Two Attempts at Building Ancient Chinese Texts Knowledge Bases (談古籍知識庫建設的兩次嘗試)

https://mp.weixin.qq.com/s?__biz=MzU5NjM0MjE3MA==&mid=2247519983&idx=1&sn=ef8b640cfbc8d8518396af1d4f6c0dc3

[AI-Generated Summary]: This article by Sun Xianbin and You Xingchao reviews two pioneering attempts from 15 years ago at Peking University’s Data Analysis Research Center to build ancient Chinese texts knowledge bases - the “Zizhi Tongjian Knowledge Service System” and the “China Historical Bibliography Database System” - discussing the theoretical framework of ontology-based semantic networks, practical challenges in text annotation, and the critical importance of knowledge base construction as foundation for AI applications in ancient texts digitization.

Language: Chinese (中文)

Authors

  • 孫顯斌 (Sun Xianbin) - Chinese Academy of Sciences Institute for the History of Natural Sciences
  • 攸興超 (You Xingchao) - Dougong Technology Co., Ltd.

Key Projects Discussed

1. Zizhi Tongjian Knowledge Service System (資治通鑒知識服務系統)

  • Developed 2008-2019 with Zhonghua Book Company
  • Comprehensive text annotation with 7 ontology types
  • Semantic network analysis capabilities
  • Full-text semantic search functionality

2. China Historical Bibliography Database System (中國歷代典籍總目系統)

  • Developed 2009 with National Library of China
  • 27 bibliography sources, over 2 million entries
  • FRBR-based structural analysis
  • Multi-dimensional navigation and statistical analysis

Core Concepts

  • Ontology (本體): Semantic referents in texts
  • Annotation (標記): Text markers linked to ontologies
  • Four Requirements for Computability: Structuralization, Unification, Standardization, Relationalization

Published in: 《中國古典學》第七輯 (Chinese Classical Studies, Vol. 7)