A method and device for constructing a knowledge graph in the field of Pu'er tea

Abstract

The present invention discloses a method and device for constructing a knowledge graph in the field of Pu'er tea, which relates to the technical field of Pu'er tea. The method comprises: establishing a first-level knowledge ontology classification model in the field of Pu'er tea; establishing a second-level knowledge ontology classification model in the field of Pu'er tea; establishing a Pu'er tea field knowledge entity model associated with the first and second-level knowledge ontology classification models, for constructing a knowledge graph in the field of Pu'er tea. The present invention adopts a construction method that expands from the center to the periphery, performs two-level knowledge ontology classification on Pu'er tea field knowledge, extracts entity models in combination with the knowledge ontology classification model feature vocabulary, and associates them with the knowledge ontology classification model to form a knowledge graph in the field of Pu'er tea. The present invention can manage the complex relationships between entity models through knowledge ontology classification, reduce the dependence on professionals in the process of incremental knowledge extraction, and improve the accuracy of knowledge modeling.

Description

Technical Field

[0001] The present invention relates to the technical field of Pu'er tea, and in particular to a method and device for constructing a knowledge graph in the field of Pu'er tea. Background Art

[0002] Knowledge graphs describe concepts, entities, and their relationships in the objective world in a structured manner, expressing internet information in a form closer to human cognition and providing a way to better organize, manage, and understand the vast amount of information on the internet. Key technologies for knowledge graphs include knowledge acquisition, knowledge representation, knowledge storage, knowledge modeling, knowledge fusion, knowledge understanding, and knowledge operations. By constructing knowledge graphs for structured, semi-structured, and unstructured data, they support applications in diverse fields.

[0003] The knowledge graph is a vast data model that can construct a vast "knowledge" network encompassing numerous entities, attributes, and relationships in the objective world, providing a fast and convenient way for people to retrieve and reason about knowledge. The recent boom in artificial intelligence is essentially a knowledge revolution, centered on observing and perceiving the world through data to enable intelligent services such as classification, prediction, and automation. As a crucial vehicle for describing human knowledge, the knowledge graph is driving numerous intelligent applications, including information retrieval and intelligent question-answering.

[0004] Companies across the Pu'er tea industry chain generally have weak information technology and automation foundations. Constrained by cost pressures, technological transformation risks, and current technological capabilities, their accumulation of data and information, as well as their understanding and application of domain knowledge, are currently at a nascent stage. Looking at the entire Pu'er tea industry chain, the knowledge and information involved in every stage of the entire Pu'er tea lifecycle—from cultivation, primary processing, fine processing, deep processing, product sales, distribution, and use—is dispersed among various processing companies, sales and distribution companies, research institutes, universities, and end users. Consequently, the comprehensive utilization of data and information is low. The value of knowledge is not yet fully reflected in driving the Pu'er tea industry.

[0005] Therefore, it is an urgent problem to study the classification expression and modeling methods of various types of knowledge in the Pu'er tea industry chain, construct a knowledge graph in the field of Pu'er tea, and gradually realize the value-added of each link in the Pu'er tea industry chain through the application of knowledge graphs, and promote the digital transformation of enterprises. Summary of the Invention

[0006] The Pu'er tea domain knowledge graph is a domain-specific knowledge graph with specialized characteristics. Its application scenarios and services are primarily targeted at specific businesses and groups. Professional knowledge graphs are typically constructed using a top-down approach, with layer-by-layer modeling and refinement. Their core lies in the abstraction of various knowledge ontologies and the logical expression of their relationships.

[0007] The construction of the knowledge graph in the field of Pu'er tea first requires the establishment of a knowledge ontology model and framework that conforms to the characteristics of the industry to ensure the accuracy of the description of each knowledge ontology. Then, by integrating structured and unstructured data to identify the knowledge entity model, and through incremental model construction, gradually establish a knowledge graph model in the field of Pu'er tea to support various knowledge application scenarios.

[0008] To achieve the above objectives, the present invention proposes a method and device for constructing a knowledge graph in the field of Pu'er tea. The construction method specifically includes the following steps:

[0009] S10, establish the first-level knowledge ontology classification model in the field of Pu'er tea;

[0010] S20, establish the second-level knowledge ontology classification model in the field of Pu'er tea;

[0011] S30, establishing a Pu'er tea domain knowledge entity model associated with the first and second layer knowledge ontology classification models, for constructing a Pu'er tea domain knowledge graph.

[0012] Preferably, the step S10 of establishing the first-level knowledge ontology classification model in the field of Pu'er tea specifically includes the following steps:

[0013] S101, creating a root node of the Pu'er tea domain knowledge graph, where the root node of the Pu'er tea domain knowledge graph is Pu'er tea;

[0014] S102, based on public information or data such as national standards, industry standards, and academic papers, establish the first-level knowledge ontology classification nodes in the field of Pu'er tea. The first-level knowledge ontology classification nodes in the field of Pu'er tea include: product, enterprise, brand, ingredient, origin, variety category, and processing and manufacturing;

[0015] S103, for each first-level knowledge ontology classification node, a feature word library is established. The feature word library is attached to the corresponding first-level knowledge ontology classification node in the form of a triple structure of the first-level knowledge ontology classification node, attribute, and attribute value;

[0016] S104, using a star structure, connecting the first-level knowledge ontology classification node to the root node, and establishing a relationship between the root node and the first-level knowledge ontology classification node; the star structure means that each first-level knowledge ontology classification node is only associated with the root node, and the knowledge ontology classification nodes are not associated with each other;

[0017] S105, constructing a first-level knowledge ontology classification model in the field of Pu'er tea, which is composed of a root node, a relationship, and a triple model of first-level knowledge ontology classification nodes.

[0018] Preferably, the step S20 of establishing the second-level knowledge ontology classification model in the field of Pu'er tea specifically includes the following steps:

[0019] S201, for each knowledge domain involved in the first-level knowledge ontology classification node in the field of Pu'er tea, determine whether it is necessary to establish a second-level knowledge ontology classification model;

[0020] S202, when it is necessary to establish a second-level knowledge ontology classification model, a second-level knowledge ontology classification node belonging to the first-level knowledge ontology classification node in the field of Pu'er tea is established based on national standards, industry standards, and public information or data of academic papers;

[0021] S203, establishing a feature word library for each second-level knowledge ontology classification node. The feature word library is attached to the corresponding second-level knowledge ontology classification node in the form of a triple structure of the second-level knowledge ontology classification node, attribute, and attribute value;

[0022] S204, connecting the second-layer knowledge ontology classification nodes with the corresponding first-layer knowledge ontology classification nodes to form a directed graph of the network knowledge ontology classification structure in the field of Pu'er tea;

[0023] S205, constructing a second-layer knowledge ontology classification model for the Pu'er tea field consisting of first-layer knowledge ontology classification nodes, relationships, and a second-layer knowledge ontology classification node triple model.

[0024] The first-level and second-level knowledge ontology classification node feature word library in the field of Pu'er tea is mainly used to store a group of keywords, words, symbols and other contents in the classification field, and is used to judge the attribution and association relationship between the entity model and the ontology model when extracting the entity model.

[0025] Preferably, the step S30 of establishing a Pu'er tea domain knowledge entity model associated with the first and second layer knowledge ontology classification models for constructing a Pu'er tea domain knowledge graph specifically includes the following steps:

[0026] S301, extracting a Pu'er tea domain knowledge entity model, using a machine learning algorithm to extract a Pu'er tea domain knowledge entity model from existing semi-structured and unstructured text data. The Pu'er tea domain knowledge entity model consists of a triple structure of entity, attribute, and attribute value;

[0027] S302, associating the Pu'er tea domain knowledge entity model with the ontology model, associating the Pu'er tea domain knowledge entity model extracted in step S301 with one or more nodes of the first-layer and second-layer knowledge ontology classification models to form a Pu'er tea domain knowledge graph;

[0028] S303, storing: storing the Pu'er tea domain knowledge ontology classification model and the Pu'er tea domain knowledge entity model data into a graph database;

[0029] S304, operation and maintenance, includes: maintaining the first-level and second-level knowledge ontology classification models and feature lexicons in the field of Pu'er tea, adjusting the association between the knowledge entity model in the field of Pu'er tea and the knowledge ontology classification model in the field of Pu'er tea, and reviewing and revising the content of the knowledge entity model in the field of Pu'er tea.

[0030] The present invention also provides a Pu'er tea field knowledge graph construction device, comprising:

[0031] An entity model extraction module is used to extract the Pu'er tea domain knowledge entity model from semi-structured and unstructured texts to form a triple data structure consisting of the Pu'er tea domain knowledge entity model name, attribute and attribute value;

[0032] The first classification determination module is used to determine, through a feature word library, which classification the Pu'er tea domain knowledge entity model belongs to in the first-level knowledge ontology classification model of the Pu'er tea domain;

[0033] The second classification determination module is used to judge and determine whether the Pu'er tea domain knowledge entity model belongs to a certain classification of the second-level knowledge ontology classification model in the Pu'er tea domain through the feature vocabulary;

[0034] The model relationship construction module is used to establish the directed relationships between the first-level knowledge ontology classification model, the second-level knowledge ontology classification model, and the knowledge entity model in the Pu'er tea field;

[0035] The knowledge operation and maintenance module is used to store, update, and maintain knowledge graph data in the field of Pu'er tea.

[0036] The present invention also provides a partial structure of a computer, including a memory, a processor, a bus, and a computer program stored in the memory and run on the processor. The processor and the memory communicate via a bus. When the processor executes the computer program, the steps of the method for constructing a knowledge graph in the Pu'er tea field as described in any one of the above items are implemented.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] The method for constructing a Pu'er tea domain knowledge graph, provided by the present invention, adopts a construction method that expands from the center to the periphery, classifying Pu'er tea domain knowledge into two levels. A first-level knowledge ontology classification model for the Pu'er tea domain is first established. Then, based on the first-level knowledge ontology classification model, a second-level knowledge ontology classification model for the Pu'er tea domain is established. Entity models are extracted by combining the feature vocabulary of the ontology classification model and are associated with the ontology classification model to form a Pu'er tea domain knowledge graph. The present invention manages the complex relationships between entity models through knowledge ontology classification, reduces reliance on professionals during incremental knowledge extraction, and improves the accuracy of knowledge modeling. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the contents of the embodiments of the present invention and the structures shown in these drawings without any creative work.

[0040] Figure 1 This is a schematic diagram of the process of constructing a knowledge graph in the field of Pu'er tea according to the present invention;

[0041] Figure 2 A schematic diagram of the process of establishing the first-level knowledge ontology classification model in the field of Pu'er tea according to the present invention;

[0042] Figure 3 This is a schematic diagram of the structure of the first-level knowledge ontology classification model in the field of Pu'er tea of the present invention;

[0043] Figure 4 A schematic diagram of the process of establishing the second-level knowledge ontology classification model in the field of Pu'er tea according to the present invention;

[0044] Figure 5 This is a partial structural diagram of the second-level knowledge ontology classification model in the field of Pu'er tea of the present invention;

[0045] Figure 6 A schematic diagram of the process of establishing the knowledge entity model of Pu'er tea field of the present invention;

[0046] Figure 7 This is an example diagram of the knowledge graph in the field of Pu'er tea of the present invention;

[0047] Figure 8 This is a schematic diagram of the structure of the device for constructing the knowledge graph in the field of Pu'er tea according to the present invention;

[0048] Figure 9 This is a partial structural diagram of the computer provided by the present invention. DETAILED DESCRIPTION

[0049] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0050] In the description of the present invention, unless otherwise specified, "plurality" means two or more. Terms such as "inner," "upper," and "lower" indicating positions or states are based on those shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They do not indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and are therefore not to be construed as limiting the present invention.

[0051] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "provided with" should be understood broadly. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; and direct or indirect connections through an intermediary. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0052] The knowledge graph in the field of Pu'er tea is a limited-domain knowledge graph with domain-specific characteristics. It is usually constructed by expanding from the center to the periphery, modeling and refining layer by layer. Its core lies in the abstraction of various knowledge ontologies and the logical expression of relationships.

[0053] In the process of constructing the knowledge graph in the field of Pu'er tea, we first need to establish a knowledge ontology model and framework that conforms to the characteristics of the industry, and then identify the knowledge entity model by integrating structured and unstructured data, and gradually establish the knowledge graph model in the field of Pu'er tea through incremental model construction.

[0054] To this end, the present invention provides a method for constructing a knowledge graph in the field of Pu'er tea. Figure 1 As shown in FIG, the method for constructing a knowledge graph in the field of Pu'er tea includes the following steps:

[0055] S10, establish the first-level knowledge ontology classification model in the field of Pu'er tea. The process of establishing the first-level knowledge ontology classification model in the field of Pu'er tea is as follows: Figure 2 As shown, the specific steps include:

[0056] S101, creating a root node of the Pu'er tea domain knowledge graph, where the root node of the Pu'er tea domain knowledge graph is Pu'er tea;

[0057] S102, based on public information or data such as national standards, industry standards, and academic papers, establish the first-level knowledge ontology classification nodes in the field of Pu'er tea;

[0058] It should be noted that the knowledge ontology is mainly used to describe the concepts and relationships between concepts in a certain field, so that they have a commonly recognized, clear and unique definition within a shared scope;

[0059] The first-level knowledge ontology classification node in the Pu'er tea field needs to classify the knowledge in the Pu'er tea field. Each first-level knowledge ontology classification node corresponds to domain knowledge content with clear concepts and scope.

[0060] Specifically, in this embodiment, based on the analysis of public information or data, the first-level knowledge ontology classification nodes of Pu'er tea field are classified, abstracted, and constructed, including seven aspects: product, enterprise, brand, ingredient, origin, variety category, processing and manufacturing, such as Figure 3 As shown, where:

[0061] The product node is mainly a conceptual abstraction of all Pu'er tea products sold in the market;

[0062] Enterprise nodes are mainly conceptual abstractions of various enterprise entities related to the Pu'er tea industry chain, such as tea-making enterprises, tea-selling enterprises, and tea-making equipment manufacturers.

[0063] Brand nodes are mainly abstract brand concepts of Pu'er tea or Pu'er tea-related products recognized by the market;

[0064] The ingredient nodes are mainly conceptual abstractions of various substances or ingredients contained in Pu'er tea;

[0065] The origin node is mainly the conceptual abstraction of the Pu'er tea geographical indication product protection area specified in the national standard "Geographical Indication Product Pu'er Tea" (GBT 22111-2008);

[0066] Variety category nodes are mainly abstract concepts of Pu'er tea products and Pu'er tea tree species;

[0067] The processing and manufacturing nodes are mainly conceptual abstractions of different processing techniques or processes for Pu'er tea;

[0068] S103, establishing a feature word library for each first-level knowledge ontology classification node. The feature word library is attached to the corresponding first-level knowledge ontology classification node in the form of a triple structure of the first-level knowledge ontology classification node, attribute, and attribute value;

[0069] In this embodiment, the feature word library of the first-level knowledge ontology classification node belongs to the category of domain feature word library in the field of natural language technology. Specifically, it refers to a set of words that are strongly related to the field and have the ability to distinguish domain knowledge. For example, in the field of Pu'er tea, "Pu'er tea", "Iceland", "Ancient Tree Tea", "Menghai", "Brown Mountain" and other common words can exist as classification features.

[0070] Table 1 is a schematic diagram of the structural relationship of the first-level knowledge ontology classification node feature word library of this embodiment:

[0071] Table 1

[0072]

[0073] As shown in Table 1, the feature word library of the knowledge ontology classification node "ingredients" consists of a set of feature word attributes and feature word attribute values. According to needs, the feature word attributes include sentences, terms, nouns, verbs, symbols, etc.; the feature word attribute values are composed of keywords, keywords, and proprietary names in the knowledge domain to which the knowledge ontology classification node belongs. In this embodiment, the feature words related to the knowledge ontology classification node "ingredients" are mainly indicators, substances, units, and other elements related to the physical and chemical testing of Pu'er tea. The feature word library is mainly used to determine the attribution and association relationship between the entity model and the ontology model when extracting the knowledge entity model.

[0074] S104, using a star structure, connect the first-level knowledge ontology classification node with the root node, and establish a relationship between the root node and the first-level knowledge ontology classification node, such as Figure 3 As shown; the star structure means that each first-level knowledge ontology classification node is only associated with the root node, and the knowledge ontology classification nodes are not associated with each other;

[0075] It should be noted that the star-structure connection method adopted in this embodiment can ensure the independence of knowledge ontology classification, facilitate the management of secondary ontology classification models and entity models corresponding to knowledge fields of classification nodes, reduce the complexity of knowledge graph semantic network, and avoid knowledge ambiguity;

[0076] S105, constructing a first-level knowledge ontology classification model for the Pu'er tea field consisting of a root node, a relationship, and a triplet model of the first-level knowledge ontology classification node. The first-level knowledge ontology classification model for the Pu'er tea field constructed in this embodiment is as follows: Figure 3 As shown;

[0077] S20, establish the second-level knowledge ontology classification model in the field of Pu'er tea. The establishment process of the second-level knowledge ontology classification model in the field of Pu'er tea is as follows: Figure 4 As shown, the specific steps include:

[0078] S201, for each knowledge domain involved in the first-level knowledge ontology classification node in the field of Pu'er tea, determine whether it is necessary to establish a second-level knowledge ontology classification model;

[0079] It should be noted that, according to the evolution of the knowledge graph construction process and the application requirements of the knowledge graph, under the first-level knowledge ontology classification node, if its knowledge domain is relatively simple or too complex, it is not necessary to build the second-level knowledge ontology classification temporarily, and divide it after the knowledge entity model accumulates to a certain scale;

[0080] S202, when it is necessary to establish a second-level knowledge ontology classification model, a second-level knowledge ontology classification node belonging to the first-level knowledge ontology classification node in the field of Pu'er tea is established based on public information or data such as national standards, industry standards, and academic papers;

[0081] Specifically, Table 2 is a list 2 of the relationship between the first-layer knowledge ontology classification nodes and the second-layer knowledge ontology classification nodes in this embodiment:

[0082] Table 2

[0083]

[0084] As shown in Table 2, based on the first-level knowledge ontology classification nodes, the second-level knowledge ontology classification nodes are constructed to manage and maintain the knowledge entity models under them;

[0085] S203, for each second-level knowledge ontology classification node, a corresponding feature word library is established. The feature word library is attached to the corresponding second-level knowledge ontology classification node in the form of a triple structure of second-level knowledge ontology classification node, attribute and attribute value, to construct a second-level knowledge ontology classification model for Pu'er tea field, such as Figure 5 The following is a partial structural diagram of the second-level knowledge ontology classification model in the field of Pu'er tea;

[0086] S204, connecting the second-layer knowledge ontology classification nodes with the corresponding first-layer knowledge ontology classification nodes to form a directed graph of the network knowledge ontology classification structure in the field of Pu'er tea, such as Figure 7 , as shown in Table 2;

[0087] S205, constructing a second-layer knowledge ontology classification model for the Pu'er tea field consisting of a first-layer knowledge ontology classification node, a relationship, and a second-layer knowledge ontology classification node triple model;

[0088] S30, establishing a Pu'er tea domain knowledge entity model associated with the first and second level knowledge ontology classification models, for constructing a Pu'er tea domain knowledge graph. The Pu'er tea domain knowledge entity model establishment process is as follows: Figure 6 As shown, the specific steps include:

[0089] S301, extracting a Pu'er tea domain knowledge entity model, using a machine learning algorithm to extract a Pu'er tea domain knowledge entity model from existing semi-structured and unstructured text data. The Pu'er tea domain knowledge entity model consists of a triple structure of entity, attribute, and attribute value;

[0090] S302, the Pu'er tea domain knowledge entity model is associated with the ontology model, and the Pu'er tea domain knowledge entity model is associated with multiple nodes of the first layer and the second layer knowledge ontology classification model to form the Pu'er tea domain knowledge graph, such as Figure 7 As shown; such as:

[0091] The first-level knowledge ontology classification node - origin is connected to the second-level knowledge ontology classification nodes: Pu'er tea area, Xishuangbanna tea area, Lincang tea area, Baoshan tea area;

[0092] The first-level knowledge ontology classification node - ingredient is connected to the second-level knowledge ontology classification nodes: dry tea, aroma, and water extract. Among them, dry tea includes ash, refined fiber, and water; aroma includes linalool, methoxybenzene, benzaldehyde, ionone, and soluble total sugar; water extract includes soluble total sugar, catechins, soluble protein, caffeine, free amino acids, and tea polyphenols.

[0093] The first-level knowledge ontology classification node - enterprise is connected to the second-level knowledge ontology classification nodes: industrial chain configuration enterprises, tea machinery and equipment manufacturing enterprises, Pu'er tea deep processing enterprises, Pu'er tea fine processing enterprises, Pu'er tea primary processing enterprises;

[0094] The first-level knowledge ontology classification node - brand is connected to the second-level knowledge ontology classification node: brand A;

[0095] The first-level knowledge ontology classification node - product is connected to the second-level knowledge ontology classification nodes: Pu'er tea products, Pu'er tea deep-processing products, Pu'er tea production equipment;

[0096] The first-level knowledge ontology classification node - processing and manufacturing is connected to the second-level knowledge ontology classification nodes: sun-dried green tea, Pu'er tea (raw tea) compressed tea, Pu'er tea (cooked tea) loose tea, Pu'er tea (cooked tea) compressed tea;

[0097] The first-level knowledge ontology classification node - Variety Category is connected to the second-level knowledge ontology classification nodes: Ancient Tree Tea, Large Tree Tea, Terrace Tea, Pu'er Tea Compacted Tea, Pu'er Tea Loose Tea; Among them: Pu'er Compacted Tea includes Raw Pu'er and Cooked Pu'er; Pu'er Loose Tea includes Pu'er Black Tea, Pu'er Black Tea, Pu'er Oolong Tea, Pu'er White Tea, Pu'er Yellow Tea, Pu'er Green Tea;

[0098] S303, storing the Pu'er tea domain knowledge ontology classification model and knowledge entity model data into the graph database;

[0099] S304, operation and maintenance, includes: maintaining the first-level and second-level knowledge ontology classification models and feature lexicons in the field of Pu'er tea, adjusting the association between the knowledge entity model in the field of Pu'er tea and the knowledge ontology classification model in the field of Pu'er tea, and reviewing and revising the content of the knowledge entity model in the field of Pu'er tea.

[0100] Based on the above embodiments, the present invention provides a device for constructing a knowledge graph in the field of Pu'er tea, such as Figure 8 As shown, including the following settings in the computer:

[0101] The entity model extraction module 401 is used to extract the Pu'er tea domain knowledge entity model from the semi-structured and unstructured text to form a triple data structure consisting of the Pu'er tea domain knowledge entity model name, attribute and attribute value;

[0102] The first classification determination module 402 is used to determine, through the feature word library, which classification of the first-level knowledge ontology classification model of the Pu'er tea field the Pu'er tea field knowledge entity model belongs to;

[0103] The second classification determination module 403 is used to judge and determine whether the Pu'er tea domain knowledge entity model belongs to a certain classification of the second-level knowledge ontology classification model in the Pu'er tea domain through the feature word library;

[0104] The model relationship building module 404 is used to establish a directed relationship between the first-level knowledge ontology classification model, the second-level knowledge ontology classification model, and the knowledge entity model in the Pu'er tea field;

[0105] The knowledge operation and maintenance module 405 is used to store, update, and maintain the knowledge graph data in the field of Pu'er tea.

[0106] The knowledge graph in the field of Pu'er tea constructed by this device is as follows: Figure 7 shown.

[0107] Figure 9 Figure 5 is a schematic diagram of the partial structure of a computer provided by the present invention, which includes: a processor 501, a memory 502, a communication interface 503, a bus 504, and a computer program stored in the memory 502 and executable by the processor 501. The processor 501, the memory 502, and the communication interface 503 communicate with each other via the bus 504. The communication interface 503 processes external instructions, triggering the processor 501 to call the computer program instructions stored in the memory 502 to execute all steps of the method for constructing a knowledge graph in the field of Pu'er tea according to the above-described embodiment of the present invention.

[0108] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A device for constructing a knowledge graph in the field of Pu'er tea, characterized in that: Using the knowledge graph construction method in the field of Pu'er tea, S10, establish the first-level knowledge ontology classification model in the field of Pu'er tea; S20, establish the second-level knowledge ontology classification model in the field of Pu'er tea; S30, establishing a Pu'er tea domain knowledge entity model associated with the first and second layer knowledge ontology classification models, for constructing a Pu'er tea domain knowledge graph; The step S10 of establishing the first-level knowledge ontology classification model in the field of Pu'er tea specifically includes the following steps: S101, creating a root node of the Pu'er tea domain knowledge graph, which is Pu'er tea; S102: Based on the information or data published in national standards, industry standards, and academic papers, establish the first-level knowledge ontology classification nodes in the field of Pu'er tea. The first-level knowledge ontology classification nodes include: product, enterprise, brand, ingredient, origin, variety category, and processing and manufacturing; S103, establishing a feature word library based on each first-level knowledge ontology classification node in the field of Pu'er tea, wherein the feature word library is attached to the corresponding first-level knowledge ontology classification node in the form of a triple structure of first-level knowledge ontology classification node, attribute, and attribute value; S104, using a star structure, connecting the first-layer knowledge ontology classification nodes with the root node, and establishing a relationship between the root node and the first-layer knowledge ontology classification nodes; S105, constructing a first-level knowledge ontology classification model for the Pu'er tea field consisting of a root node, a relationship, and a triplet model of first-level knowledge ontology classification nodes; The step S20 of establishing the second-level knowledge ontology classification model in the field of Pu'er tea specifically includes the following steps: S201, determining whether it is necessary to establish a second-level knowledge ontology classification model for each knowledge domain involved in the first-level knowledge ontology classification node in the field of Pu'er tea; S202, when it is necessary to establish a second-level knowledge ontology classification model, a second-level knowledge ontology classification node belonging to the first-level knowledge ontology classification node in the field of Pu'er tea is established based on national standards, industry standards, and public information or data of academic papers; S203, establishing a feature word library for each second-level knowledge ontology classification node. The feature word library is attached to the corresponding second-level knowledge ontology classification node in the form of a triple structure of the second-level knowledge ontology classification node, attribute, and attribute value; S204, connecting the second-layer knowledge ontology classification nodes with the corresponding first-layer knowledge ontology classification nodes to form a directed graph of the network knowledge ontology classification structure in the field of Pu'er tea; S205, constructing a second-layer knowledge ontology classification model for the Pu'er tea field consisting of a first-layer knowledge ontology classification node, a relationship, and a second-layer knowledge ontology classification node triple model; The step S30 of establishing a Pu'er tea domain knowledge entity model associated with the first and second layer knowledge ontology classification models for constructing a Pu'er tea domain knowledge graph specifically includes the following steps: S301, extracting a Pu'er tea domain knowledge entity model, using a machine learning algorithm to extract a Pu'er tea domain knowledge entity model from existing semi-structured and unstructured text data. The Pu'er tea domain knowledge entity model consists of a triple structure of entity, attribute, and attribute value; S302, associating the Pu'er tea domain knowledge entity model with the ontology model, associating the Pu'er tea domain knowledge entity model extracted in step S301 with one or more nodes of the first-layer and second-layer knowledge ontology classification models to form a Pu'er tea domain knowledge graph; S303, storing: storing the Pu'er tea domain knowledge ontology classification model and the Pu'er tea domain knowledge entity model data into a graph database; S304, Operation and Maintenance, including: maintaining the first-level and second-level knowledge ontology classification models and feature lexicons in the Pu'er tea field, adjusting the relationship between the Pu'er tea field knowledge entity model and the Pu'er tea field knowledge ontology classification model, and reviewing and revising the content of the Pu'er tea field knowledge entity model; The Pu'er tea field knowledge graph construction device includes: An entity model extraction module is used to extract the Pu'er tea domain knowledge entity model from semi-structured and unstructured texts to form a triple data structure consisting of the Pu'er tea domain knowledge entity model name, attribute and attribute value; The first classification determination module is used to determine, through a feature word library, which classification the Pu'er tea domain knowledge entity model belongs to in the first-level knowledge ontology classification model of the Pu'er tea domain; The second classification determination module is used to judge and determine whether the Pu'er tea domain knowledge entity model belongs to a certain classification of the second-level knowledge ontology classification model in the Pu'er tea domain through the feature vocabulary; The model relationship construction module is used to establish the directed relationships between the first-level knowledge ontology classification model, the second-level knowledge ontology classification model, and the knowledge entity model in the Pu'er tea field; The knowledge operation and maintenance module is used to store, update, and maintain knowledge graph data in the field of Pu'er tea.

Images