Construction, retrieval and report generation method and system of safety monitoring knowledge graph

Abstract

The invention provides a construction, retrieval and report generation method and system for a safety monitoring knowledge graph, and relates to the field of safety monitoring. The construction method comprises the following steps: extracting entity information from safety monitoring data of a target project and determining entity nodes; determining a subordinate relationship among the entity nodes, and connecting the entity nodes through edges; generating a unique index identifier for the monitoring result node as an attribute value, taking the index identifier as an index or a key, and setting the monitoring result content as an attribute node; determining an attribute relationship between the monitoring result node and the attribute node, and performing connection through edges to form a security monitoring knowledge graph; and the user is helped to better organize, understand and utilize knowledge. According to the retrieval method, the information retrieval efficiency is improved, and the understanding and reasoning capabilities are enhanced. According to the generation method, dynamic automatic generation of the safety monitoring data analysis report can be realized, and the working efficiency is improved. The building system is used for providing an operating system for a user.

Description

Technical Field

[0001] The present application relates to the field of security monitoring, and specifically, to a method and system for constructing, retrieving and generating reports of a security monitoring knowledge graph. Background Art

[0002] The safety of water conservancy and hydropower projects is receiving more and more attention, which requires the project construction and operation management units to carry out safety monitoring and data compilation and analysis in strict accordance with the relevant requirements of the specifications, and submit safety monitoring data analysis reports on a regular basis; however, since such reports cover a wide range of areas and involve many results, the documents are often hundreds of pages long, or even thousands of pages long. The traditional method of manually compiling reports is inefficient, time-consuming, and prone to low-level errors, making it difficult to provide timely technical support for the safe operation of the project. Summary of the invention

[0003] The purpose of this application is to provide a method and system for constructing, retrieving and generating reports for a safety monitoring knowledge graph, and to establish an engineering safety monitoring knowledge graph through knowledge extraction, association, fusion and other methods.

[0004] This application is implemented as follows:

[0005] In a first aspect, the present application provides a method for constructing a security monitoring knowledge graph, comprising:

[0006] Extract entity information from the safety monitoring data of the target project and determine entity nodes, wherein the entity nodes include monitoring object nodes, monitoring part nodes, monitoring project nodes and monitoring result nodes;

[0007] Determine the subordinate relationship between each entity node and connect them through edges;

[0008] Generate a unique index identifier as an attribute value for the monitoring achievement node, the attribute value and the attribute relationship of the monitoring achievement node are a one-to-one mapping relationship, use the index identifier as an index or key, present the monitoring achievement content associated with the index identifier, and set the monitoring achievement content as an attribute node, the monitoring achievement content includes characteristic text, graphics, and statistical reports;

[0009] Determine the attribute relationship between monitoring result nodes and attribute nodes, connect them through edges, and form a safety monitoring knowledge graph;

[0010] The knowledge graph model is used to describe the relationship between nodes to form a triple data structure, and the security monitoring knowledge graph is saved in the graph database in the form of a triple data structure.

[0011] Based on the first aspect, the specific steps of saving the safety monitoring knowledge graph into the graph database in the form of a triple data structure include:

[0012] The triple data is disassembled to obtain the information of entities, relationships and attributes. The entities, relationships and attributes are input into the Neo4j graph database using the graph query language, and an index is created in Neo4j to form a complete tree-structured security monitoring knowledge graph.

[0013] Based on the first aspect, safety monitoring data includes design drawings and observation data.

[0014] Based on the first aspect, the specific steps of determining the subordinate relationship between the entity nodes include:

[0015] The monitoring object-monitoring part, monitoring part-monitoring project, and monitoring project-monitoring results are all one-to-many hierarchical subordinate relationships.

[0016] In the second aspect, the present application provides a retrieval method of a safety monitoring knowledge graph, which is applied to the above-mentioned construction method, including:

[0017] Step 1: Visit the initial vertex node A and mark node A as visited;

[0018] Step 2: If node A is not a monitoring project node, execute step 3 to perform a depth-first traversal recursively on node A to find the subordinate node B of node A; if node A is a monitoring project node, execute step 5;

[0019] Step 3: If node B exists, continue to step 4; if node B does not exist, return to step 1 and continue with the next subordinate node of node A;

[0020] Step 4: If node B has not been visited and is not a monitoring project node, perform a depth-first traversal recursively on node B, treat node B as a new vertex node A, and execute steps 1 to 3; if node B is a monitoring project node, execute step 5;

[0021] Step 5: Create a monitoring results list, use a breadth-first search algorithm to traverse all subordinate nodes of the monitoring project node to monitor the results nodes, and add them to the list;

[0022] Step 6: After the query is completed, the graph directory structure with vertex node A as the root node and all monitoring result node queues under it are obtained.

[0023] In a third aspect, the present application provides a method for generating a safety monitoring data analysis report based on a knowledge graph, applying the above-mentioned construction method, the generation method includes:

[0024] Determine the report content based on the data compilation and analysis requirements of the target project;

[0025] Select one or more entity nodes in the graph database as analysis objects based on the report content, and set the statistical time range of the report;

[0026] According to the selected analysis object, the graph query language is used to search in the graph database to obtain the analysis scope of the report;

[0027] Based on the analysis scope and the set statistical time range, the monitoring results content in the report is generated, and the monitoring results content includes characteristic text, display graphics and statistical reports;

[0028] Based on the safety monitoring knowledge graph structure and the selected analysis scope, an outline of the safety monitoring data analysis report is generated. The outline directory is combined according to the association relationship in the safety monitoring data analysis report graph, and the monitoring results are placed in the analysis report according to the outline;

[0029] Write the generated report outline into the document file in the form of a document title directory;

[0030] Add the relevant monitoring results to the corresponding directory of the document;

[0031] Edit the document style, generate and save a standardized safety monitoring data analysis report document;

[0032] The generated safety monitoring data analysis report document is exported to the location specified by the user, and the integrity of the exported file is verified after the export is completed.

[0033] In a fourth aspect, the present application provides a system for constructing a security monitoring knowledge graph, including:

[0034] An entity node determination module is used to extract entity information from the safety monitoring data of the target project and determine entity nodes, wherein the entity nodes include monitoring object nodes, monitoring part nodes, monitoring project nodes and monitoring result nodes;

[0035] A subordinate edge establishment module is used to determine the subordinate relationship between each entity node and connect them through edges;

[0036] The attribute node generation module is used to generate a unique index identifier as an attribute value for the monitoring achievement node. The attribute value and the attribute relationship of the monitoring achievement node are a one-to-one mapping relationship. The index identifier is used as an index or key to present the monitoring achievement content associated with the index identifier, and the monitoring achievement content is set as an attribute node. The monitoring achievement content includes characteristic text, graphics, and statistical reports;

[0037] The knowledge graph generation module is used to determine the attribute relationship between the monitoring result nodes and the attribute nodes, connect them through edges, and form a safety monitoring knowledge graph;

[0038] The storage module is used to use the knowledge graph model to describe the association relationship between nodes to form a triple data structure, and save the security monitoring knowledge graph in the form of a triple data structure into the graph database.

[0039] Compared with the prior art, the present invention has at least the following advantages or beneficial effects:

[0040] On the first aspect, this application proposes a method for constructing a safety monitoring knowledge graph, and establishes a safety monitoring knowledge graph for the target project through knowledge extraction, association combination, etc. Help users better organize, understand and use knowledge. On the second aspect, this application provides a retrieval method for a safety monitoring knowledge graph, which improves information retrieval efficiency, enhances understanding and reasoning capabilities, and helps break down data barriers and achieve knowledge fusion. On the third aspect, this application provides a method for generating a safety monitoring data analysis report based on a knowledge graph, which can realize the dynamic and automatic generation of safety monitoring data analysis reports, facilitate safety monitoring compilation and analysis, increase work efficiency, and provide decision makers with strong data support. On the fourth aspect, this application provides a system for constructing a safety monitoring knowledge graph, and provides users with an operating system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 This is a flowchart of an embodiment of a method for constructing a security monitoring knowledge graph in this application;

[0043] Figure 2 A schematic diagram of a knowledge graph in an embodiment of a method for constructing a safety monitoring knowledge graph in this application;

[0044] Figure 3 A schematic diagram of monitoring results in an embodiment of a method for constructing a security monitoring knowledge graph in the present application.

[0045] Figure 4 This is a flowchart of an embodiment of a method for retrieving a security monitoring knowledge graph in the present application;

[0046] Figure 5 This is a flowchart of an embodiment of a method for generating a safety monitoring data analysis report based on a knowledge graph in this application;

[0047] Figure 6This is a schematic diagram of the monitoring results content of an embodiment of a method for generating a safety monitoring data analysis report based on a knowledge graph in the present application. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0049] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0050] Example

[0051] The present application embodiment provides a method for constructing a safety monitoring knowledge graph, which can establish a safety monitoring knowledge graph for a target project through knowledge extraction, association combination, etc., to help users better organize, understand and use knowledge. Figure 1 , the method comprises the following steps:

[0052] In a first aspect, the present application provides a method for constructing a security monitoring knowledge graph, comprising:

[0053] S101: extracting entity information from the safety monitoring data of the target project and determining entity nodes, wherein the entity nodes include monitoring object nodes, monitoring part nodes, monitoring project nodes and monitoring result nodes;

[0054] In this step, the safety monitoring data includes design drawings, observation data, etc., and the content directly related to safety monitoring is obtained from these data. Based on these contents, entity information is extracted and entity nodes are determined. The entity information includes monitoring objects, monitoring parts, monitoring projects and monitoring results. The entity nodes include monitoring object nodes, monitoring part nodes, monitoring project nodes and monitoring results nodes; please refer to Figure 2The monitoring object can be a water conservancy project hub (marked 1 in the figure), and the monitoring part can be a hydraulic structure engineering entity. The monitoring parts are divided according to the design drawings and the actual project, such as the main dam, AA section, BB section and CC section (marked 2 in the figure); by deconstructing the monitoring object, one or more monitoring items are obtained, such as seepage pressure monitoring, crack monitoring, vertical displacement monitoring and horizontal displacement monitoring (marked 3 in the figure), which are used to describe the safety status of the monitoring object; the monitoring results belong to the monitoring project, and the results information related to the monitoring project is displayed in the form of text description, results graphics, compiled reports, etc. After integrating all the monitoring results information in a monitoring project, the safety status of the monitoring project is obtained.

[0055] S102: Determine the subordinate relationship between each entity node and connect them through edges;

[0056] In this step, please refer to Figure 2 In this embodiment, there are four entities: monitoring object, monitoring part, monitoring project and monitoring result. Among them, monitoring object-monitoring part, monitoring part-monitoring project, and monitoring project-monitoring result are all one-to-many hierarchical subordinate relationships, and they are connected by edges. Edges are used to describe the subordinate relationship of entity nodes. In the knowledge graph, edges are used to connect different nodes and represent the relationship between them. Once the subordinate relationship between entity nodes is determined, these nodes can be connected through edges.

[0057] S103: Generate a unique index identifier as an attribute value for the monitoring achievement node, the attribute value and the attribute relationship of the monitoring achievement node are a one-to-one mapping relationship, use the index identifier as an index or key, present the monitoring achievement content associated with the index identifier, and set the monitoring achievement content as an attribute node, the monitoring achievement content includes feature text, graphics, and statistical reports;

[0058] In this step, please refer to Figure 2, the monitoring result node is the outermost circle, and each monitoring result node (representing a specific monitoring result) is assigned a unique index identifier (UID). This index identifier is stored as an attribute value of the node and is used to uniquely identify the monitoring result. Each monitoring result node corresponds to only one unique index identifier (UID), and each index identifier (UID) corresponds to only one monitoring result node. This one-to-one mapping relationship ensures the accuracy and traceability of the data. The index identifier (UID) is used as an index or key to quickly retrieve and locate information associated with a specific monitoring result. This indexing mechanism improves the efficiency of data access. Through the UID index or key, the system can present detailed content associated with a specific monitoring result. These contents may include monitoring results, analysis reports, charts and other forms, which together constitute a comprehensive display of the monitoring results. Please refer to Figure 3 , Figure 3 This is a schematic diagram of the information associated with the monitoring results of this embodiment. In some cases, the content of the monitoring results may be further organized into attribute nodes. These attribute nodes may contain more detailed monitoring data, analysis results or other related information, which are connected to the monitoring results nodes through attribute relationships, forming a more complex and rich data structure.

[0059] S104: Determine the attribute relationship between the monitoring result node and the attribute node, connect them through edges, and form a safety monitoring knowledge graph;

[0060] In this step, the process of determining the attribute relationship is to clarify how the monitoring result nodes and these attribute nodes are related to each other. For example, a monitoring result may have a "monitoring time" attribute, which is associated with an attribute node representing a specific time. Connecting multiple monitoring result nodes, attribute nodes, and the edges between them forms a structured knowledge representation system, namely the safety monitoring knowledge graph. This knowledge graph not only contains rich monitoring data and information, but also displays the associations and hierarchical structures between these data and information in an intuitive and easy-to-understand way.

[0061] S105: Use the knowledge graph model to describe the association relationship between nodes to form a triple data structure, and save the security monitoring knowledge graph in the form of a triple data structure into the graph database.

[0062] In this step, the knowledge graph model is a data model used to represent the relationship between entities (nodes). By using the knowledge graph model, we can clearly describe the association relationship between each node (such as monitoring results node, attribute node, etc.) in the safety monitoring data analysis report. These association relationships reflect the intrinsic connection between monitoring data and are the basis for building a knowledge graph. A triple consists of three parts: head entity, relationship, and tail entity. In the knowledge graph, each relationship can be represented as a triple, in which the head entity and the tail entity are connected by a relationship. For example, in the safety monitoring knowledge graph, a triple may be represented as a monitoring result node (head entity) connected to an attribute node (tail entity) representing the monitoring time through a "has" relationship. This triple data structure is concise and easy to store and query. When saving the safety monitoring knowledge graph to the graph database, first convert each node and relationship in the knowledge graph into a triple data structure. Then use the interface or tool provided by the graph database to batch import these triple data into the graph database. Through such settings, we can easily query, analyze and utilize these knowledge graph data in the graph database.

[0063] Based on the first aspect, the specific steps of saving the safety monitoring knowledge graph into the graph database in the form of a triple data structure include:

[0064] The triple data is disassembled to obtain the information of entities, relationships and attributes. The entities, relationships and attributes are input into the Neo4j graph database using the graph query language, and an index is created in Neo4j to form a complete tree-structured security monitoring knowledge graph.

[0065] By decomposing the triple data, we get the basic elements that make up the knowledge graph: entities, relationships, and attributes. Before inputting the entity, relationship, and attribute information into the Neo4j graph database, you need to use a graph query language supported by Neo4j (such as Cypher) to write the corresponding data insertion statements. Cypher is a declarative graph query language that allows users to express complex graph queries and update operations in a way that is close to natural language. Using the data insertion statements written in the Cypher language, the decomposed entity, relationship, and attribute information is input into the Neo4j graph database. Create indexes for important entities or relationships in the Neo4j graph database. Indexes can speed up the database's search for specific nodes or relationships, thereby optimizing query performance.

[0066] Based on the first aspect, safety monitoring data includes design drawings and observation data.

[0067] Based on the first aspect, the specific steps of determining the subordinate relationship between the entity nodes include:

[0068] The monitoring object-monitoring part, monitoring part-monitoring project, and monitoring project-monitoring results are all one-to-many hierarchical subordinate relationships.

[0069] In the second aspect, this application provides a retrieval method for a security monitoring knowledge graph, which is applied to the above-mentioned construction method. Please refer to Figure 4 , S1 is step one, S2 is step two, S3 is step three, S4 is step four, S5 is step five, S6 is step six, the method comprises:

[0070] Step 1: Visit the initial vertex node A and mark node A as visited;

[0071] In this step, starting from the specified starting node A, A is marked as visited to ensure that the node will not be visited repeatedly during subsequent traversal processes.

[0072] Step 2: If node A is not a monitoring project node, execute step 3 to perform a depth-first traversal recursively on node A to find the subordinate node B of node A; if node A is a monitoring project node, execute step 5;

[0073] In this step, if node A is a monitoring project node, directly execute step 5, because at this time a starting point of a monitoring project has been found, and then all monitoring results under the monitoring project need to be collected.

[0074] Step 3: If node B exists, continue to step 4; if node B does not exist, return to step 1 and continue with the next subordinate node of node A;

[0075] In this step, if there is no subordinate node B, return to step 1 and try to continue searching from the next sibling node of node A (that is, other nodes that have the same parent node as A).

[0076] Step 4: If node B has not been visited and is not a monitoring project node, perform a depth-first traversal recursively on node B, treat node B as a new vertex node A, and execute steps 1 to 3; if node B is a monitoring project node, execute step 5;

[0077] In this step, if node B has not been visited and is not a monitoring item node, a depth-first traversal recursively is performed on node B, B is regarded as a new starting point (ie, a new node A), and steps one to three are repeated.

[0078] Step 5: Create a monitoring results list, use a breadth-first search algorithm to traverse all subordinate nodes of the monitoring project node to monitor the results nodes, and add them to the list;

[0079] In this step, breadth-first search helps to find all nodes reachable from the starting point in a shorter time.

[0080] Step 6: After the query is completed, the graph directory structure with vertex node A as the root node and all monitoring result node queues under it are obtained.

[0081] At this step, the structure and queue can be used for further analysis or presentation.

[0082] In summary, this method effectively traverses the security monitoring knowledge graph by combining the depth-first search and breadth-first search algorithms, especially for the query and collection of monitoring project nodes and their subordinate monitoring result nodes.

[0083] In the third aspect, this application provides a method for generating a safety monitoring data analysis report based on a knowledge graph. Applying the above construction method, please refer to Figure 5 , the generating method comprises:

[0084] S301: Determine the report content based on the data compilation and analysis requirements of the target project;

[0085] In this step, based on the data compilation and analysis requirements of the target project, it is determined what the report needs to include. This involves a comprehensive analysis of the safety monitoring data of the target project and determining which aspects of the data need to be included in the report.

[0086] S302: Select one or more entity nodes in the graph database as analysis objects according to the report content, and set the statistical time range of the report;

[0087] In this step, in a graph database (such as Neo4j), one or more entity nodes are selected as analysis objects according to the report content. These entity nodes represent the objects or concepts that need to be analyzed in the report.

[0088] Set the statistical time range of the report, that is, determine the time period for analyzing the monitoring data.

[0089] S303: searching the graph database using a graph query language according to the selected analysis object to obtain the analysis scope of the report;

[0090] In this step, according to the selected analysis object and time range, a graph query language (such as Cypher) is used to search in the graph database to obtain the analysis range required for the report. This step uses the relevance and query capabilities of the knowledge graph to quickly locate data related to the analysis object and within the time range.

[0091] S304: Based on the analysis scope and the set statistical time range, the monitoring results content in the report is generated, and the monitoring results content includes characteristic text, display graphics and statistical reports;

[0092] In this step, based on the retrieved analysis scope and the set statistical time range, the monitoring results content in the report is generated. These contents may include characteristic text, such as text descriptions of monitoring data, display graphics, such as line graphs, bar graphs, etc., and statistical reports, such as statistical analysis results of monitoring data. For example, the statistical time is set to 2020 and 2021, and the characteristic values ​​of the horizontal displacement monitoring points on the main monitoring section (AA) of the upstream slope of the main dam are shown in Table 1

[0093] Table 1 Horizontal displacement characteristic values ​​of the main dam section AA Unit: mm

[0094]

[0095] Please refer to the display graphics Figure 6 , Figure 6 It is a schematic diagram of the monitoring results, showing the statistical information of the monitoring data: it shows that the range of horizontal displacement changes of the measuring points on the AA section of the main dam during the year is 4.2 to 17.4 mm, 92% of the changes of the measuring points during the year are negative (showing upstream displacement changes), and the range of extreme amplitude changes during the year are 29.6 to 34.0 mm and 17.0 to 29.0 mm respectively.

[0096] S305: Based on the safety monitoring knowledge graph structure and the selected analysis scope, an outline of the safety monitoring data analysis report is generated, the outline directory is combined according to the association relationship in the safety monitoring data analysis report graph, and the monitoring results content is placed in the analysis report according to the outline;

[0097] In this step, the outline of the report is automatically generated based on the structure of the security monitoring knowledge graph and the selected analysis scope. The outline directory is combined according to the association relationship in the knowledge graph to ensure the logic and coherence of the report content. The monitoring results generated previously are placed according to the outline to ensure that the content of each part corresponds to its position in the outline.

[0098] S306: Write the generated report outline into the document file in the form of a document title directory; add the relevant monitoring results content to the corresponding directory of the document; edit the document style, generate a standardized safety monitoring data analysis report document and save it;

[0099] In this step, the generated report outline is written into the document file (such as Word, PDF, etc.) in the form of a document title directory. Add relevant monitoring results content, including feature text, display graphics and statistical reports, to the corresponding directory of the document. Edit the document style, including font, font size, paragraph format, chart layout, etc., to generate a standardized and beautiful safety monitoring data analysis report document.

[0100] S307: Export the generated safety monitoring data analysis report document to the location specified by the user, and perform integrity verification on the exported file after the export is completed.

[0101] In this step, the generated security monitoring data analysis report document is exported to the location specified by the user. This may be a local folder, a network share, or a cloud storage service. After the export is complete, the exported file is verified for integrity to ensure that all content has been exported correctly and nothing is missing.

[0102] In summary, this method greatly improves the efficiency and accuracy of report generation by automatically utilizing the information in the knowledge graph to generate reports, while also making the content of the report more comprehensive and the structure clearer.

[0103] In a fourth aspect, the present application provides a system for constructing a safety monitoring knowledge graph, including: an entity node determination module, which is used to extract entity information from the safety monitoring data of the target project and determine the entity nodes, wherein the entity nodes include monitoring object nodes, monitoring part nodes, monitoring project nodes and monitoring result nodes; a subordinate edge establishment module, which is used to determine the subordinate relationship between each entity node and connect them through edges; an attribute node generation module, which is used to generate a unique index identifier as an attribute value for the monitoring result node, and the attribute relationship between the attribute value and the monitoring result node is a one-to-one mapping relationship. The index identifier is used as an index or key to present the monitoring result content associated with the index identifier, and the monitoring result content is set as an attribute node, wherein the monitoring result content includes feature text, graphics, and statistical reports; a knowledge graph generation module, which is used to determine the attribute relationship between the monitoring result node and the attribute node, and connect them through edges to form a safety monitoring knowledge graph; a storage module, which is used to use the knowledge graph model to describe the association relationship between nodes to form a triple data structure, and save the safety monitoring knowledge graph in the form of a triple data structure to a graph database.

[0104] For the specific implementation of the system, please refer to the specific implementation of a method for constructing a security monitoring knowledge graph, which will not be elaborated here.

[0105] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for constructing a safety monitoring knowledge graph, characterized in that: include: Extract entity information from the safety monitoring data of the target project and determine entity nodes, wherein the entity nodes include monitoring object nodes, monitoring part nodes, monitoring project nodes and monitoring result nodes; Determine the subordinate relationship between each entity node and connect them through edges; Generate a unique index identifier as an attribute value for the monitoring achievement node, the attribute value and the attribute relationship of the monitoring achievement node are a one-to-one mapping relationship, use the index identifier as an index or key, present the monitoring achievement content associated with the index identifier, and set the monitoring achievement content as an attribute node, the monitoring achievement content includes characteristic text, graphics, and statistical reports; Determine the attribute relationship between monitoring result nodes and attribute nodes, connect them through edges, and form a safety monitoring knowledge graph; The knowledge graph model is used to describe the relationship between nodes to form a triple data structure, and the security monitoring knowledge graph is saved in the graph database in the form of a triple data structure.

2. The method according to claim 1, characterized in that The specific steps of saving the safety monitoring knowledge graph into the graph database in the form of a triple data structure include: The triple data is disassembled to obtain the information of entities, relationships and attributes. The entities, relationships and attributes are input into the Neo4j graph database using the graph query language, and an index is created in Neo4j to form a complete tree-structured security monitoring knowledge graph.

3. The method according to claim 1, characterized in that Safety monitoring data include design drawings and observation data.

4. The method according to claim 1, characterized in that: The specific steps of determining the subordinate relationship between each entity node include: The monitoring object-monitoring part, monitoring part-monitoring project, and monitoring project-monitoring results are all one-to-many hierarchical subordinate relationships.

5. A method for retrieving a safety monitoring knowledge graph, characterized in that: The construction method applied to any one of claims 1 to 4, wherein the retrieval method comprises: Step 1: Visit the initial vertex node A and mark node A as visited; Step 2: If node A is not a monitoring project node, execute step 3 to perform a depth-first traversal recursively on node A to find the subordinate node B of node A; if node A is a monitoring project node, execute step 5; Step 3: If node B exists, continue to step 4; if node B does not exist, return to step 1 and continue with the next subordinate node of node A; Step 4: If node B has not been visited and is not a monitoring project node, perform a depth-first traversal recursively on node B, treat node B as a new vertex node A, and execute steps 1 to 3; if node B is a monitoring project node, execute step 5; Step 5: Create a monitoring results list, use a breadth-first search algorithm to traverse all subordinate nodes of the monitoring project node to monitor the results nodes, and add them to the list; Step 6: After the query is completed, the graph directory structure with vertex node A as the root node and all monitoring result node queues under it are obtained.

6. A method for generating a safety monitoring data analysis report based on a knowledge graph, characterized in that: The construction method applied to any one of claims 1 to 4, wherein the generation method comprises: Determine the report content based on the data compilation and analysis requirements of the target project; Select one or more entity nodes in the graph database as analysis objects based on the report content, and set the statistical time range of the report; According to the selected analysis object, the graph query language is used to search in the graph database to obtain the analysis scope of the report; Based on the analysis scope and the set statistical time range, the monitoring results content in the report is generated, and the monitoring results content includes characteristic text, display graphics and statistical reports; Based on the safety monitoring knowledge graph structure and the selected analysis scope, an outline of the safety monitoring data analysis report is generated. The outline directory is combined according to the association relationship in the safety monitoring data analysis report graph, and the monitoring results are placed in the analysis report according to the outline; Write the generated report outline into the document file in the form of a document title directory; add the relevant monitoring results content to the corresponding directory of the document; edit the document style, generate a standardized safety monitoring data analysis report document and save it; The generated safety monitoring data analysis report document is exported to the location specified by the user, and the integrity of the exported file is verified after the export is completed.

7. A system for constructing a safety monitoring knowledge graph, characterized in that: include: An entity node determination module is used to extract entity information from the safety monitoring data of the target project and determine entity nodes, wherein the entity nodes include monitoring object nodes, monitoring part nodes, monitoring project nodes and monitoring result nodes; A subordinate edge establishment module is used to determine the subordinate relationship between each entity node and connect them through edges; The attribute node generation module is used to generate a unique index identifier as an attribute value for the monitoring achievement node. The attribute value and the attribute relationship of the monitoring achievement node are a one-to-one mapping relationship. The index identifier is used as an index or key to present the monitoring achievement content associated with the index identifier, and the monitoring achievement content is set as an attribute node. The monitoring achievement content includes characteristic text, graphics, and statistical reports; The knowledge graph generation module is used to determine the attribute relationship between the monitoring result nodes and the attribute nodes, connect them through edges, and form a safety monitoring knowledge graph; The storage module is used to use the knowledge graph model to describe the association relationship between nodes to form a triple data structure, and save the security monitoring knowledge graph in the form of a triple data structure into the graph database.

Images