语料库语言学视域下数据驱动的数字人文研究--以《数字人文季刊》为例

Data-driven Digital Humanities from the Perspective of Corpus Linguistics--A Case Study on the Digital Humanities Quarterly

基于语料库语言学和数字人文之间的内在联系,运用语料库语言学对数据处理的理念和工具,实现数字人文领域不同学科视角的文本挖掘,可为数字人文发展热点和趋势提供更全面的论据。文章以语料库语言学为理论背景,基于数据驱动的研究范式,通过构建《数字人文季刊》语料库,运用语料库分析工具Wmatrix和文本可视化工具Voyant,综合采用定性和定量研究方法,对语料库进行关键词、词性、语义域3个维度的语料分析,旨在开展数字人文领域的历时性语言分析来窥探数字人文10多年来的衍变路径和未来趋势。通过对语料库语言学视角的DHQ历时语料库进行研究对象、研究模式、数字技术、研究导向和研究群体5个维度的分析,发现:对数据的形式化要求、众包新模式的探索、自然语言处理和可视化数字技术的刚需、人文性的重视和对不同角色的女性群体的关注是目前数字人文领域研究的研究重点,这给未来数字人文的发展提供了方向。

Since there is a kind of internal connection between Corpus Linguistics and Digital Humanities,it is feasible to introduce the concepts and tools of data processing in Corpus Linguistics into Digital Humanities,so as to realize text mining from different disciplinary perspectives and provide a more comprehensive argument for determining the hot spots and trends of Digital Humanities.In this paper,Corpus Linguistics is selected as the theoretical background,a data-driven research paradigm is adopted,and both qualitative and quantitative research methods are used.By means of the corpus analysis tool Wmatrix and the text visualization tool Voyant,it makes a diachronic linguistic analysis of a self-constructed Digital Humanities Quarterly(DHQ)corpus in the following three dimensions,i.e.,keyword analysis,part-of-speech analysis,and semantic domain analysis.It aims to find out the development path and trends of Digital Humanities in the past ten years,focusing on research objects,research models,digital technologies,research orientation and research community.The result shows that there are five major hot topics in current Digital Humanities,i.e.,the formalization of data,the new models of crowdsourcing,natural language processing and visual digital technology,the emphasis on humanity,and more attention to female groups with different roles.Such could show a direction for the future development of Digital Humanities.