A Fine-grained Knowledge Graph of Traditional Chinese Medicine Ancient Books

This dataset is in CSV format, consisting of two files: nodes.csv and relations.csv, which are encoded in UTF-8. When in use, two files can be imported into graph database tools such as Neo4j to form a visual knowledge graph, which can be used for further research.nodes.csv contains 26771 traditional Chinese medicine related entities extracted from five traditional Chinese medicine ancient books: 伤寒论, 伤寒类方, 伤寒悬解, 伤寒论浅注, and 伤寒九十论. Among them, each row represents an entity, the first column is the entity ID, the second column is the entity name, and the third column is the entity type.relationships.csv contains 8272 triplets of relationships between entities in nodes.csv. Among them, each row represents a pair of relationships, with the first column being the head entity ID, the second column being the tail entity ID, and the third column being the relationship type.