滑动时间窗视角下科学基金研究前沿探测

研究前沿是某领域处于先锋地位的研究内容。想要更好的理解当下和未来的研究趋势,需要时刻关注领域的研究前沿。文章综述了前人的研究方法,指出其共有问题,提出滑动时间窗视角下基于科学基金的前沿探测方法。并在选取最优聚类算法的基础上,以肿瘤学为例探测其研究前沿。结果表明Agglomerative的聚类效果最优,对4个滑动时间窗内的数据进行共词聚类分析,探测其研究前沿并分析其演化规律。

Detection of scientific funds-supported research fronts using sliding window

Research fronts represent cutting edge studies in specific fields. One can better understand current and future development trends in the relevant field when updated with trends in research fronts. Based on the summary of existing methods, this paper analyzes the main problems, and proposes a method of fronts detecting which based on funds with sliding window. Taking the field of oncology as example to verify, we select the optimal clustering algorithm. It turns out that Agglomerative is the best, and we conduct the co-word clustering to detect and analyze the evolution in four sliding windows.