fronts From The Article Downloading Data

For scientific researchers, timely and accurately knowing the current topics conducted by other researchers in the same field helps obtaining the latest trend of scientific research and the research frontier, to grasp the advantage in the global science and technology competition. Previous analyses of research trends are based on the backtracking of those published literature and have long period of time lag, which often makes the results being the study of the past research trends.In this study, we propose a new idea:through collecting the downloading information of scientific literature by researchers worldwide, to estimate their research topics currently engaged in, trace real-timely the research trend in a specific area, and to detect the research hotspots and research frontiers. Based on this idea, we present the new scientific research trend detecting model by analyzing the downloading data of scientific papers. Using this model and taking the Scientometrics journal for example, we build the new scientific research trend knowledge mining framework. By24/7uninterrupted long-term collecting of downloading data, with text mining and information metrology methods, to detect the weekly and monthly research hotspots and new research trends. Taking the data in July and August in2003as examples, we find that in scientometrics field, indicator, citation analysis, h-index, collaboration, impact factor, technology, innovation, text mining are research hot topics, Parts of speech, Data mining, Synergy, Network analysis are emerging fields that scientometrics researchers are focusing on.