Intelligent question and answer method and system based on traditional Chinese medicine classics meta-term engine

Abstract

The invention provides an intelligent question answering method and system based on a traditional Chinese medicine classical meta-term engine. The method comprises the following steps: identifying and extracting core traditional Chinese medicine terms and a user query intention in a user question; the core traditional Chinese medicine terms are input into a pre-constructed traditional Chinese medicine classical meta-term engine to be processed, normalized mapping and semantic association extension of the terms are completed, and the engine is constructed based on a classification-term-word list-semantic CTVS association framework. Comprising a traditional Chinese medicine classical book term ancient and modern double-classification system C layer, a meta-term extraction and standardization T layer, a dynamic word list V layer and a semantic association network S layer. Carrying out graph pattern matching and reasoning in the S layer, traversing and integrating a plurality of deep association paths by taking classical element terms and corresponding modern clinical terms as double center nodes, and retrieving answer information associated with the query intention; and carrying out integration sorting and structured assembly on the answer information to generate a structured answer. Intelligent questions and answers can be performed on traditional Chinese medicine related questions of the user.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent question and answer of traditional Chinese medicine, in particular to an intelligent question and answer method and system based on a traditional Chinese medicine classic book meta-terminology engine. BACKGROUND

[0002] Traditional Chinese medicine classics are important carriers of the essence of Chinese traditional culture. The classical theories and treatment experiences recorded in them still have important practical value today and are an important source of inheritance and innovative development of traditional Chinese medicine. Effectively obtaining high-quality traditional Chinese classic knowledge and transforming and utilizing it to solve current clinical problems are the key to supporting clinical practice and technological innovation with classics. Under the background of the digital era, it is of great significance to realize the representation of the content structure of traditional Chinese medicine classic literature and the sequential organization of knowledge elements, but the complex semantic expression and various terminologies pose challenges to knowledge classification, retrieval and application.

[0003] With the development of artificial intelligence technology, intelligent question and answer systems are increasingly widely used in the medical field. However, in the field of traditional Chinese medicine, due to the unique knowledge system, the complex evolution of terminology over time, and the existence of a large number of synonymous phenomena, the direct application of general intelligent question and answer technology faces great challenges. The main problems existing at present include low standardization of terminology, shallow semantic understanding, and fragmented knowledge sources. SUMMARY

[0004] In order to solve the technical problems existing in the prior art, the present application provides an intelligent question and answer method and system based on a traditional Chinese medicine classic book meta-terminology engine, and the technical solution is as follows: On the one hand, an intelligent question and answer method based on a traditional Chinese medicine classic book meta-terminology engine is provided, which comprises: S1, receiving a natural language question input by a user, performing semantic analysis on it, and identifying and extracting core traditional Chinese medicine terminology and user query intent in the question; S2, inputting the core traditional Chinese medicine terminology into a pre-constructed traditional Chinese medicine classic book meta-terminology engine for processing, completing standardized mapping and semantic association expansion of the terminology, the traditional Chinese medicine classic book meta-terminology engine being constructed based on a classification-terminology-vocabulary-semantic CTVS association framework, comprising a C layer of ancient and modern dual classification systems of traditional Chinese medicine classic book terminology, a T layer of meta-terminology extraction and standardization, a V layer of dynamic vocabulary, and an S layer of semantic association network; S3, performing graph pattern matching and reasoning in the semantic association network S layer of the traditional Chinese medicine classic book meta-terminology engine according to the standardized mapping and semantic association expanded terminology, traversing and integrating multiple depth association paths with the dual center nodes of classic book meta-terminology and corresponding modern clinical terminology, and retrieving answer information associated with the query intent; S4, integrating, sorting and structuring the answer information, generating a structured answer, and returning it to the user for visual presentation.

[0005] Optionally, the pre-construction of the TCM classics meta-terminology engine specifically includes: (a) Constructing a dual classification system of ancient and modern terminology in traditional Chinese medicine classics, Level C: Establish a classification system for TCM classic terminology that includes multiple categories, and based on the current national standards for TCM clinical diagnosis and treatment, generate a standardized mapping table between classic terminology and clinical terminology to achieve the mapping and association between ancient and modern terminology; (b) Meta-term extraction and normalization layer T: Core meta-terms are extracted from traditional Chinese medicine classics using a natural language processing model, and terminology standardization is performed based on the mapping table obtained in step (a), the pre-constructed alias-standard name correspondence table, and the characteristic alias-standard name detection model. (c) Constructing the dynamic vocabulary V layer: A dynamic thesaurus is constructed for the terms normalized in step (b), the thesaurus including multiple core data items for describing the multidimensional attributes of the terms; (d) Construct the S layer of the semantic association network: Based on the classification and relation data in the dynamic vocabulary obtained in step (c), a semantic association network of terms is constructed. Steps (a), (b), (c), and (d) are sequentially linked and collaboratively construct the meta-terminology engine for traditional Chinese medicine classics.

[0006] Optionally, the construction of the ancient and modern dual classification system C-layer of TCM classic terminology in step (a) specifically includes: (i) Construct a classification system for terms in TCM classics, including 15 primary categories: basics, diagnostic methods, drugs, prescriptions, diseases, symptoms, syndromes, acupuncture, massage, health preservation, medical books, physicians, time, space and others, which are further expanded into 64 secondary categories; (ii) Establish a mapping and association of TCM clinical diagnosis and treatment terms based on the current national standards GB / T 16751.1-2023 "TCM Clinical Diagnosis and Treatment Terminology Part 1: Diseases", GB / T 16751.2-2021 "TCM Clinical Diagnosis and Treatment Terminology Part 2: Syndromes", and GB / T 16751.3-2023 "TCM Clinical Diagnosis and Treatment Terminology Part 3: Treatment Methods"; (iii) Generate a standardized mapping table of classical and clinical terms, including encoding and definition, as a framework for semantic conversion between ancient and modern terms.

[0007] Optionally, the extraction and normalization of meta-terms in step (b) are achieved through a multi-strategy fusion processing method, including the following steps: (i) Initialization construction: Based on the pre-built core dictionary of traditional Chinese medicine and the standardized mapping table generated in step (a), an initial term seed set is constructed, and an alias-standard name mapping library is constructed according to the pre-built alias-standard name correspondence table and the characteristic alias-standard name detection model; (ii) Multi-strategy collaborative extraction: The target classical text is initially screened using a sequence labeling model based on active learning to automatically obtain candidate terms with high confidence; at the same time, a pattern matching method based on semantic roles and dependency parsing is used to extract terms and their associated information from specific sentence structures in the original text. (iii) Human-machine collaborative annotation and optimization: For the low-confidence results and conflicting results in the above steps, domain experts will manually verify and annotate them; the manually annotated results will be fed back to the sequence annotation model as training samples and the rule base for pattern matching will be updated to achieve iterative optimization of the model and self-improvement of annotation efficiency. (iv) Knowledge-guided terminology normalization: For the extracted candidate terms, the alias-standard name mapping library is first queried to merge synonyms; for terms that fail to match, the semantic similarity between the term seed set and the term seed set in the vector space and the contextual dependency path features in the classics are calculated to make a comprehensive judgment, thereby completing the term disambiguation and standardization mapping.

[0008] Optionally, the alias-standard name correspondence table includes the standard name of each term and its corresponding common aliases; The distinctive alias-canonical name detection model is used to further calculate potential aliases for each term, and the steps are as follows: Constructing a feature set: For each term Wi, construct a basic feature set Fwi, which includes the corresponding structured attribute information, Fwi={c1,c2,c3,...}, where c1,c2,c3 are attribute lists; The calculation constructs are as follows: Intersection: the number of features shared by two terms |Fwa∩Fwb|=Fshare; Union: the total number of features of all two terms |Fwa∪Fwb|=Fall; Symmetric difference: the union and intersection of the two term sets, Fall-Fshare=Fsym; Total number: FWa+FWb=Fnum; Perform the calculation: Sim(Wa,Wb) = ((2 Fshare) / Fnum)-(Fsym / Fall), where Sim(Wa,Wb) represents the similarity between the two terms, the first part ((2 Fshare / Fnum emphasizes the importance of shared features; the more shared features, the higher the similarity. The second part, Fsym / Fall, slightly penalizes the mismatched parts, making the results more balanced. When calculating similarity, this part of the value is subtracted, thus slightly reducing the similarity score. This ensures that even if two terms have many shared features, their differences will still be reflected in the score, while helping to avoid misclassification or excessively high similarity scores due to ignoring differences. Similarity normalization: The similarity results of the two terms are normalized: Sim(Wa,Wb)=max(0,min(1,Sim(Wa,Wb))), ensuring that the similarity score is between [0, 1]. Output results: Two terms with similarity scores higher than a preset threshold will be output as their canonical names and corresponding aliases.

[0009] Optionally, the dynamic vocabulary in step (c) includes 16 core data items: V-1 Serial Number; V-2 Primary Classification; V-3 Secondary Classification; V-4 See Classification; V-5 Meta-Term ID; V-6 Terminology; V-7 Pinyin; V-8 Definition; V-9 Variant Characters; V-10 Synonyms; V-11 Near-Synonyms; V-12 Hypernyms; V-13 Hyponyms; V-14 Connectives; V-15 Modern Medical Terminology; V-16 Source Texts.

[0010] Optionally, the step of constructing the semantic association network in step (d) includes: (ii) Parse the relation fields in the dynamic vocabulary and directly extract the predefined hierarchical relationships; (iii) Analyze the definition fields in the dynamic vocabulary and the original text context of the terms to extract the attribute relationships and function relationships of the terms; (iv) Calculate and optimize the confidence scores of the extracted relations to form a semantic association network rich in various relation types.

[0011] Optionally, in step S3, using classical meta-terms and corresponding modern clinical terms as dual-center nodes, multiple deeply related paths are traversed and integrated to retrieve answer information associated with the query intent, specifically including: In the S layer of the semantic association network, using classical meta-terms and their corresponding modern clinical terms as dual-center nodes, the following deep association paths are traversed and integrated: 1) Mapping path between ancient and modern disease names: directly obtain the predefined equivalent mapping relationship between classical meta-terms and their corresponding modern clinical terms; 2) Pathogenesis-Pathology Mapping Path: Traverse the semantic relationships from classical meta-terms to their core pathogenesis, and then from the pathogenesis to the modern pathophysiological nodes. 3) The path of diagnosis and treatment: Traversing the relationship from the meta-terms in classical texts to their various treatment methods, and then to the treatment of various representative prescriptions; 4) Symptom-sign association pathway: Traversing the relationship between symptom descriptions in ancient books and the manifestations of modern clinical examination indicators; 5) Biomarker association pathways: Traversing the corresponding biomarker relationships from traditional Chinese medicine pathogenesis to various modern biomarkers; Based on the traversal results of each associated path, retrieve the answer information related to the query intent.

[0012] On the other hand, an intelligent question-answering system based on a meta-terminology engine from traditional Chinese medicine classics is provided, the system comprising: The parsing module is used to receive natural language questions input by users, perform semantic parsing on them, identify and extract core TCM terms and user query intent from the questions; The processing module is used to input the core TCM terms into a pre-built TCM classic meta-term engine for processing, and to complete the standardized mapping and semantic association expansion of the terms. The TCM classic meta-term engine is built on the classification-term-vocabulary-semantic CTVS association framework, including a TCM classic term dual classification system (ancient and modern) layer C, a meta-term extraction and standardization layer T, a dynamic vocabulary layer V, and a semantic association network layer S. The retrieval module is used to perform graph pattern matching and reasoning in the semantic association network S layer of the TCM classic meta-term engine based on the terms after normalization mapping and semantic association expansion. With the classic meta-terms and corresponding modern clinical terms as dual center nodes, it traverses and integrates multiple deep association paths to retrieve answer information related to the query intent. The generation module is used to integrate, sort, and structurally assemble the answer information to generate a structured answer, which is then returned to the user for visual presentation.

[0013] On the other hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, which is loaded and executed by the processor to implement the above-mentioned intelligent question-answering method based on the meta-terminology engine of traditional Chinese medicine classics.

[0014] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement the above-mentioned intelligent question-answering method based on the meta-terminology engine of traditional Chinese medicine classics.

[0015] The beneficial effects of the technical solution provided by this invention include at least the following: This invention, through the CTVS four-layer linkage of the TCM classic meta-term engine, can successfully transform a user's simple query about a modern medical question into a semantic calculation that deeply traverses the knowledge network of ancient books, ultimately achieving a true "fusion of ancient and modern knowledge" and intelligent question-and-answer service. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of an intelligent question-answering method based on a meta-term engine of traditional Chinese medicine classics, provided by an embodiment of the present invention. Figure 2 This is an overall block diagram of an intelligent question-answering method based on a meta-term engine of traditional Chinese medicine classics provided in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the process of constructing a meta-terminology engine for traditional Chinese medicine classics, as provided in this embodiment of the invention. Figure 4 This is a schematic diagram of the T-layer for meta-term extraction and normalization provided in an embodiment of the present invention; Figure 5 This is a block diagram of an intelligent question-answering system based on a meta-terminology engine of traditional Chinese medicine classics, provided in an embodiment of the present invention. Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0019] This invention provides an intelligent question-answering method based on a meta-terminology engine for traditional Chinese medicine (TCM) classics. Its core lies in first constructing a structured TCM classics meta-terminology engine, and then using this engine to empower the intelligent question-answering process. It innovatively constructs a four-layer progressive framework (CTVS), which, through structured hierarchies, connects the entire path from basic classification to deep semantic association. By deeply integrating terminology processing technology with semantic network construction, it achieves a breakthrough leap from discrete storage to deep semantic reasoning of TCM classics terminology, significantly improving knowledge retrieval accuracy. Furthermore, it solves the problem of numerous synonyms and homonyms existing in current technologies through a unique alias-standard name detection model.

[0020] This invention provides an intelligent question-answering method based on a meta-terminal engine of traditional Chinese medicine classics. This method can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The flowchart of this method is shown below. Figure 2 The diagram shown is an overall block diagram of the method. The processing flow may include the following steps: S1. Receive natural language questions input by users, perform semantic analysis on them, identify and extract core TCM terms and user query intent from the questions; In this embodiment of the invention, a large language model can be used to perform semantic parsing on natural language questions input by users, and to identify and extract core TCM terms and user query intent from the questions.

[0021] S2. Input the core TCM terms into the pre-built TCM classic meta-term engine for processing to complete the standardized mapping and semantic association expansion of the terms. The TCM classic meta-term engine is built on the classification-term-vocabulary-semantic CTVS association framework, including the ancient and modern dual classification system of TCM classic terms (C layer), meta-term extraction and standardization (T layer), dynamic vocabulary (V layer), and semantic association network (S layer). Optionally, such as Figure 3 As shown, the pre-construction of the TCM classics meta-terminology engine specifically includes: (a) Constructing a dual classification system of ancient and modern terminology in traditional Chinese medicine classics, Level C: Establish a classification system for TCM classic terminology that includes multiple categories, and based on the current national standards for TCM clinical diagnosis and treatment, generate a standardized mapping table between classic terminology and clinical terminology to achieve the mapping and association between ancient and modern terminology; (b) Meta-term extraction and normalization layer T: Core meta-terms are extracted from traditional Chinese medicine classics using a natural language processing model, and terminology standardization is performed based on the mapping table obtained in step (a), the pre-constructed alias-standard name correspondence table, and the characteristic alias-standard name detection model. (c) Constructing the dynamic vocabulary V layer: A dynamic thesaurus is constructed for the terms normalized in step (b), the thesaurus including multiple core data items for describing the multidimensional attributes of the terms; (d) Construct the S layer of the semantic association network: Based on the classification and relation data in the dynamic vocabulary obtained in step (c), a semantic association network of terms is constructed. Steps (a), (b), (c), and (d) are sequentially linked and collaboratively construct the meta-terminology engine for traditional Chinese medicine classics.

[0022] Optionally, the construction of the ancient and modern dual classification system C-layer of TCM classic terminology in step (a) specifically includes: (i) Construct a classification system for terms in TCM classics, including 15 primary categories: basics, diagnostic methods, drugs, prescriptions, diseases, symptoms, syndromes, acupuncture, massage, health preservation, medical books, physicians, time, space and others, which are further expanded into 64 secondary categories; (ii) Establish a mapping and association of TCM clinical diagnosis and treatment terms based on the current national standards GB / T 16751.1-2023 "TCM Clinical Diagnosis and Treatment Terminology Part 1: Diseases", GB / T 16751.2-2021 "TCM Clinical Diagnosis and Treatment Terminology Part 2: Syndromes", and GB / T 16751.3-2023 "TCM Clinical Diagnosis and Treatment Terminology Part 3: Treatment Methods"; (iii) Generate a standardized mapping table of classical and clinical terms, including encoding and definition, as a framework for semantic conversion between ancient and modern terms.

[0023] Optionally, such as Figure 4 As shown, the extraction and normalization of meta-terms in step (b) is achieved through a multi-strategy fusion processing method, including the following steps: (i) Initialization construction: Based on the pre-built core dictionary of traditional Chinese medicine and the standardized mapping table generated in step (a), an initial term seed set is constructed, and an alias-standard name mapping library is constructed according to the pre-built alias-standard name correspondence table and the characteristic alias-standard name detection model; (ii) Multi-strategy collaborative extraction: A sequence labeling model based on active learning (such as a BERT-BiLSTM-CRF joint model) is used to initially screen the target classical text and automatically obtain candidate metaterms with high confidence; at the same time, a pattern matching method based on semantic roles and dependency parsing is used to extract terms and their associated information from specific sentence structures in the classical text. (iii) Human-machine collaborative annotation and optimization: For the low-confidence results and conflicting results in the above steps, domain experts will manually verify and annotate them; the manually annotated results will be fed back to the sequence annotation model as training samples and the rule base for pattern matching will be updated to achieve iterative optimization of the model and self-improvement of annotation efficiency. (iv) Knowledge-guided terminology normalization: For the extracted candidate meta-terms, the alias-standard name mapping library is first queried to merge synonyms; for terms that fail to match, the semantic similarity between the term seed set and the term seed set in the vector space and the contextual dependency path features in the classics are calculated to make a comprehensive judgment, thereby completing the term disambiguation and standardization mapping.

[0024] Optionally, the alias-standard name correspondence table includes the standard name of each term and its corresponding common aliases; The distinctive alias-canonical name detection model is used to further calculate potential aliases for each term, and the steps are as follows: Constructing a feature set: For each term Wi, construct a basic feature set Fwi, which includes the corresponding structured attribute information, Fwi={c1,c2,c3,...}, where c1,c2,c3 are attribute lists; The calculation constructs are as follows: Intersection: the number of features shared by two terms |Fwa∩Fwb|=Fshare; Union: the total number of features of all two terms |Fwa∪Fwb|=Fall; Symmetric difference: the union and intersection of the two term sets, Fall-Fshare=Fsym; Total number: FWa+FWb=Fnum; Perform the calculation: Sim(Wa,Wb) = ((2 Fshare) / Fnum)-(Fsym / Fall), where Sim(Wa,Wb) represents the similarity between the two terms, the first part ((2 Fshare / Fnum emphasizes the importance of shared features; the more shared features, the higher the similarity. The second part, Fsym / Fall, slightly penalizes the mismatched parts, making the results more balanced. When calculating similarity, this part of the value is subtracted, thus slightly reducing the similarity score. This ensures that even if two terms have many shared features, their differences will still be reflected in the score, while helping to avoid misclassification or excessively high similarity scores due to ignoring differences. Similarity normalization: The similarity results of the two terms are normalized: Sim(Wa,Wb)=max(0,min(1,Sim(Wa,Wb))), ensuring that the similarity score is between [0, 1]. Output results: Two terms with similarity scores higher than a preset threshold will be output as their canonical names and corresponding aliases.

[0025] The unique alias-standard name detection model of this invention not only considers the intersection and union of the terms themselves, but also incorporates the influence of symmetry difference and total number, thereby improving the accuracy and interpretability of term similarity assessment. For example, the features of "Huangqi" are {Qi deficiency, Qi tonification, Leguminosae, Root, Unique Feature 1}, and the features of "Mianqi" are {Qi deficiency, Qi tonification, Leguminosae, Root, Unique Feature 2}. Then "Huangqi" and "Mianqi" each have 5 features, Fwa=5, Fwb=5. The intersection is {Qi deficiency, Qi tonification, Leguminosae, Root}, so Fshare = 4; the union is {Qi deficiency, Qi tonification, Leguminosae, Root, Unique Feature 1, Unique Feature 2}, so Fall = 6; Fsym = Fall - Fshare = 6 - 4 = 2; Fnum = FWa + FWb = 5 + 5 = 10; similarity Sim(Wa, Wb) = ((2 Fshare) / Fnum)-(Fsym / Fall)=((2 4) / 10)=0.8-(2 / 6)≈0.467>Preset threshold 0.4, output standard name: "Astragalus", alias: "Mianqi".

[0026] Optionally, the dynamic vocabulary in step (c) includes 16 core data items: V-1 Serial Number; V-2 Primary Classification; V-3 Secondary Classification; V-4 See Classification; V-5 Meta-Term ID; V-6 Terminology; V-7 Pinyin; V-8 Definition; V-9 Variant Characters; V-10 Synonyms; V-11 Near-Synonyms; V-12 Hypernyms; V-13 Hyponyms; V-14 Connectives; V-15 Modern Medical Terminology; V-16 Source Texts.

[0027] Optionally, the step of constructing the semantic association network in step (d) includes: (ii) Parse the relation fields in the dynamic vocabulary and directly extract the predefined hierarchical relationships; (iii) Analyze the definition fields in the dynamic vocabulary and the original text context of the terms to extract the attribute relationships and function relationships of the terms; (iv) Calculate and optimize the confidence scores of the extracted relations to form a semantic association network rich in various relation types.

[0028] S3. Based on the terms after normalization mapping and semantic association expansion, graph pattern matching and reasoning are performed in the S layer of the semantic association network of the TCM classic meta-term engine. Using the classic meta-terms and the corresponding modern clinical terms as dual center nodes, multiple deep association paths are traversed and integrated to retrieve answer information related to the query intent. Optionally, in step S3, using classical meta-terms and corresponding modern clinical terms as dual-center nodes, multiple deeply related paths are traversed and integrated to retrieve answer information associated with the query intent, specifically including: In the S layer of the semantic association network, using classical meta-terms and their corresponding modern clinical terms as dual-center nodes, the following deep association paths are traversed and integrated: 1) Mapping path between ancient and modern disease names: directly obtain the predefined equivalent mapping relationship between classical meta-terms and their corresponding modern clinical terms; 2) Pathogenesis-Pathology Mapping Path: Traverse the semantic relationships from classical meta-terms to their core pathogenesis, and then from the pathogenesis to the modern pathophysiological nodes. 3) The path of diagnosis and treatment: Traversing the relationship from the meta-terms in classical texts to their various treatment methods, and then to the treatment of various representative prescriptions; 4) Symptom-sign association pathway: Traversing the relationship between symptom descriptions in ancient books and the manifestations of modern clinical examination indicators; 5) Biomarker association pathways: Traversing the corresponding biomarker relationships from traditional Chinese medicine pathogenesis to various modern biomarkers; Based on the traversal results of each associated path, retrieve the answer information related to the query intent.

[0029] S4. Integrate, sort, and structurally assemble the answer information to generate a structured answer, and return it to the user for visual presentation.

[0030] The concept of meta-terms in traditional Chinese medicine (TCM) classics in this invention refers to the core terms found in TCM classics, which are also the core concepts constituting the TCM academic theoretical system. These include basic terms such as viscera, meridians, and acupoints, as well as applied terms such as diseases, prescriptions, and drugs. These terms are highly abstract, representative, and academic. Centered on these terms, a multi-level, multi-dimensional, and interconnected conceptual system is formed, supporting and structuring the overall TCM traditional academic network.

[0031] Classification of meta-terms in traditional Chinese medicine classics: see Table 1, which mainly includes 15 first-level categories, 64 second-level categories, and 51 third-level categories.

[0032] Table 1. Classification Framework of Terminology in Traditional Chinese Medicine Classics

[0033] Principles for Defining Meta-Terminology in Traditional Chinese Medicine (TCM) Classics: 1. **Clarity of Connotation:** Meta-terminology in TCM classics possesses clear conceptual connotations, occupies a core position in the TCM academic theoretical system, and permeates the entire theoretical system. Examples include Zang-Fu organs (liver), acupoints (Hegu), diseases (Xiao Ke), and drugs (Huang Qi). 2. **Academic Representativeness:** Meta-terminology in TCM classics represents the most widely used and representative core terminology across multiple expressions of the same academic concept. For example, Xiao Ke has different expressions in TCM classics such as Xiao Dan, Fei Xiao, and Shang Xiao, with Xiao Ke being the most widely used and representative. 3. **Correspondence Between Ancient and Modern Times:** Meta-terminology in TCM classics possesses a certain degree of historical continuity and stability, and can provide guidance in the clinical application of modern diseases as much as possible. For example, the modern disease diabetes can find corresponding diagnostic and treatment methods in TCM classics through the meta-terminology Xiao Ke and related terms.

[0034] The concept of the TCM classic meta-term engine in this invention embodiment: The TCM classic meta-term engine refers to the identification, extraction and integration of core terms from TCM classics based on information technologies such as artificial intelligence, big data, and natural language processing, and the realization of systematic expression and transformation application of TCM classic knowledge through semantic association, knowledge graph construction and other methods.

[0035] The core functions of the Traditional Chinese Medicine Classics Meta-Terminology Engine include: 1. Meta-terminology identification and standardization: Integrating multiple methods such as batch extraction based on terminology lists, automatic identification based on model algorithms, and manual review, the engine identifies and integrates terms in Traditional Chinese Medicine classics, selects meta-terms with theoretical foundation, high frequency of use, and interdisciplinary relevance, and standardizes their definitions. 2. Semantic association and knowledge network construction: Embedding meta-terms into a multi-dimensional semantic space, establishing a network of associations between them and different conceptual terms, and constructing a knowledge graph of Traditional Chinese Medicine classics based on the full-text database of Traditional Chinese Medicine classics.

[0036] Application Scenarios of the Meta-Terminology Engine for Traditional Chinese Medicine Classics: 1. Thematic Ancient Book Retrieval and Knowledge Discovery: Based on a standardized meta-terminology system built upon a "classification-terminology-thesaurus-semantics" framework, it enables deep indexing and organization of massive and heterogeneous traditional Chinese medicine classics. It supports users in conducting precise, semantic, and associative searches based on dimensions such as disease, syndrome, treatment method, prescription, Chinese medicine, and theoretical concepts. Breaking through the limitations of traditional keyword matching, it effectively mines relevant discussions scattered across different classics, significantly improving the efficiency and depth of ancient book knowledge discovery. 2. Automated / Semi-Automated Construction of Traditional Chinese Medicine Knowledge Graphs: Using the "meta-terms" extracted and structured by the engine and their inherent semantic relationships (such as synonymy, hierarchical, compatibility, indications, contraindications, etiological and pathogenesis associations, etc.) as core nodes and edges, it provides a structured and standardized semantic foundation for constructing large-scale, high-quality knowledge graphs in the field of traditional Chinese medicine classics. It significantly reduces the difficulty and cost of entity recognition and relationship extraction during knowledge graph construction, accelerating the generation, enrichment, and updating of the graph. 3. Intelligent Question-Answering System: Transforms obscure and difficult-to-understand traditional Chinese medicine classics into structured data elements, supports multi-channel question-and-answer retrieval, and promotes cross-temporal and cross-cultural dialogue between traditional Chinese medicine theory and modern medicine. 4. Clinical Decision Support: Through the correspondence between ancient and modern terminology, comprehensively and systematically retrieves similar case studies and treatment methods in traditional Chinese medicine classics to assist clinical decision-making.

[0037] The intelligent question-answering process of this invention is as follows: After the user inputs a question, the question is first parsed to identify core terms and query intent. Then, the service of the Traditional Chinese Medicine Classics Meta-Terminology Engine is invoked: the V-layer vocabulary is used to standardize and semantically expand the terms. Next, the standardized query is mapped to a C-layer dual-classification system, and graph pattern matching and reasoning are performed in the S-layer semantic network to find the optimal answer path. Finally, by integrating information such as definitions and original text fragments from the V-layer, a structured and easily understandable answer is generated and visualized.

[0038] The following two specific embodiments illustrate the construction process of the TCM classic meta-terminology engine and the intelligent question-answering method based on the TCM classic meta-terminology engine of this invention: Example 1: This embodiment uses type 2 diabetes (TCM disease name: Xiao Ke) as the application object to fully demonstrate the construction process of the CTVS four-layer framework of the TCM classic meta-term engine: a) Construct a dual classification system (C-level) for TCM classical terminology to achieve a mapping and association between classical terminology and clinical terminology from ancient to modern times.

[0039] First, a classification system for classical terminology was constructed to determine the location path of "Xiao Ke" (wasting and thirsting disease): first-level category "diseases" → second-level category "disease names" → third-level category "Xiao Ke disease", and the classification code DJFL_JB_0103 was assigned. Synchronously map to clinical term classification: primary category "Qi, Blood, Body Fluid and Disease Terms" → secondary category "Consumptive Thirst Diseases" → tertiary category "Consumptive Thirst", code LCFL_QXJY_7.9; Establish an ancient and modern mapping table, and record the mapping relationship as an equivalent mapping.

[0040] (b)Extract core meta-terms through semi-supervised learning and manual annotation to complete the meta-term extraction and normalization T layer.

[0041] Use the BERT-BiLSTM-CRF joint model to process classical texts, and extract core meta-terms including "Consumptive Thirst" (high confidence), "Consumptive Heat" (low confidence), etc. For terms with low confidence, confirm through expert review that "Consumptive Heat" is a synonym of "Consumptive Thirst"; Execute term normalization: eliminate the ambiguity of "Consumptive Thirst" as a disease name and symptom description through dependency analysis, and determine the standardized term as "Consumptive Thirst (disease)"; establish a synonym mapping "Consumptive Heat → Consumptive Thirst"; generate a hierarchical relationship "Qi, Blood, Body Fluid Diseases (hypernym) → Consumptive Thirst → Upper Consumptive Thirst / Middle Consumptive Thirst / Lower Consumptive Thirst (hyponyms)".

[0042] (c)Construct a dynamic vocabulary V layer including 16 core data items for the standardized terms.

[0043] Create a complete dynamic vocabulary for the standardized term "Consumptive Thirst", including 16 core data: V-1 Serial number: 156 V-2 Primary classification: Diseases V-3 Secondary classification: Disease names V-4 See classification: Symptom - Thirst V-5 Meta-term ID: T2024_0156 V-6 Term word: Consumptive Thirst V-7 Pinyin: xiao ke V-8 Definition: Refers to a class of diseases caused by excessive consumption of fatty and sweet foods, or extreme emotions, excessive sexual intercourse, or damage by warm-heat pathogens, or excessive use of minerals and drugs, resulting in dryness of gastric fluid due to stomach heat, or dryness transformed from lung heat, excessive heart fire, or damage to kidney yin, resulting in abnormal qi transformation and involuntary leakage of body fluid essence, characterized by excessive drinking, excessive eating, and excessive urination V-9 Variant characters: 痟渴 V-10 Synonyms: Consumptive Heat, Lung Consumption, Upper Consumptive Thirst V-11 Near synonyms: Thirst, Excessive Drinking V-12 Hypernym: Qi, Blood, Body Fluid Diseases V-13 Hyponyms: Upper Consumptive Thirst, Middle Consumptive Thirst, Lower Consumptive Thirst V-14 Related words: Yin deficiency and dryness-heat, Liuwei Dihuang Pills, Coptis chinensis V-15 Modern medical terms: Type 2 diabetes (ICD-11: 5A11) V-16 is sourced from the classic text: *Synopsis of Prescriptions of the Golden Chamber*, Volume 2, Chapter 13: Diabetes Mellitus, Dysuria, and Urinary Tract Diseases (Pulse, Symptoms, and Treatments). (d) Construct a semantic association network S layer of terms by refining the conceptual relationships at each level.

[0044] Based on dynamic vocabulary generation, the <subject, predicate, object> triplet is generated, including: hierarchical relationship <diabetes, is_a, qi, blood, body fluid disease>; attribute relationship <diabetes, pathogenesis, yin deficiency and dryness-heat>; and action relationship <Liuwei Dihuang Wan, treatment, kidney yin deficiency type diabetes> and <Coptis chinensis, contraindications, spleen and stomach deficiency-cold type diabetes>.

[0045] Example 2: Taking a user's query "integration of ancient and modern knowledge about coronary heart disease and angina pectoris" as an example, the following steps are used to achieve in-depth analysis of the user's question and generate an answer: 1. Question Analysis and Terminology Recognition: The user enters a natural language question: "Please query information on the integration of ancient and modern knowledge about coronary heart disease and angina pectoris, including the corresponding ancient disease name, pathogenesis, prescriptions, and modern medical explanations." Semantic analysis of the question: Identifying core terms: Through semantic role labeling and dependency analysis, the core keyword was accurately identified as "coronary heart disease and angina pectoris". Simultaneously, key limiting conditions were identified: "integration of ancient and modern knowledge", "ancient disease name", "pathogenesis", "prescription", and "modern medical explanation".

[0046] Identify query intent: Determine if the user's intent is a complex, multi-dimensional, and comprehensive query. This requires the system to not only perform simple term mapping but also provide a complete chain of connections from ancient times to the present in terms of disease names, theories, treatments, and modern scientific explanations.

[0047] 2. Terminology standardization and semantic expansion: The identified core term "coronary heart disease and angina pectoris" is input into the meta-term engine of traditional Chinese medicine classics for processing.

[0048] The engine first performs queries and mappings in the V layer (dynamic vocabulary): Terminology standardization: The "Modern Medical Terms (V-15)" field was queried to confirm that "coronary heart disease and angina pectoris" is a modern standard term. Subsequently, through the "Synonyms (V-10)" field and based on the ancient and modern terminology standardization mapping table generated in S1, the core standard meta-term corresponding to it in ancient books was located as "chest pain and heart pain".

[0049] Semantic expansion: The engine automatically queries the hypernym (V-12), hyponym (V-13), and related word (V-14) fields of the meta-term to perform semantic expansion.

[0050] By using the disambiguation markers and polysemous word associations pre-stored in the related word field, the differences in contextual usage between historical synonyms such as "heartache" and "heartache due to fainting" and "chest pain and heartache" were clarified.

[0051] By using predefined "concept clusters", the pathogenesis concepts related to "chest pain" such as "yang deficiency and yin excess" (derived from "Essential Prescriptions of the Golden Chamber"), "phlegm and turbidity obstruction", and "blood stasis in the heart" are included in the search scope.

[0052] At this point, the original query was standardized and expanded to: taking the ancient term "chest pain and heart pain" as the core, associating it with all its historical variants and core pathogenesis concepts, and searching for related "prescriptions", "traditional Chinese medicines" and their "modern medical explanations".

[0053] 3. Semantic query and reasoning: The normalized query is submitted to the S layer (semantic association network) for graph pattern matching and reasoning. Using "chest pain" and "coronary heart disease angina" as dual central nodes, the following deep association paths are traversed and integrated: Mapping path between ancient and modern disease names: Directly obtain the predefined "equivalent mapping" relationship between "chest pain and heart pain" and "coronary heart disease and angina pectoris".

[0054] The pathogenesis-pathology mapping path: traversing from "chest pain and heart pain" to its core pathogenesis "yang deficiency and yin excess", and then the semantic relationship of the modern pathophysiological nodes "coronary atherosclerosis" and "myocardial ischemia" associated with this pathogenesis.

[0055] The path of diagnosis and treatment: traversing from "chest pain and heart pain" to its treatment methods "activating blood circulation and removing blood stasis" and "relieving pain and promoting yang", and then to the "treated" relationship of representative formulas "Gualou Xiebai Banxia Decoction" and "Xuefu Zhuyu Decoction".

[0056] Symptom-sign association pathway: Traversing the relationship between the symptom description of "chest pain radiating to the back" in ancient books and the "manifestations" of the modern clinical examination indicator "ST segment depression on electrocardiogram".

[0057] Biomarker association pathway: Traversing the "corresponding marker" relationship from the TCM pathogenesis of "phlegm and blood stasis obstructing each other" to modern biomarkers such as "high-sensitivity C-reactive protein" and "platelet aggregation rate".

[0058] Based on the traversal results of each associated path, retrieve the answer information related to the query intent.

[0059] 4. Answer generation and output: The answer information is integrated, sorted, and structurally assembled. The sorting criteria include: the semantic relevance to core terms, the authority of the evidence, and its relevance to modern clinical practice.

[0060] Finally, the following structured, integrated answer is generated and returned: Integration of ancient and modern disease names: Modern diseases: Coronary heart disease and angina pectoris Corresponding ancient disease name: Chest pain and heart pain (core meta-term) Historical alternative names (aliases): Heartache, sudden heartache, fainting heartache (the distinction should be made based on the context). Classification Code: Ancient Books System DJFL_JB_0103 / Clinical System LCFL_ZFB_5.1.1 Integration of theory and pathogenesis: Core pathogenesis: Yang deficiency and Yin excess (heart Yang deficiency, Yin cold and phlegm turbidity rising upwards) Modern pathological explanation: Coronary atherosclerotic stenosis or spasm leads to insufficient blood supply to the myocardium.

[0061] Integration of treatment methods and prescriptions: Common treatment methods: relieve pain and promote yang, invigorate blood circulation and remove blood stasis.

[0062] Recommended classical formulas: Gualou Xiebai Banxia Decoction and Xuefu Zhuyu Decoction.

[0063] Commonly used Chinese herbs: Trichosanthes kirilowii fruit, Allium macrostemon, Pinellia ternata, Prunus persica kernel, Carthamus tinctorius, Ligusticum chuanxiong.

[0064] Symptom and indicator integration: Typical symptoms described in ancient texts: "chest and back pain" and "heart pain radiating to the back, back pain radiating to the heart".

[0065] Modern corresponding signs: angina attack, abnormal ST-T segment changes on electrocardiogram.

[0066] The answer is ultimately presented to users in the form of a complete and interactive "integration of ancient and modern knowledge" map. The map clearly shows the mapping from "coronary heart disease and angina pectoris" to "chest pain and heart pain", and radiates out the ancient and modern nodes and connections of related pathogenesis, prescriptions, and symptoms, enabling users to gain a systematic and in-depth understanding of this issue.

[0067] like Figure 5 As shown, this embodiment of the invention also provides an intelligent question-answering system based on a meta-terminology engine from traditional Chinese medicine classics. The system includes: The parsing module 510 is used to receive natural language questions input by users, perform semantic parsing on them, identify and extract core TCM terms and user query intent from the questions; The processing module 520 is used to input the core TCM terms into a pre-built TCM classic meta-term engine for processing, and to complete the standardized mapping and semantic association expansion of the terms. The TCM classic meta-term engine is built on the classification-term-vocabulary-semantic CTVS association framework, including a TCM classic term dual classification system (ancient and modern) layer C, a meta-term extraction and standardization layer T, a dynamic vocabulary layer V, and a semantic association network layer S. The retrieval module 530 is used to perform graph pattern matching and reasoning in the semantic association network S layer of the TCM classic meta-term engine based on the terms after normalization mapping and semantic association expansion. With the classic meta-terms and corresponding modern clinical terms as dual center nodes, it traverses and integrates multiple deep association paths to retrieve answer information related to the query intent. The generation module 540 is used to integrate, sort and structure the answer information to generate a structured answer, and return it to the user for visualization.

[0068] The intelligent question-answering system based on the meta-terminology engine of traditional Chinese medicine classics provided in this embodiment of the invention has a functional structure that corresponds to the intelligent question-answering method based on the meta-terminology engine of traditional Chinese medicine classics provided in this embodiment of the invention, and will not be described again here.

[0069] Figure 6 This is a schematic diagram of the structure of an electronic device 600 provided in an embodiment of the present invention. The electronic device 600 may vary considerably due to different configurations or performance. It may include one or more central processing units (CPUs) 601 and one or more memories 602. The memory 602 stores at least one instruction, which is loaded and executed by the processor 601 to implement the steps of the above-mentioned intelligent question-answering method based on the meta-terminology engine of traditional Chinese medicine classics.

[0070] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including instructions that can be executed by a processor in a terminal to complete the intelligent question-answering method based on a meta-terminal engine of traditional Chinese medicine classics. For example, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0071] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0072] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent question and answer method based on traditional Chinese medicine classic book element term engine, characterized in that, The method comprises: S1, receiving a natural language question sentence input by a user, performing semantic analysis on the question sentence, and identifying and extracting core traditional Chinese medicine terms and user query intentions in the question sentence; S2, inputting the core traditional Chinese medicine terms into a pre-constructed traditional Chinese medicine book meta-term engine for processing, completing standardized mapping and semantic association expansion of the terms, and the traditional Chinese medicine book meta-term engine is constructed based on a classification-term-vocabulary-semantic (CTVS) association framework, including a traditional Chinese medicine book term ancient and modern double classification system C layer, a meta-term extraction and standardization T layer, a dynamic vocabulary V layer, and a semantic association network S layer; S3, according to the standardized mapping and semantic association expansion of the terms, performing graph pattern matching and reasoning in the semantic association network S layer of the traditional Chinese medicine book meta-term engine, traversing and integrating multiple depth association paths with the dual center nodes of the book meta-term and the corresponding modern clinical term, and retrieving answer information associated with the query intention; S4, integrating, sorting and structuring the answer information, generating a structured answer, and returning it to the user for visual presentation.

2. The method of claim 1, wherein, The pre-construction of the traditional Chinese medicine book meta-term engine specifically comprises: (a) Constructing a traditional Chinese medicine book term ancient and modern double classification system C layer: Establish a traditional Chinese medicine book term classification system including multiple categories, and generate a standardized mapping table of book terms and clinical terms based on the current national standard for traditional Chinese medicine clinical diagnosis and treatment, to realize the mapping association of ancient and modern terms; (b) Meta-term extraction and standardization T layer: From the traditional Chinese medicine book text, extract core meta-terms using a natural language processing model, and perform term standardization based on the mapping table obtained in step (a), a pre-constructed alias-standard name correspondence table, and a characteristic alias-standard name detection model; (c) Constructing a dynamic vocabulary V layer: Construct a dynamic vocabulary for the terms standardized in step (b), which includes multiple core data items for describing the multi-dimensional attributes of the terms; (d) Constructing a semantic association network S layer: Based on the classification and relationship data in the dynamic vocabulary obtained in step (c), construct a semantic association network of the terms; the steps (a), (b), (c), (d) are associated in sequence to cooperatively construct the traditional Chinese medicine book meta-term engine.

3. The method of claim 2, wherein, The step (a) of constructing the traditional Chinese medicine book term ancient and modern double classification system C layer specifically comprises: (i) Constructing a traditional Chinese medicine book term classification system, including 15 first-level categories: basic, treatment method, medicine, prescription, disease, symptom, syndrome, acupuncture, massage, health preservation, medical book, medical expert, time, space, and others, and expanding 64 second-level categories thereunder; (ii) Establishing a mapping association of traditional Chinese medicine clinical diagnosis and treatment terms based on the current national standards GB / T 16751.1-2023 "Traditional Chinese Medicine Clinical Diagnosis and Treatment Terms Part 1: Diseases", GB / T 16751.2-2021 "Traditional Chinese Medicine Clinical Diagnosis and Treatment Terms Part 2: Syndromes", and GB / T 16751.3-2023 "Traditional Chinese Medicine Clinical Diagnosis and Treatment Terms Part 3: Treatment Methods"; (iii) generating a standardization mapping table including the defined canonical terms and clinical terms as the framework basis for the semantic conversion of the terms.

4. The method of claim 2, wherein, The extraction and standardization of the meta-terms in step (b) is achieved by a multi-strategy fusion processing method, including the following steps: (i) initialization construction: based on the pre-set core dictionary of traditional Chinese medicine field and the standardization mapping table generated in step (a), an initial term seed set is constructed, and an alias-standard name mapping library is constructed according to the pre-constructed alias-standard name correspondence table and the characteristic alias-standard name detection model; (ii) multi-strategy collaborative extraction: the sequence labeling model based on active learning is used to preliminarily screen the target classics text, and the candidate terms with high confidence are automatically obtained; at the same time, the pattern matching method based on semantic role and dependency grammar analysis is used to extract the terms and their associated information from the specific sentence structure of the classics original text; (iii) human-computer collaborative labeling and optimization: the low-confidence results and conflict results in the above steps are manually checked and labeled by domain experts; the manual labeling results are fed back to the sequence labeling model as training samples, and the rule library of pattern matching is updated to realize the iterative optimization of the model and the self-improvement of the labeling efficiency; (iv) knowledge-guided term normalization: for the extracted candidate terms, first query the alias-standard name mapping library for synonym merging; for the terms that cannot be matched, the semantic similarity in the vector space and the context dependence path features in the classics are calculated to make a comprehensive judgment, and the disambiguation and standardization mapping of the terms are completed.

5. The method of claim 2, wherein, The alias-standard name correspondence table includes the standard name of each term and the corresponding various conventional aliases; The characteristic alias-standard name detection model is used to further calculate the potential aliases of each term, and the steps are as follows: Construct a feature set: for each term Wi, construct a basic feature set Fwi, which includes corresponding structured attribute information, Fwi={c1,c2,c3,...}, c1,c2,c3 are attribute lists; Calculate the intersection: the number of features shared by two terms |Fwa∩Fwb|=Fshare; Union: the total number of all features of two terms |Fwa∪Fwb|=Fall; Symmetric difference: the union-set-intersection of two term sets, Fall-Fshare=Fsym; Total number: FWa+FWb=Fnum; Perform the calculation: Sim(Wa, Wb) = ((2 Fshare) / Fnum)-(Fsym / Fall), where Sim(Wa, Wb) represents the similarity of the two terms, the first part ((2 Fshare) / Fnum) emphasizes the importance of common features, the more common features, the higher the similarity, and the second part -(Fsym / Fall) slightly punishes the mismatched part, making the result more balanced, when calculating the similarity, this part value will be subtracted, so that the similarity score is slightly reduced, ensuring that even if two terms have many common features, the difference between them will also be reflected in the score, which helps to avoid false classification or too high similarity score due to ignoring differences; Similarity normalization: normalize the similarity results of two terms: Sim(Wa,Wb)=max(0,min(1,Sim(Wa,Wb))), to ensure that the similarity score is between [0, 1]; Result output: output the two terms with a similarity score higher than the preset threshold as the standard name and the corresponding alias.

6. The method of claim 2, wherein, The dynamic vocabulary in step (c) includes 16 core data items: V-1 serial number; V-2 primary classification; V-3 secondary classification; V-4 refer to classification; V-5 meta-term ID; V-6 term word; V-7 pinyin; V-8 definition; V-9 variant; V-10 synonym; V-11 near synonym; V-12 hypernym; V-13 hyponym; V-14 associated word; V-15 modern medical term; V-16 source classic.

7. The method of claim 2, wherein, The step of constructing a semantic association network in step (d) comprises: (ii) parsing the relationship field in the dynamic vocabulary to directly extract the pre-defined hierarchical relationship; (iii) analyzing the definition field in the dynamic vocabulary and the context of the original text of the term to extract the attribute relationship and role relationship of the term; (iv) calculating and optimizing the confidence of the extracted various relationships to form a semantic association network rich in multiple relationship types.

8. The method of claim 1, wherein, In S3, with the dual center nodes of the classic meta-term and the corresponding modern clinical term, multiple depth association paths are traversed and integrated to retrieve the answer information associated with the query intent, specifically including: In the semantic association network S layer, with the dual center nodes of the classic meta-term and the corresponding modern clinical term, the following depth association paths are traversed and integrated: 1) Ancient and modern disease name mapping path: directly obtaining the pre-defined equivalent mapping relationship between the classic meta-term and the corresponding modern clinical term; 2) Pathogenesis-pathology mapping path: traversing the semantic relationship from the classic meta-term to its core pathogenesis, and then from the pathogenesis to the modern pathophysiological node associated with it; 3) Syndrome-treatment association path: traversing from the classic meta-term to various treatments, and then to various representative prescription treatment relationships; 4) Symptom-sign association path: traversing from the symptom description in ancient books to the performance relationship of modern clinical examination indicators; 5) Biomarker association path: traversing from the TCM pathogenesis to the corresponding biomarker relationship of various modern biomarkers; According to the traversal results of each association path, the answer information associated with the query intent is retrieved.

9. An intelligent question and answer system based on a traditional Chinese medicine textbook meta-terminology engine, characterized in that, The system comprises: An analysis module for receiving a natural language question input by a user, performing semantic analysis on the question, and identifying and extracting core TCM terms and user query intent in the question; A processing module for inputting the core TCM terms into a pre-constructed TCM classic meta-term engine for processing, completing the standardized mapping and semantic association expansion of the terms, wherein the TCM classic meta-term engine is constructed based on a classification-term-vocabulary-semantic (CTVS) association framework, including a C layer of ancient and modern dual classification system of TCM classic terms, a T layer of meta-term extraction and standardization, a V layer of dynamic vocabulary, and a S layer of semantic association network; A retrieval module for performing graph pattern matching and reasoning in the semantic association network S layer of the TCM classic meta-term engine based on the standardized mapping and semantic association expanded terms, with the dual center nodes of the classic meta-term and the corresponding modern clinical term, traversing and integrating multiple depth association paths, and retrieving the answer information associated with the query intent; A generation module for integrating, sorting, and structuring the answer information, generating a structured answer, and returning it to the user for visual presentation. 10.An electronic device, comprising a processor and a memory having stored therein at least one instruction, wherein, The at least one instruction is loaded and executed by the processor to realize the intelligent question and answer method based on the traditional Chinese medicine classic book element term engine according to any one of claims 1-8.

Images