数字人文视域下的古文献文本标注与可视化研究∗ ———以«左传»知识库为例.

The traditional research paradigms based on digitalization and full-text retrieval are not sufficient for history, philology and linguistic studies within the paradigm of digital humanities. For quantitative analysis and visualization, the words, time, places, persons and events in the ancient texts need to be constructed into a historical digital database. By applying the natural language processing technology of ancient Chinese, the word segmentation, parts-of-speech, time, person ID, location ID and GIS information are annotated with manual checking. Thus, the Zuozhuan digital humanity knowledgebase is constructed. Then the key persons, social networks, personal traveling tracks and distances in Zuozhuan are calculated, which may propose new methods to related ancient texts annotation, structural humanity knowledge base construction in history and philology research. Finally, the potential improvements and applications of the database are discussed furthermore. [ABSTRACT FROM AUTHOR]