An automated construction method and system for a knowledge graph in the field of geological disasters

Abstract

The present invention discloses a method and system for automatically constructing a knowledge graph in the field of geological disasters. The method includes the following steps: constructing a geological disaster chain ontology model by using a top-down method; describing the logical structure of the ontology model by using a five-tuple method; expressing the semantic relationship of the ontology model according to prior knowledge; using a bottom-up method, according to the established ontology model, extracting geological disaster entities, attributes and their relationships from existing geological disaster reports by using knowledge extraction technology; decomposing the extracted geological disaster entities, attributes and their relationships, establishing a mapping relationship between specific elements and the ontology model, and finally automatically generating a knowledge graph through alignment and fusion. The beneficial effect of the present invention is that it effectively solves the limitations such as the inability of single-disaster body modeling to accurately describe geological disasters.

Description

Technical Field

[0001] The present invention relates to the field of earthquake prediction, and in particular to a method and system for automatically constructing a knowledge graph for the field of geological disasters. Background Art

[0002] The concept of the knowledge graph originated in May 2012 and was initially proposed by Google to reveal the dynamic development law of the knowledge field. In the knowledge graph, entities, relationships, and attributes are the constituent elements. Usually, semantic symbols are represented by nodes, and the relationships between semantics are represented by edges. With the transformation of information into knowledge, the combination of the knowledge graph and geoscience knowledge integrates the characteristics of semantic relationships and spatial information, thus having the ability of large-scale association and causal analysis of multi-level knowledge of people, land, and space.

[0003] In the field of geoscience, Zhou Chenghu et al. proposed to extract geoscience knowledge from a large number of existing geoscience literatures to construct a geoscience knowledge graph, expand the unique spatio-temporal characteristics of geoscience knowledge, integrate multi-source geoscience elements, and establish a geoscience knowledge expression model. In the construction of the disaster knowledge graph, generally, ontology is used as the basic theory for disaster knowledge modeling to form a visual expression of disaster knowledge. In the construction of the disaster knowledge graph, many scholars have done relevant research. Generally, ontology is used as the basic theory for disaster knowledge modeling to form a visual expression of disaster knowledge. However, due to the multi-source heterogeneous nature of geological disaster data, from the perspective of natural language expression, constructing a geological disaster chain knowledge graph to associate the environment where disasters occur, disaster ontology, geographical objects, and emergency treatment to achieve the integration, fusion, and storage of data from different sources. Currently, most studies describe single elements of disaster events by constructing ontology knowledge from different sources, analyze the evolution process and associated relationships of disaster events, but cannot comprehensively analyze the process of secondary disasters caused by primary disasters, and it is difficult to describe the occurrence mechanism of geological disasters as a whole and accurately. Summary of the Invention

[0004] To solve the above problems, a method for automatically constructing a knowledge graph for the field of geological disasters provided by this application specifically includes the following steps:

[0005] S101: Construct a geological disaster chain ontology model by using a top-down method; describe the logical structure of the ontology model by using a five-tuple method; express the semantic relationship of the ontology model according to prior knowledge;

[0006] S102: Adopt a bottom-up method. According to the established ontology model, use knowledge extraction technology to extract geological disaster entities, attributes, and their relationships from existing geological disaster reports;

[0007] S103: Decompose the extracted geological disaster entities, attributes, and their relationships, establish the mapping relationship between specific elements and the ontology model, and finally automatically generate a knowledge graph through alignment and fusion methods.

[0008] Furthermore, the ontology model includes: geological environment ontology, geological disaster ontology, geographical object ontology, and emergency response ontology.

[0009] Furthermore, describe the logical structure of the ontology model, specifically:

[0010] Onto = (Con, Rel, Prop, Rule, Ins) (1)

[0011] Among them, Con refers to concepts, representing the collective term for a set of things with the same characteristics; Rel refers to relationships, representing the hierarchical relationships between concepts, between concepts and instances, as well as the spatio-temporal relationships and semantic relationships between instances; Prop refers to attributes, representing the relevance between instance objects and the relevance between instances and values; Rule refers to rules, representing the constraint expressions for the value ranges, types, and combination methods of domain concepts and instances.

[0012] Furthermore, express the semantic relationships of the ontology model according to prior knowledge, specifically:

[0013] The geological environment ontology is the disaster-bearing environment of the geological disaster ontology;

[0014] The emergency response ontology is the response and countermeasure for the geological disaster ontology;

[0015] The geographical object ontology is the disaster-affected body of the geological disaster ontology;

[0016] The geographical object ontology is also the processing object of the emergency response ontology.

[0017] A knowledge graph automatic construction system for the geological disaster field, the system includes:

[0018] Geological disaster chain ontology model construction module: Build a geological disaster chain ontology model using a top-down method; Describe the logical structure of the ontology model using a five-tuple method; Express the semantic relationships of the ontology model according to prior knowledge;

[0019] Knowledge extraction module: Use a bottom-up method, according to the established ontology model, use knowledge extraction technology to extract geological disaster entities, attributes, and their relationships from existing geological disaster reports;

[0020] Knowledge graph generation module: Decompose the extracted geological disaster entities, attributes, and their relationships, establish the mapping relationship between specific elements and the ontology model, and finally automatically generate a knowledge graph through alignment and fusion methods.

[0021] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for automatically constructing a knowledge graph in the field of geological disasters are implemented.

[0022] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the steps of the method for automatically constructing a knowledge graph in the field of geological disasters are implemented.

[0023] Compared with the prior art, the beneficial effects of the present invention include:

[0024] 1. Guided by the theory and method of knowledge graph, the present invention constructs a unified logical expression framework for geological disaster chains, defines and constructs four core elements in the field of geological disasters, namely geological disaster events, geological environment ontology, geographical object ontology, and emergency response ontology, and classifies and defines the attributes and semantic relationships between concepts and instances at the ontology level.

[0025] 2. The present invention effectively solves the problem that the modeling of a single disaster body cannot accurately describe the limitations of geological disasters.

[0026] 3. It can clearly and accurately express the rich semantic relationships between geological disaster entities and between entities and attributes, effectively verifying the feasibility and effectiveness of the method for constructing a geological disaster knowledge graph proposed by the present invention, and providing an idea for the construction and research of a geoscience knowledge graph. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flowchart of the method of the present invention;

[0028] Figure 2 is a schematic diagram of the construction of a geological disaster knowledge graph;

[0029] Figure 3 is a schematic diagram (partial) of the construction of geological disaster chain ontology instances;

[0030] Figure 4 is a schematic diagram (partial) of a geological disaster knowledge graph based on spatio-temporal evolution;

[0031] Figure 5 is a schematic diagram (partial) of a process-oriented geological disaster chain knowledge graph. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] To make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0033] The present invention provides a method for automatically constructing a knowledge graph in the field of geological disasters. Please refer to Figure 1 , Figure 1 which is a flowchart of the method of the present invention; the method of the present invention includes the following steps:

[0034] S101: Construct a geological disaster chain ontology model by using a top-down method; describe the logical structure of the ontology model by using a five-tuple method; express the semantic relationship of the ontology model according to prior knowledge;

[0035] It should be noted that the modeling of the geological disaster chain ontology takes geological disaster events as the core, and realizes the formalization and logical expression of geological disaster concepts and instances through the ontology, so as to support geoscience knowledge discovery and knowledge reasoning.

[0036] The basic elements of geological disaster events include time, space, phenomena, etc., which have typical spatio-temporal characteristics, and the extraction and knowledge expression of their information need to consider spatio-temporal problems.

[0037] In this application, the geological disaster chain ontology model includes four parts: geological environment ontology, geological disaster ontology, geographical object ontology and emergency response ontology. Please refer to Figure 2 shown in Figure 2 which is a specific visualization process schematic diagram of knowledge graph construction.

[0038] In order to summarize and sort out geological domain knowledge under a unified semantic expression framework, so as to construct the logical association between knowledge systems and ultimately serve information extraction and knowledge reasoning, the present invention conducts a logical structure expression on the geological disaster chain ontology;

[0039] Specifically, since the five-tuple representation method can satisfy the holistic description of geological disasters and the geological environment and geological objects under their action, the present invention selects the five-tuple as the description framework of the ontology, and finally forms a unified expression of knowledge, which is expressed as:

[0040] Onto=(Con, Rel, Prop, Rule, Ins) (1)

[0041] Among them, Con refers to concept, representing the collective term for a set of things with the same characteristics; Rel refers to relationship, representing the hierarchical relationships between concepts, between concepts and instances, as well as the spatio-temporal relationships and semantic relationships between instances; Prop refers to property, representing the relevance between instance objects and the relevance between instances and values; Rule refers to rule, representing the constraint expression on the value range, type and combination method of domain concepts and instances, so as to support semantic reasoning;

[0042] It should be noted that since the geological disaster knowledge graph schema level includes a set of concept nodes and a combination of concept edge relationships, which represents the representation of concept nodes and the relationships between concepts in the field of geological disasters. Therefore, based on the existing prior knowledge, this invention divides the four types of elements in the field of geological disasters, namely geological disaster events, geological disaster environments, geographical objects, and emergency responses, into concept levels, and defines the attribute relationships and semantic relationships between concepts.

[0043] The geological environment ontology is the disaster-bearing environment of the geological disaster ontology;

[0044] The emergency response ontology is the response and countermeasures of the geological disaster ontology;

[0045] The geographical object ontology is the disaster-affected body of the geological disaster ontology;

[0046] The geographical object ontology is also the processing object of the emergency response ontology. Specifically:

[0047] (1) For the geological disaster ontology, the focus of its description is on the classification of geological disasters and the expression of the relationships between disasters. According to the specific concepts and hierarchical relationships in the "Classification and Grading of Geological Disasters in the National Land Resources Industry Standard of the People's Republic of China (DZ - 2000)" and the "Classification and Grading of Geological Disasters in the Geological and Mineral Industry Standard of the People's Republic of China (DZ0238 - 2004)", geological disasters can be classified into 13 types such as landslides, collapses, debris flows, ground fissures, land subsidence, and ground collapses according to categories, and each category can be further divided into sub-categories. For example, collapses can be divided into giant collapses, large collapses, medium collapses, and small collapses, etc. In addition, there are often induced relationships between different types of geological disasters, which ultimately lead to the generation of disaster chains. At the same time, the generation mechanisms and affected objects of different types of geological disasters are different, and the generated disaster chains are also different.

[0048] The basic framework in the geological disaster ontology includes concepts and related relationships, while instances, attributes, and constraints further enrich and improve the logical structure of the ontology framework. In terms of the attributes of geological disasters, spatio-temporal attributes are a typical characteristic, and non-spatio-temporal attributes are also included. At the same time, the general attributes and specific attributes between different geological disasters need to be considered. Some general attributes and specific attributes of geological disasters are shown in Table 1.

[0049] Table 1 Description of Geological Disaster Attributes (Partial)

[0050]

[0051] (2) Geographic object ontology. Its description is mainly based on the classification hierarchy of geographic information elements and related concepts in "Classification and Codes for Fundamental Geographic Information Elements" (GB13923 - 2006). The attributes of concepts in the geographic object ontology include both geometric measurement attributes such as area, distance, length, quantity, etc., and semantic descriptions such as name. Geographic object relationships include two major categories: spatial relationships and non - spatial relationships. The former includes azimuth relationships, topological relationships, etc., and the latter includes equivalence relationships, subordination relationships, part / whole relationships, etc., which describe the semantic relationships between concepts and concepts, concepts and instances, and instances and instances.

[0052] (3) Geological environment ontology. Since the modeling of the geological environment ontology under the action of geological disasters involves many geological environment objects caused by natural and human factors, the formulation of its basic concepts and relationships is based on the concepts and specific hierarchical relationships in the "Geological Disaster Investigation Specification".

[0053] The description of geological environment relationships includes two types: typical spatial relationship descriptions and semantic relationship descriptions. The spatial relationship descriptions in the geological environment mainly adopt three types of common topological relationships, azimuth relationships, and metric relationships in geography. The topological relationship designs six common basic relationships (such as containment, being contained, equality, etc.), the azimuth relationship represents eight common basic relationships (such as east, south, west, north, etc.), and the metric relationship represents the measurement of the relative distance between two spatial positions; the description of semantic relationships represents the relationships between concepts, between concepts and entities, and between instances and instances. Common semantic relationships include parent - child relationships, whole - part relationships, mutually exclusive relationships, and equivalence relationships. The semantic relationships between instances include functional relationships, attribute relationships, and structural relationships, etc.

[0054] (4) Emergency response ontology. Its description is based on relevant documents such as the "Regulations on the Prevention and Control of Geological Disasters" and the "Overall Emergency Plan for National Sudden Public Events". The whole process of geological disasters is divided into three different stages: pre - disaster, in - disaster, and post - disaster, and the goals and tasks of each stage are different.

[0055] S102: Adopt a bottom - up method. According to the established ontology model, use knowledge extraction technology to extract geological disaster entities, attributes, and their relationships from existing geological disaster reports;

[0056] S103: Decompose the extracted geological disaster entities, attributes, and their relationships, establish the mapping relationship between specific elements and the ontology model, and finally automatically generate a knowledge graph through alignment and fusion methods.

[0057] An automated knowledge graph construction system for the field of geological disasters, the system includes:

[0058] Geological disaster chain ontology model construction module: Construct a geological disaster chain ontology model using a top-down method; Describe the logical structure of the ontology model using a five-tuple method; Express semantic relationships for the ontology model based on prior knowledge;

[0059] Knowledge extraction module: Adopt a bottom-up method, and according to the established ontology model, use knowledge extraction technology to extract geological disaster entities, attributes and their relationships from existing geological disaster reports;

[0060] Knowledge graph generation module: Decompose the extracted geological disaster entities, attributes and their relationships, establish the mapping relationship between specific elements and the ontology model, and finally automatically generate a knowledge graph through alignment and fusion methods.

[0061] A computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the automated knowledge graph construction method for the field of geological disasters are implemented.

[0062] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the automated knowledge graph construction method for the field of geological disasters are implemented.

[0063] Example 1

[0064] Based on the conceptual semantic framework of the ontology library in the field of geological disaster chains, the present invention extracts entities and relationships based on rules and machine learning algorithms for the knowledge source of regional geological disaster reports, and finally stores the extracted entity relationships as triples in the graph database Neo4j database. The specific steps are as follows:

[0065] (1) Data source

[0066] The original data selected in the present invention is geological disaster reports from the National Geological Archives of China (http: / / www.ngac / cn / ), with a total of 350 pages and approximately 1,100,000 words. In the present invention, we convert it into a word format as the format of the input data.

[0067] (2) Entity and relationship extraction

[0068] When performing entity and relationship extraction, the annotation system adopted by the present invention is BIO, where B represents the starting position of the entity, I represents other positions in the entity except the starting position, and O represents non-entity; the evaluation indicators mainly adopt accuracy rate, recall rate and comprehensive metric F value. Finally, the present invention adopts the BERT-BiLSTM-Attention-CRF model. In addition, in order to verify the effect of this model, four experiments of rule matching, BiLSTM, BiLSTM-CRF, and BiLSTM-Attention-CRF are carried out at the same time. The experimental results are compared as shown in Tables 2 and 3. Compared with the other four models, the BERT-BiLSTM-Attention-CRF model has significantly better effects.

[0069] Table 2 Experimental Results of Entity Recognition by Model Method

[0070] Model Accuracy / % Recall / % F - measure Rule matching 88.10 89.20 88.65 BiLSTM 93.30 93.80 93.55 BiLSTM - CRF 94.50 94.60 94.55 BiLSTM - Attention - CRF 95.20 95.60 95.40 BERT - BiLSTM - Attention - CRF 96.10 96.50 96.30

[0071] Table 3 Experimental Results of Entity Relationship Recognition by Model Method

[0072] Model Accuracy / % Recall / % F - measure BiLSTM 68.10 65.10 66.57 BiLSTM - CRF 69.23 68.45 68.84 BiLSTM - Attention - CRF 71.56 70.23 70.89 BERT - BiLSTM - Attention - CRF 75.19 74.23 74.71

[0073] (3) Knowledge Storage

[0074] Through the above-described extraction and processing processes, the information data in the geological report is converted into structured knowledge through algorithms. Since the graph database has obvious storage advantages for data with clear structural levels, entity relationships, and entity attribute classes, and can realize the visualization display of the geological disaster chain knowledge graph from multiple dimensions such as the concept level, entity level, and attribute level, for entity relationship and entity attribute information, when storing in the graph database, multiple pieces of knowledge are formed in the form of (entity, relationship, entity) and (entity, attribute, attribute value) triples. The head and tail parts are stored as nodes in the graph, and the attribute information and relationship information are stored as edges, in order to realize the mapping between structured knowledge and triple knowledge in the graph database. Based on the constructed geological disaster chain knowledge graph, applications such as knowledge graph completion and knowledge reasoning can be realized based on graph query languages and mining algorithms.

[0075] Example Two

[0076] The present invention uses the Protégé ontology modeling tool combined with the OWL DL language to model the geological disaster chain ontology. In the present invention, the structured and standardized expression of geological disaster knowledge in the regional geological disaster report in combination with debris flow is elaborated, and a geological disaster chain formed by a series of secondary disasters caused by different disasters is constructed, such as Figure 3As shown in the figure. Earthquakes and rainfall can trigger disasters such as landslides, collapses, and debris flows, which will have a greater impact on geographical objects in the surrounding areas. For example, landslides can cause crop failures, traffic disruptions, and villages and towns may be buried, resulting in a large amount of economic losses. At the same time, rescue agencies will take corresponding treatment and disposal measures, including relocation, professional monitoring, engineering treatment, mass monitoring and prevention, etc. These form the basis for the extraction of geological disaster information.

[0077] Using the deep learning method proposed in the present invention, entities and relationships are extracted from the processed geological disaster reports. Finally, the graph database Neo4j is used to store the extracted entities, relationships, and attribute values. Part of the geological disaster knowledge graph is as Figure 4 shown.

[0078] Example 3

[0079] The occurrence of geological disasters not only includes single-type disasters, but often multiple types of secondary or derivative disasters occur. Disasters of the same type also have different states, and their corresponding attributes and impacts are also different. Therefore, taking the Jiuzhaigou earthquake as an example, the present invention constructs a process-oriented geological disaster chain knowledge graph, as Figure 5 shown.

[0080] Generally speaking, the present invention is guided by the knowledge graph theory and method, constructs a unified logical expression framework for geological disaster chains, defines and constructs four core elements in the field of geological disasters, namely geological disaster events, geological environment ontology, geographical object ontology, and emergency disposal ontology, and classifies and defines the attributes and semantic relationships between concepts and instances at the ontology level.

[0081] Taking the regional geological report as an example, an experimental case analysis is carried out, and a geological disaster knowledge graph is constructed through data preprocessing and the extraction of geological disaster entities and relationships based on deep learning.

[0082] The beneficial effects of the present invention are:

[0083] 1. Guided by the knowledge graph theory and method, the present invention constructs a unified logical expression framework for geological disaster chains, defines and constructs four core elements in the field of geological disasters, namely geological disaster events, geological environment ontology, geographical object ontology, and emergency disposal ontology, and classifies and defines the attributes and semantic relationships between concepts and instances at the ontology level.

[0084] 2. The present invention effectively solves the problems such as the limitations that the modeling of a single disaster body cannot accurately describe geological disasters.

[0085] 3. It can clearly and accurately express the rich semantic relationships between geological hazard entities and between entities and attributes, effectively verifying the feasibility and effectiveness of the geological hazard knowledge graph construction method proposed by the present invention, and providing an idea for the construction and research of geoscience knowledge graphs.

[0086] The specific implementation manners of the present invention described above do not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and deformations made according to the technical concept of the present invention shall be included in the protection scope of the claims of the present invention.

Claims

1. A method for automatically constructing a knowledge graph in the field of geological disasters, characterized in that: It includes the following steps: S101: Construct a geological disaster chain ontology model by using a top-down method; describe the logical structure of the ontology model by using a five-tuple method; express the semantic relationship of the ontology model according to prior knowledge; S102: Adopt a bottom-up method. According to the established ontology model, use knowledge extraction technology to extract geological disaster entities, attributes and their relationships from existing geological disaster reports; S103: Decompose the extracted geological disaster entities, attributes and their relationships, establish the mapping relationship between specific elements and the ontology model, and finally automatically generate a knowledge graph through alignment and fusion; The ontology model includes: geological environment ontology, geological disaster ontology, geographical object ontology and emergency response ontology; The geological disaster ontology is classified according to the specific concepts and hierarchical relationships in the "Geological Disaster Classification and Grading" of the Land and Resources Industry Standard of the People's Republic of China (DZ-2000) and the "Geological Disaster Classification and Grading" of the Geological and Mineral Industry Standard of the People's Republic of China (DZ0238-2004), including: landslide, collapse, debris flow, ground fissure, land subsidence and ground collapse; The geographical object ontology is described according to the hierarchical structure of geographical information elements classification and its related concepts in the "Classification and Codes of Fundamental Geographical Information Elements" (GB 13923-2006). The attributes of geographical objects include: area, distance, length, quantity and name; The geological environment ontology is described according to the concepts and specific hierarchical relationships in the "Geological Disaster Investigation Specification", including: two types of typical spatial relationships and semantic relationships; The emergency response ontology divides the whole process of geological disasters into three different stages: pre-disaster, in-disaster and post-disaster according to the "Regulations on the Prevention and Control of Geological Disasters" and the "Overall Emergency Plan for National Sudden Public Events". The goals and tasks of each stage are different; Describing the logical structure of the ontology model specifically as: Onto=(Con,Rel,Prop,Rule,Ins) (1) Among them, Con refers to concepts, representing the general term of a set of things with the same characteristics; Rel refers to relationships, representing the hierarchical relationships between concepts, between concepts and instances, as well as the spatio-temporal relationships and semantic relationships between instances; Prop refers to attributes, representing the relevance between instance objects and the relevance between instances and values; Rule refers to rules, representing the constraint expressions on the value range, type and combination method of domain concepts and instances; Expressing the semantic relationship of the ontology model according to prior knowledge, specifically as: The geological environment ontology is the disaster-bearing environment of the geological disaster ontology; The emergency response ontology is the response and countermeasure of the geological disaster ontology; The geographical object ontology is the disaster-bearing body of the geological disaster ontology; The geographical object ontology is also the processing object of the emergency response ontology.

2. A system for automatically constructing a knowledge graph in the field of geological disasters, characterized in that: The system includes: Geological disaster chain ontology model construction module: Construct a geological disaster chain ontology model using a top-down approach; describe the logical structure of the ontology model using a five-tuple method; express the semantic relationships of the ontology model based on prior knowledge; Knowledge extraction module: Adopt a bottom-up approach. According to the established ontology model, use knowledge extraction technology to extract geological disaster entities, attributes, and their relationships from existing geological disaster reports; Knowledge graph generation module: Decompose the extracted geological disaster entities, attributes, and their relationships, establish the mapping relationship between specific elements and the ontology model, and finally automatically generate a knowledge graph through alignment and fusion methods; The ontology model includes: geological environment ontology, geological disaster ontology, geographical object ontology, and emergency response ontology; The geological disaster ontology is classified according to the specific concepts and hierarchical relationships in the "Classification and Grading of Geological Disasters in the Industry Standard of the Ministry of Land and Resources of the People's Republic of China (DZ - 2000)" and the "Classification and Grading of Geological Disasters in the Industry Standard of the Ministry of Geology and Mineral Resources of the People's Republic of China (DZ0238 - 2004)", including: landslides, collapses, debris flows, ground fissures, land subsidence, and ground collapses; The geographical object ontology is described based on the classification hierarchical structure of geographical information elements and their related concepts in the "Classification and Coding of Fundamental Geographical Information Elements" (GB 13923 - 2006). The attributes of geographical objects include: area, distance, length, quantity, and name; The geological environment ontology is described based on the concepts and specific hierarchical relationships in the "Geological Disaster Investigation Specification", including: two categories of typical spatial relationships and semantic relationships; The emergency response ontology divides the whole process of geological disasters into three different stages: pre-disaster, in-disaster, and post-disaster according to the "Regulations on the Prevention and Control of Geological Disasters" and the "Overall Emergency Plan for National Sudden Public Events", and the goals and tasks of each stage are different; Describe the logical structure of the ontology model, specifically: Onto=(Con,Rel,Prop,Rule,Ins) (1) Among them, Con refers to concepts, representing the general term of a set of things with the same characteristics; Rel refers to relationships, representing the hierarchical relationships between concepts, between concepts and instances, as well as the spatio-temporal relationships and semantic relationships between instances; Prop refers to attributes, representing the relevance between instance objects and the relevance between instances and values; Rule refers to rules, representing the constraint expressions on the value range, type, and combination method of domain concepts and instances; Express the semantic relationships of the ontology model based on prior knowledge, specifically: The geological environment ontology is the disaster-bearing environment of the geological disaster ontology; The emergency response ontology is the response and countermeasure for the geological disaster ontology; The geographical object ontology is the disaster-affected body of the geological disaster ontology; The geographical object ontology is also the processing object of the emergency response ontology.

3. A computer device, Characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method for automatically constructing a knowledge graph for the geological disaster field as described in claim 1.

4. A computer-readable storage medium, characterized in that, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for automatically constructing a knowledge graph for the field of geological disasters as described in claim 1 are implemented.

Images