A method for constructing a safety management knowledge graph

Abstract

A method for constructing a safety management knowledge graph provided by this application includes: obtaining target data required for constructing the safety management knowledge graph from a preset database; the preset database includes a coal preparation standard database, an accident case database, and / or an operating procedure database; the target data is data related to coal preparation; using a preset large model to identify target entities in the target data and generate corresponding triples; using a deep learning model to determine the relevance between target entities based on the triples; constructing a safety management knowledge graph according to the target entities and the relevance between each target entity. The method for constructing a safety management knowledge graph of this application can improve the query efficiency and the processing efficiency of safety problems by managing safety data in the way of constructing a knowledge graph.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method for constructing a security management knowledge graph. Background Art

[0002] With the continuous development of the coal industry, the amount of data generated in the coal preparation process is becoming increasingly large. This data is of great significance for improving coal preparation efficiency and ensuring production safety. However, traditional data management and analysis methods make it difficult to effectively integrate and utilize this information.

[0003] At present, traditional safety manuals, safety training and other materials on the market are all stored in databases or binary files as standard data. The knowledge obtained when consulting the data is relatively scattered, and the consulting efficiency is low, which affects the efficiency of handling safety issues.

[0004] Therefore, the current data management has the problem of low access efficiency, which affects the efficiency of handling security issues. Summary of the invention

[0005] The present application provides a method for constructing a security management knowledge graph to solve the problem that current data management has low access efficiency, which affects the efficiency of handling security issues.

[0006] The first aspect of the present application provides a method for constructing a security management knowledge graph, comprising:

[0007] Acquire target data required for constructing a safety management knowledge graph from a preset database; the preset database includes a coal preparation standard database, an accident case database and / or an operating procedure database; the target data is data related to coal preparation;

[0008] Using a preset large model to identify the target entity in the target data and generate a corresponding triple;

[0009] Determine the association between target entities based on the triples using a deep learning model;

[0010] Construct a security management knowledge graph based on the target entities and the associations between them.

[0011] Furthermore, in the above method, the step of obtaining target data required for constructing a security management knowledge graph from a preset database includes:

[0012] Acquire corresponding initial data from the coal preparation standard database; the initial data includes equipment information, operating procedure data, and emergency treatment measures data of the coal preparation industry;

[0013] Remove noise and process missing values ​​from the initial data to generate corresponding intermediate data;

[0014] In response to the user's approval instruction for the intermediate data, determine the corresponding approved intermediate data as the target data.

[0015] Further, in the method described above, the step of using a preset large model to identify target entities in the target data and generate corresponding triples includes:

[0016] Perform normalization processing on the target data to generate standard data corresponding to the target data;

[0017] Input the preset coal preparation term categories, preset prompt words, and standard data into a preset large model to extract target entities in the standard data by using named entity recognition, and generate the triples by using natural language processing; the preset coal preparation term categories include coal types, coal preparation processes, and pollutant emissions.

[0018] Further, in the method described above, the step of using a deep learning model to determine the relevance between target entities based on the triples includes:

[0019] Input the triples into a deep learning model, and use word vector technology to identify the similarity between different target entities to determine the relevance.

[0020] Further, in the method described above, after constructing the safety management knowledge graph according to the target entities and the relevance between each target entity, it further includes:

[0021] In response to the user's knowledge retrieval operation on the safety management knowledge graph, retrieve and determine whether the retrieved corresponding data is abnormal;

[0022] If it is determined that the corresponding data is abnormal, give a warning prompt to the user.

[0023] Further, in the method described above, before obtaining the target data required for constructing the safety management knowledge graph from a preset database, it further includes:

[0024] Obtain the basic knowledge of coal preparation standards from a knowledge collection platform; the basic knowledge is manually collected and input into the knowledge collection platform;

[0025] Use a large model to augment the basic knowledge to generate the required coal preparation standard data;

[0026] Store the coal preparation standard data in a preset database.

[0027] The second aspect of this application provides a safety management knowledge graph construction device, including:

[0028] An acquisition module for acquiring target data required for constructing a safety management knowledge graph from a preset database; the preset database includes a coal preparation standard database, an accident case database, and / or an operation procedure database; the target data is coal preparation-related data;

[0029] An identification module for identifying target entities in the target data using a preset large model and generating corresponding triples;

[0030] A determination module for determining the relevance between target entities based on the triples using a deep learning model;

[0031] A construction module for constructing a safety management knowledge graph based on the target entities and the relevance between each target entity.

[0032] Further, for the device as described above, the acquisition module is specifically used for:

[0033] Acquiring corresponding initial data from the coal preparation standard database; the initial data includes equipment information, operation procedure data, and emergency treatment measure data in the coal preparation industry; performing noise removal and missing value processing on the initial data to generate corresponding intermediate data; and in response to the user's approval instruction for the intermediate data, determining the passed intermediate data as the target data.

[0034] Further, for the device as described above, the identification module is specifically used for:

[0035] Performing standardization processing on the target data to generate standard data corresponding to the target data; inputting the preset coal preparation term categories, preset prompt words, and the standard data into the preset large model to extract target entities in the standard data by using the named entity recognition method and generate the triples by using the natural language processing method; the preset coal preparation term categories include coal types, coal preparation processes, and pollutant emissions.

[0036] Further, for the device as described above, the determination module is specifically used for:

[0037] Inputting the triples into the deep learning model and using the word vector technology to identify the similarity between different target entities to determine the relevance.

[0038] Further, for the device as described above, the device further includes:

[0039] An early warning module for, in response to the user's knowledge retrieval operation on the safety management knowledge graph, retrieving and determining whether the retrieved corresponding data is abnormal; if it is determined that the corresponding data is abnormal, giving an early warning prompt to the user.

[0040] Further, for the device as described above, the device further includes:

[0041] A knowledge collection module is used to obtain the basic knowledge of coal preparation standards from a knowledge collection platform; the basic knowledge is manually collected and input into the knowledge collection platform; a large model is used to augment the basic knowledge to generate the required coal preparation standard data; the coal preparation standard data is stored in a preset database.

[0042] A third aspect of this application provides an electronic device, including: a memory and a processor;

[0043] The memory stores computer-executable instructions;

[0044] The processor executes the computer-executable instructions stored in the memory to implement the security management knowledge graph construction method according to any one of the first aspects.

[0045] A fourth aspect of this application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the security management knowledge graph construction method according to any one of the first aspects.

[0046] A fifth aspect of this application provides a computer program product, including a computer program, which when executed by a processor implements the security management knowledge graph construction method according to any one of the first aspects.

[0047] A security management knowledge graph construction method provided by this application includes: obtaining target data required for constructing a security management knowledge graph from a preset database; the preset database includes a coal preparation standard database, an accident case database, and / or an operation procedure database; the target data is coal preparation-related data; using a preset large model to identify target entities in the target data and generate corresponding triples; using a deep learning model to determine the relevance between target entities based on the triples; constructing a security management knowledge graph according to the target entities and the relevance between each target entity. The security management knowledge graph construction method of this application obtains the target data required for constructing a security management knowledge graph from a preset database, uses a preset large model to identify target entities in the target data, and generates corresponding triples. At the same time, a deep learning model is used to determine the relevance between target entities based on the triples, so that a security management knowledge graph can be constructed according to the target entities and the relevance between each target entity. By managing security data in the form of a knowledge graph, the query efficiency and the processing efficiency of security problems can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings here are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application.

[0049] Figure 1 It is a scenario diagram for implementing the security management knowledge graph construction method of the embodiments of the present application;

[0050] Figure 2 It is a schematic flowchart of the security management knowledge graph construction method provided by the present application Figure 1 ;

[0051] Figure 3 It is a schematic flowchart of the security management knowledge graph construction method provided by the present application Figure 2 ;

[0052] Figure 4 It is a schematic diagram of the overall process of the security management knowledge graph construction method provided by the present application;

[0053] Figure 5 It is a schematic diagram of the structure of the security management knowledge graph construction device provided by the present application;

[0054] Figure 6 It is a schematic diagram of the structure of the electronic device provided by the present application.

[0055] Through the above-mentioned drawings, the specific embodiments of the present application have been shown, and there will be more detailed descriptions hereinafter. These drawings and text descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Specific Embodiments

[0056] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0057] It should be noted that a security management knowledge graph construction method of the present disclosure can be used in the field of artificial intelligence technology. It can also be used in any field other than the field of artificial intelligence technology. The application field of a security management knowledge graph construction method of the present disclosure is not limited.

[0058] Hereinafter, the technical solutions of the present application will be described in detail with specific embodiments. The following several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. Hereinafter, the embodiments of the present application will be described with reference to the drawings.

[0059] To clearly understand the technical solution of this application, the concept of this solution will be introduced in detail first. Traditional data management methods are difficult to cope with the complex data environment in coal preparation plants, and there are problems such as data islands and information asymmetry.

[0060] With the continuous development of the coal industry, the amount of data generated in the coal preparation process is increasing day by day. These data are of great significance for improving the coal preparation efficiency and ensuring production safety. However, traditional data management and analysis methods are difficult to effectively integrate and utilize this information.

[0061] Currently, traditional safety manuals, safety training materials, etc. on the market are stored in databases or binary files in standard data. The knowledge obtained when consulting data is relatively scattered, and the consulting efficiency is relatively low, which affects the processing efficiency of safety issues.

[0062] Therefore, the current data management has the problem of low consulting efficiency, which affects the processing efficiency of safety issues. So, aiming at the problems in the existing technology, the inventor found in the research that to solve this problem, the safety knowledge data of coal preparation can be integrated by means of a knowledge graph, thereby improving the information consulting efficiency. At the same time, through the knowledge graph, users can quickly locate safety issues, and thus also improve the processing efficiency of safety issues. Among them, the Knowledge Graph is a technology used to represent and organize information, aiming to describe entities and their mutual relationships in the form of nodes and edges.

[0063] Specifically, obtain the target data required to construct the safety management knowledge graph from a preset database. The preset database includes a coal preparation standard database, an accident case database, and / or an operation procedure database. The target data is data related to coal preparation. Use a preset large model to identify the target entities in the target data and generate corresponding triples. Use a deep learning model to determine the relevance between the target entities based on the triples. Construct a safety management knowledge graph according to the target entities and the relevance between each target entity.

[0064] The method for constructing the safety management knowledge graph of this application obtains the target data required to construct the safety management knowledge graph from a preset database, uses a preset large model to identify the target entities in the target data, and generates corresponding triples. At the same time, use a deep learning model to determine the relevance between the target entities based on the triples, so that a safety management knowledge graph can be constructed according to the target entities and the relevance between each target entity. By managing safety data in the way of constructing a knowledge graph, the query efficiency and the processing efficiency of safety issues can be improved.

[0065] Based on the above creative discovery, the inventor proposed the technical solution of this application.

[0066] The application scenarios of the method for constructing a safety management knowledge graph provided in the embodiments of the present application are introduced below. As Figure 1 shown, where 1 is the first electronic device and 2 is the second electronic device. Among them, the first electronic device 1 can be a computer, a tablet computer, etc. The second electronic device 2 can be various database servers, etc. The database server can be a coal standard database, an accident case database, an operation procedure database, etc.

[0067] Exemplarily, when constructing a safety management knowledge graph, the first electronic device 1 obtains the target data required for constructing the safety management knowledge graph from the second electronic device 2, where the target data is data related to coal preparation. That is, ① the second electronic device 2 sends the target data to the first electronic device 1, and the first electronic device 1 performs the following processing:

[0068] ② Use a preset large model to identify the target entities in the target data and generate corresponding triples.

[0069] ③ Use a deep learning model to determine the relevance between the target entities based on the triples.

[0070] ④ Construct a safety management knowledge graph according to the target entities and the relevance between the target entities.

[0071] After constructing the safety management knowledge graph, the safety management knowledge graph can be further audited and verified by means of expert review to improve the accuracy of the safety management knowledge graph.

[0072] The embodiments of the present application are introduced below with reference to the accompanying drawings of the specification.

[0073] Figure 2 is a flowchart of the method for constructing a safety management knowledge graph provided by the present application Figure 1 As Figure 2 shown, in this embodiment, the execution subject of the embodiments of the present application is a safety management knowledge graph construction device, and the safety management knowledge graph construction device can be integrated in an electronic device, such as a tablet computer, a computer, etc. Then the method for constructing a safety management knowledge graph provided in this embodiment includes the following steps:

[0074] S101, obtain the target data required for constructing the safety management knowledge graph from a preset database. The preset database includes a coal preparation standard database, an accident case database, and / or an operation procedure database. The target data is data related to coal preparation.

[0075] In this embodiment, the preset database can be a combination of a coal standard database, an accident case database, and an operating procedure database, or any one of them can be used. The coal standard database is used to store standard data for coal preparation, such as data on coal types, coal preparation processes, etc. The accident case database is used to store case data related to safety occurrences, and the operating procedure database is used to store operating procedures related to safety.

[0076] S102, Use the preset large model to identify the target entities in the target data and generate corresponding triples.

[0077] In this embodiment, the preset large model can be a language large model. The language large model can use natural language processing (NLP) technology and named entity recognition (NER) technology to identify the target entities in the target data and generate corresponding triples. Named entity recognition technology is a technical task in natural language processing (NLP) aimed at identifying and classifying named entities from text. Named entities usually include words or phrases in specific categories such as person names, place names, organization names, time expressions, quantity expressions, etc. In this embodiment, the named entities can be entities such as coal types, coal preparation processes, and safety specifications.

[0078] In this embodiment, a triple is a basic structure for representing knowledge. A triple consists of three parts: a subject, a predicate, and an object. This structure is usually used to describe entities and their relationships, similar to a simple sentence in natural language. Generating the triples corresponding to the target data can provide a basis for determining the relationships between entities in the subsequent steps.

[0079] S103, Use a deep learning model to determine the relevance between the target entities based on the triples.

[0080] In this embodiment, the deep learning model can use open-source versions of models such as BERT (Bidirectional Encoder Representations from Transformers) and GPT (Generative Pre-trained Transformer) to improve the effect of determining the relevance between the target entities.

[0081] S104, Construct a safety management knowledge graph based on the target entities and the relevance between each target entity.

[0082] In this embodiment, a basic framework schema can be created based on the target entities and the relationships between the target entities. According to the schema, tools such as Excel can be used to standardize the cleaned data. For example, the correspondence between rows and columns is mapped into corresponding entity types.

[0083] Based on the above knowledge representation, a safety management knowledge graph is constructed. The graph represents entities through nodes and relationships between entities through edges. Then, the extracted data is presented in the form of tables and Excel. Using the discrimination ability of the large model or manual review, the data with high credibility is selected and then stored in the graph database, such as Neo4j, to store and manage the knowledge graph, thus forming a knowledge network containing rich semantic information.

[0084] Optionally, the latest standards and safety management information can also be obtained from the coal preparation standard database regularly to automatically update the knowledge graph and ensure the timeliness and accuracy of the graph. At the same time, a maintenance mechanism is established to correct the error information in the graph and improve the graph structure.

[0085] A method for constructing a safety management knowledge graph provided by an embodiment of the present application includes: obtaining target data required for constructing the safety management knowledge graph from a preset database. The preset database includes a coal preparation standard database, an accident case database, and / or an operation procedure database. The target data is coal preparation-related data. A preset large model is used to identify target entities in the target data and generate corresponding triples. A deep learning model is used to determine the relationships between the target entities based on the triples. A safety management knowledge graph is constructed according to the target entities and the relationships between the target entities.

[0086] The method for constructing a safety management knowledge graph of the present application obtains target data required for constructing the safety management knowledge graph from a preset database, uses a preset large model to identify target entities in the target data, and generates corresponding triples. At the same time, a deep learning model is used to determine the relationships between the target entities based on the triples, so that a safety management knowledge graph can be constructed according to the target entities and the relationships between the target entities. By managing safety data in the form of a knowledge graph, the query efficiency and the processing efficiency of safety problems can be improved.

[0087] Figure 3 It is a flowchart of the method for constructing a safety management knowledge graph provided by the present application Figure 2 , as Figure 3 shown, the method for constructing a safety management knowledge graph provided in this embodiment is a further refinement based on the method for constructing a safety management knowledge graph provided in the previous embodiment of the present application. Then, the method for constructing a safety management knowledge graph provided in this embodiment includes the following steps.

[0088] S201, Obtain the corresponding initial data from the coal preparation standard database. The initial data includes equipment information, operating procedure data, and emergency treatment measure data in the coal preparation industry.

[0089] In this embodiment, the operating procedure data includes data such as equipment operation and coal preparation process. The equipment information in the coal preparation industry includes various equipment information in the coal preparation industry, such as screening equipment, analyzers, etc. The emergency treatment measure data includes corresponding emergency treatment measure data in case of safety problems, such as how to operate the equipment and how to perform self-rescue.

[0090] Optionally, in this embodiment, before S201, it may also include the process of knowledge expansion and collection, which is specifically as follows:

[0091] Obtain the basic knowledge of coal preparation standards from the knowledge collection platform. The basic knowledge is manually collected and input into the knowledge collection platform.

[0092] Use a large model to augment the basic knowledge to generate the required coal preparation standard data.

[0093] Store the coal preparation standard data in a preset database.

[0094] In this embodiment, the person can be an expert in the field of coal preparation, and input the knowledge in the collected field into the knowledge collection platform.

[0095] In this embodiment, using a large model to augment the basic knowledge can augment synonyms or antonyms, thereby expanding the capacity of the basic knowledge and making the finally generated coal preparation standard data cover a wider range, providing a better data capacity basis for the subsequent generation of the knowledge graph.

[0096] For professional terms, they can be vectorized and stored in a vector database in the form of documents or slices, and then train the large model to understand and perform self-augmentation of synonyms.

[0097] When performing augmentation, it is also possible to further review and screen the augmented data through the method of expert review and verification, and retain the available data in the field of coal preparation.

[0098] S202, Perform noise removal and missing value processing on the initial data to generate corresponding intermediate data.

[0099] In this embodiment, when performing noise removal and missing value processing on the initial data, a verification process can also be added. For example, use the discrimination ability of the large model and experts to review the knowledge to ensure the reliability of the intermediate data, thereby improving the accuracy of the knowledge graph.

[0100] S203. In response to the user's approval instruction for the intermediate data, determine the corresponding approved intermediate data as the target data.

[0101] In this embodiment, through manual means, such as experts further reviewing the intermediate data, the reliability of the intermediate data can be improved, thereby improving the accuracy of the knowledge graph.

[0102] S204. Use a preset large model to identify the target entities in the target data and generate corresponding triples.

[0103] Optionally, in this embodiment, S204 can be specifically as follows:

[0104] Perform standardization processing on the target data to generate standard data corresponding to the target data.

[0105] Input the preset coal preparation term categories, preset prompt words, and standard data into the preset large model to extract the target entities in the standard data by using the named entity recognition method and generate triples by using the natural language processing method. The preset coal preparation term categories include coal types, coal preparation processes, and pollutant emissions.

[0106] In this embodiment, a) Coal types include: anthracite, bituminous coal, and lignite. Note: The characteristics and application fields of different coal types, as well as their impacts on equipment and processes during coal preparation.

[0107] b) Coal preparation processes include: heavy medium separation, flotation, and screening. Note: The process descriptions, applicable conditions of each process, and their impacts on coal type selection and product quality.

[0108] c) Pollutant emissions include: sulfur dioxide, nitrogen oxides, and dust. Note: Emission standards, detection methods, and potential impacts on the environment.

[0109] In this embodiment, examples of the preset prompt words are as follows:

[0110] Extract all entity pairs that meet the conditions and the corresponding relationship types from the given document. Your generated content must meet all the following conditions.

[0111] 1. The entity types only include {entity_list}.

[0112] {entity_detail}

[0113] 2. The relationship types only include {relation_list}.

[0114] 3. The entity types and relationship types corresponding to the entity pairs can only have the following several construction modes:

[0115] {triplet_detail}

[0116] 4. Entities must be clearly mentioned in the document. Only two entities with a certain relationship can be regarded as an entity pair.

[0117] 5. Output format: A list composed of JSON-formatted data. Note that duplicate content should not be output, and a complete list should be output.

[0118] [ {"head entity": "", "head entity type": "", "relationship type": "", "tail entity": "", "tail entity type": ""},{"head entity": "", "head entity type": "", "relationship type": "", "tail entity": "","tail entity type": ""}]

[0119] S205. Input the triples into the deep learning model, and use the word vector technology to identify the similarity between different target entities to determine the relevance.

[0120] In this embodiment, since the technical terms in the coal preparation field are relatively professional, for synonyms or ambiguous words in the coal preparation field, the word vector technology such as Word2Vec or GloVe can be used to identify the similarity between different target entities and establish a mapping relationship, so as to determine the relevance between similar target entities.

[0121] For non-similar target entities with logical relevance or specific relevance in the coal preparation field, the relevance between each entity can be directly determined through a pre-trained deep learning model.

[0122] S206. Construct a safety management knowledge graph based on the target entities and the relevance between each target entity.

[0123] Optionally, in this embodiment, after S206, an early warning function can also be added when actually using the knowledge graph, as follows:

[0124] In response to the user's knowledge retrieval operation on the safety management knowledge graph, retrieve and determine whether the retrieved corresponding data is abnormal.

[0125] If it is determined that the corresponding data is abnormal, an early warning prompt will be given to the user.

[0126] In this embodiment, the data in the knowledge graph can be updated regularly. At the same time, for the corresponding data retrieved by the user, it can be determined whether it is a normal value or an abnormal value, so that when the data is abnormal, an early warning prompt can be given to the user to avoid causing safety-related impacts.

[0127] Optionally, in this embodiment, it is also possible to use a large model and a large-screen display system at the user's site to monitor in real time whether the specific data displayed on the large screen is a normal value or an abnormal value. The background server actively calls the atlas data for query and detection, and gives possible reasons and solutions. The user can also actively ask about the reasons for the anomalies and solutions in the large model system.

[0128] To further illustrate the method for constructing a safety management knowledge atlas of the present application, the following will be combined with Figure 4 to illustrate the overall construction process.

[0129] As Figure 4 shown, the method for constructing a safety management knowledge atlas in this embodiment mainly includes six processes:

[0130] 1. Data collection: Collect relevant data from the coal preparation standard database, including equipment information, operating procedures, safety standards, and other professional books, etc. Establish a knowledge collection platform, set up a knowledge expert system to use the knowledge collection platform to collect basic knowledge. With the help of a large model, summarize and augment the basic data to obtain high-quality data that meets the field requirements.

[0131] 2. Data cleaning: Remove noise and irrelevant information, handle missing values, and ensure the quality of the data. The discrimination ability of the large model and the expert's review of knowledge need to be added in this part to ensure the reliability of the subsequent processed data.

[0132] 3. Entity recognition: Manually or with the help of a large model, identify key entities in the data, such as names of people, place names, organizations, etc.

[0133] 4. Extract triples: Based on the preset coal preparation term categories, preset prompt words, and standard data, generate triples through a large model.

[0134] 5. Relationship extraction: Analyze the relevance between entities and extract the relationships between them.

[0135] 6. Atlas construction: Integrate the above information into an atlas form, using nodes to represent entities and edges to represent the relationships between entities.

[0136] As Figure 4 shown, the data collection and data cleaning steps can be achieved manually or by using equipment automation methods to ensure the reliability and standardization of the data. This is the basis for constructing a high-quality and highly credible atlas. Collect data related to safety management from the coal preparation standard database, including but not limited to operating procedures, risk assessment standards, emergency treatment measures, etc. Clean, format, and standardize the collected data to ensure the consistency and integrity of the data. In this step, manual labor can use third-party tools to achieve it.

[0137] In the entity recognition step, technicians can start to intervene and perform standardization processing on the obtained data or materials. For example, operations such as document format conversion and document content acquisition need to be carried out through programs. Only after obtaining the data can entity recognition be performed. For example, we can abstract and extract the corresponding nouns from text materials. This process can use natural language processing (NLP) technology to extract key elements from the data.

[0138] In the process of associating entities, the following methods can also be adopted: 1. Reinforcement learning: Introduce reinforcement learning technology to optimize the decision-making of the model in the process of association and extraction. 2. Expert knowledge feedback: Establish an expert review mechanism to regularly review and correct the extracted entities and relationships to improve accuracy. 3. Dynamic update mechanism: Build a mechanism to timely introduce new terms and technological developments to keep the knowledge graph up-to-date.

[0139] In addition, a graph database (such as Neo4j) can be used to store and manage the knowledge graph, build a data layer and a schema layer, and define entities, relationships, and attributes.

[0140] The method for constructing a safety management knowledge graph in this embodiment realizes the automatic understanding and knowledge extraction of standard texts and operating procedures by introducing natural language processing technology, thereby reducing the workload of manual processing. At the same time, it solves the problems that it is difficult to apply and retrieve safety production knowledge. In addition, safety management knowledge can be classified, the essence can be abstracted, and the association relationships of knowledge can be found. Thus, it can more efficiently solve safety production problems and the standardization problems in the data management of coal preparation plants, and improve data management efficiency.

[0141] In implementation, the method of this embodiment automatically generates a safety management knowledge graph by scraping data from a coal preparation standard database, an accident case database, and operating procedure documents, combined with natural language processing technology. Users can query the safety specifications and risk assessment results related to a specific operation or equipment through the system interface. The system will generate safety management suggestions based on the information in the graph and issue early warnings when potential risks are detected. Through this system, managers can more efficiently monitor and manage safety risks in the coal preparation process to ensure the continuous stability and safety of production.

[0142] Figure 5 It is a schematic structural diagram of a safety management knowledge graph construction device provided by this application. As Figure 5 shown, in this embodiment, the safety management knowledge graph construction device 300 can be set in an electronic device. The safety management knowledge graph construction device 300 includes:

[0143] An acquisition module 301 is configured to acquire target data required for constructing a safety management knowledge graph from a preset database. The preset database includes a coal preparation standard database, an accident case database, and / or an operation procedure database. The target data is data related to coal preparation.

[0144] An identification module 302 is configured to identify target entities in the target data by using a preset large model and generate corresponding triples.

[0145] A determination module 303 is configured to determine the relevance between target entities based on the triples by using a deep learning model.

[0146] A construction module 304 is configured to construct a safety management knowledge graph according to the target entities and the relevance between each target entity.

[0147] The safety management knowledge graph construction device provided in this embodiment can execute Figure 2 the technical solution of the method embodiment shown, and its implementation principle and technical effect are similar to those of Figure 2 the method embodiment shown, which will not be elaborated here one by one.

[0148] Based on the safety management knowledge graph construction device provided in the previous embodiment, the safety management knowledge graph construction device provided in this application further refines the safety management knowledge graph construction device. The safety management knowledge graph construction device 300 includes:

[0149] Optionally, in this embodiment, the acquisition module 301 is specifically configured to:

[0150] Acquire corresponding initial data from the coal preparation standard database. The initial data includes equipment information, operation procedure data, and emergency treatment measure data in the coal preparation industry. Perform noise removal and missing value processing on the initial data to generate corresponding intermediate data. In response to the user's approval instruction for the intermediate data, determine the passed intermediate data as the target data.

[0151] Optionally, in this embodiment, the identification module 302 is specifically configured to:

[0152] Perform standardization processing on the target data to generate standard data corresponding to the target data. Input the preset coal preparation term category, preset prompt words, and standard data into the preset large model to extract target entities in the standard data by using the named entity recognition method and generate triples by using the natural language processing method. The preset coal preparation term category includes coal type, coal preparation process, and pollutant emission.

[0153] Optionally, in this embodiment, the determination module 303 is specifically configured to:

[0154] Input the triples into a deep learning model and use word vector technology to identify the similarity between different target entities to determine the relevance.

[0155] Optionally, in this embodiment, the security management knowledge graph construction device 300 further includes:

[0156] An early warning module, configured to respond to a user's knowledge retrieval operation on the security management knowledge graph, retrieve and determine whether the retrieved corresponding data is abnormal. If it is determined that the corresponding data is abnormal, an early warning prompt is sent to the user.

[0157] Optionally, in this embodiment, the security management knowledge graph construction device 300 further includes:

[0158] A knowledge collection module, configured to obtain basic knowledge of coal preparation standards from a knowledge collection platform. The basic knowledge is manually collected and input into the knowledge collection platform. Use a large model to augment the basic knowledge to generate the required coal preparation standard data. Store the coal preparation standard data in a preset database.

[0159] The security management knowledge graph construction device provided in this embodiment can execute Figures 2 - 4 the technical solutions of the method embodiment shown, and its implementation principle and technical effects are similar to those of Figures 2 - 4 the method embodiment shown, and will not be elaborated here one by one.

[0160] According to the embodiments of the present application, the present application also provides an electronic device, a computer-readable storage medium, and a computer program product.

[0161] As Figure 6 shown, Figure 6 is a schematic structural diagram of the electronic device provided by the present application. The electronic device is intended to be various forms of electronic devices, such as a tablet computer, a computer, etc. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described herein and / or required.

[0162] As Figure 6 shown, the electronic device includes: a processor 401 and a memory 402. Each component is interconnected using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the electronic device.

[0163] The memory 402 is the non-transitory computer-readable storage medium provided by this application. Among them, the memory stores instructions executable by at least one processor, so that at least one processor executes the security management knowledge graph construction method provided by this application. The non-transitory computer-readable storage medium of this application stores computer instructions, and these computer instructions are used to make a computer execute the security management knowledge graph construction method provided by this application.

[0164] As a non-transitory computer-readable storage medium, the memory 402 can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the security management knowledge graph construction method in the embodiments of this application (for example, the acquisition module 301, the recognition module 302, the determination module 303, and the construction module 304 shown in the appendix). Figure 5 By running the non-transitory software programs, instructions, and modules stored in the memory 402, the processor 401 can execute various functional applications and data processing of the electronic device, that is, implement the security management knowledge graph construction method in the above method embodiments.

[0165] In the above embodiments, it should be understood that the processor can be a central processing unit (English: Central Processing Unit, abbreviated: CPU), and can also be other general-purpose processors, digital signal processors (English: Digital Signal Processor, abbreviated: DSP), application-specific integrated circuits (English: Application Specific Integrated Circuit, abbreviated: ASIC), etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.

[0166] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0167] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience in representation, the buses in the drawings of this application are not limited to only one bus or one type of bus.

[0168] Meanwhile, this embodiment also provides a computer product. When the instructions in the computer product are executed by the processor of the electronic device, the electronic device can execute the security management knowledge graph construction method of the above embodiment.

[0169] The above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc. The readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0170] An exemplary readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an Application Specific Integrated Circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in the device.

[0171] The division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be an indirect coupling or communication connection through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0172] The unit described as a separate component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0173] In addition, each functional unit in various embodiments of the present invention may be integrated in a processing unit, may exist physically separately as individual units, or two or more units may be integrated in one unit.

[0174] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs and other various media that can store program codes.

[0175] Those of ordinary skill in the art can understand that all or part of the steps to implement the above method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments. The aforementioned storage medium includes: ROM, RAM, magnetic disks, or optical discs and other various media that can store program codes.

[0176] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other implementations of the embodiments of this application. This application is intended to cover any variations, uses, or adaptations of the embodiments of this application, which follow the general principles of the embodiments of this application and include common general knowledge or conventional technical means in the technical field not disclosed in the embodiments of this application.

[0177] It should be understood that the embodiments of this application are not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the embodiments of this application is only limited by the appended claims.

Claims

1. A method for constructing a safety management knowledge graph, characterized in that: include: Acquire basic knowledge of coal preparation standards from a knowledge collection platform; the basic knowledge is manually collected and input into the knowledge collection platform; The basic knowledge is augmented by using a large model, and the augmented data is reviewed and verified by experts in the field of coal preparation to generate the required coal preparation standard data; Storing coal preparation standard data in a preset database; Acquire target data required for constructing a safety management knowledge graph from a preset database; the preset database includes a coal preparation standard database, an accident case database and / or an operating procedure database; the target data is data related to coal preparation; Performing standardization processing on the target data to generate standard data corresponding to the target data; Input the preset coal preparation term categories, preset prompt words and standard data into the preset large model, extract the target entities in the standard data by using named entity recognition, and generate corresponding triples by using natural language processing; The preset coal preparation term categories include coal types, coal preparation processes, and pollutant emissions; The coal types include: anthracite, bituminous coal, lignite; the coal preparation process includes: heavy medium separation, flotation, screening; the pollutant emissions include: sulfur dioxide, nitrogen oxides, dust; Inputting the triples into a deep learning model and using word vector technology to identify similarities between different target entities to determine the relevance between the target entities; Construct a security management knowledge graph based on target entities and the associations between target entities; In response to the user's knowledge retrieval operation on the security management knowledge graph, the corresponding data retrieved is retrieved and judged whether it is abnormal; if it is determined that the corresponding data is abnormal, an early warning prompt is given to the user.

2. The method according to claim 1, characterized in that The step of obtaining target data required for constructing a security management knowledge graph from a preset database includes: Acquire corresponding initial data from the coal preparation standard database; the initial data includes equipment information, operating procedure data, and emergency treatment measures data of the coal preparation industry; Remove noise and process missing values ​​from the initial data to generate corresponding intermediate data; In response to a user's instruction to approve the review of the intermediate data, the corresponding approved intermediate data is determined as target data.

3. A security management knowledge graph construction device, characterized in that: include: An acquisition module is used to obtain target data required for building a security management knowledge graph from a preset database; The preset database includes a coal preparation standard database, an accident case database and / or an operating procedure database; the target data is data related to coal preparation; An identification module, used for performing standardization processing on the target data to generate standard data corresponding to the target data; Input the preset coal preparation term categories, preset prompt words and standard data into the preset large model, extract the target entities in the standard data by using named entity recognition, and generate corresponding triples by using natural language processing; The preset coal preparation term categories include coal types, coal preparation processes, and pollutant emissions; The coal types include: anthracite, bituminous coal, lignite; the coal preparation process includes: heavy medium separation, flotation, screening; the pollutant emissions include: sulfur dioxide, nitrogen oxides, dust; A determination module, used for inputting the triples into a deep learning model and using word vector technology to identify similarities between different target entities to determine the relevance between the target entities; A construction module is used to construct a security management knowledge graph based on target entities and the associations between target entities; The device also includes: The knowledge collection module is used to obtain the basic knowledge of coal preparation standards from the knowledge collection platform; the basic knowledge is manually collected and input into the knowledge collection platform; the basic knowledge is augmented by a large model, and the augmented data is reviewed and verified by experts in the coal preparation field to generate the required coal preparation standard data; the coal preparation standard data is stored in a preset database.

4. An electronic device, characterized in that: include: Memory and processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the security management knowledge graph construction method as described in claim 1 or 2.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the security management knowledge graph construction method as described in claim 1 or 2.

6. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the security management knowledge graph construction method as described in claim 1 or 2.

Images