The Knowledge Graph Dataset for Martial Arts Culture

The dataset constructs a high-quality martial arts knowledge dataset for large model applications centered on the theme of martial arts, covering multiple core dimensions such as schools, figures, and encyclopedic knowledge. The dataset mainly uses information from three authoritative platforms—the China Wushu Association, the General Administration of Sport of China, and Baidu Baike—as well as the civil platform Chinese Wushu Heritage Network as primary data sources. Through manual screening, standardized processing, and quality monitoring, approximately 1,000 original data entries were generated, including information on martial arts encyclopedic knowledge, martial arts schools, renowned martial arts masters and their lineage relationships, martial arts rules and regulations, and martial arts exchanges. Using large language models, more than 25,000 triple (subject–predicate–object) relationships were extracted from the constructed martial arts data. Among them, the inheritor files contain introductions of inheritors from various schools and their lineage relationships; the competition rules, regulations, and policy documents include competition rules within the martial arts domain; the school files contain introductions to various martial arts schools; the Martial Arts Encyclopedia and Chinese Martial Arts Encyclopedia files include knowledge about the meaning, historical development, and significance of martial arts; the martial arts masters and martial arts institutions files include well-known masters, famous institutions, and their introductions; and the martial arts exchange and information release files contain detailed information related to communication and information sharing in the martial arts field. This dataset provides a solid data foundation for the application of question-answering models in intelligent knowledge-based QA tasks and also promotes the digital dissemination of martial arts culture and the deeper application of intelligent systems in the field of traditional culture.