学科主题探测与演化分析研究述评

Recent Advances in Detection and Evolutionary Tracking of Scientific Topics:A Multi-Perspective Survey and Prospect

如何借助于有效的方法和途径在大量文献数据基础上对学科主题进行深入和精准的探测与跟踪,是以科学计量领域为核心、不同学科共同致力突破的热点和前沿研究问题。相关方法和途径主要涉及频次视角、内容视角、引证视角及融合视角4个视角。文章试图结合近年来发表于国际国内重要期刊有关学科主题探测和演化分析的最新文献,对相关视角的主要进展进行述评,归纳总结不同视角的实现路径和机制,指出已有视角可能存在的知识单元或者网络关系的异质重要性偏差问题、知识的时间衰变以及新兴主题特征的小样本弱势问题、主题自然发育和进化的拟合困境、微观层面的知识流动和变迁刻画问题等,特别是针对融合视角的研究这一总体趋势指明方向。

How to leverage instructive approaches to detect and track the topics in depth and accurately from the big data of historical literature is a hot and cuttingedge issue garnered substantial interest from different disciplines,typically scientometrics in information science.The major mechanisms behind involve four perspectives,including frequency analysis,content analysis,citation analysis and synthesized analysis.This study attempts to review the latest literature published in the prestigious international and domestic journals in recent years to update the main progress of relevant perspectives in scientific topic detection and evolution analysis by unravelling their essential implementation paths.Particularly,it raises the risks inherited in existing perspectives,including the biases in heterogeneous importance of knowledge units or network relationships,the temporal decay of knowledge,the trap of the small samples of emergent topics,the dilemma in fitting the natural development and evolution of topics,the failure in portraying knowledge flow and evolution in micro level,and so on.In response to these issues,the study urges attention to the future trend of synthesized perspective.