关键词: 自然语言处理, 知识图谱, 易水学派, 句法分析

Abstract: Entity and relationship extraction is a crucial component in natural language processing tasks such as knowledge graph construction, question answering system design, and semantic analysis. The information pertaining to Yishui school of traditional Chinese medicine primarily exists in the form of unstructured classical Chinese text, making key information extraction from TCM texts essential for mining and studying TCM academic schools. To efficiently address these challenges using artificial intelligence methods, this paper presents a word segmentation and entity relationship extraction model based on conditional random field within the framework of natural language processing technology to identify and extract entity relationships from TCM texts. Important key entity information from different ancient books is extracted using commonly employed TF-IDF information retrieval and data mining weighting techniques. Additionally, grammatical relationships between entities in each ancient book article are analyzed using a neural network dependency parsing analyzer, which are then represented as tree structures for visualization purposes. This paper lays the foundation for subsequent steps involving building a knowledge graph for Yishui school and utilizing artificial intelligence methods to conduct research on TCM academic schools.

Key words: natural language processing, knowledge graph, Yishui school, syntactic analysis