Interdisciplinary Collaboration and AI-empowered Knowledge Organization: A Digital Curation Practice for the Confucian Philosophy Initiative in China

Abstract

As a core component of Chinese culture, Confucian classics embody the philosophical thoughts and historical context of Eastern civilization. This poster presents an AI-enabled intelligent service project for philosophical concepts in Confucian classics, demonstrating a complete digital curation framework that uses generative AI to automate knowledge processing, organization, preservation, and presentation.

In the first phase, senior Confucian scholars and librarians collaboratively built and refined the domain ontology CAPOnto to systematically standardize the Confucian conceptual system. In the second phase, a technical team led by computer scientists developed CAP-Agent, an intelligent knowledge indexing tool based on the Qwen-max large language model, powered by hybrid Retrieval-Augmented Generation (RAG) and Chain-of-thought techniques. It automatically parsed texts and extracted concepts, relations, and metadata to construct a high-quality knowledge graph compatible with CAPOnto.

The third stage focused on platform engineering and deployment, delivering a reader, interactive visualization, linked open data APIs, and a knowledge-graph-enhanced QA assistant, with sustainable iteration supported by technical optimization and professional evaluation from Confucian scholars and library cataloging specialists. This interdisciplinary practice promotes open sharing of philosophical knowledge and provides a replicable AI-driven approach for efficient extraction, organization, and long-term intelligent curation of ancient Chinese documents.