Operation specification query method, computer equipment and computer storage medium

Abstract

The embodiment of the invention discloses an operation specification query method, computer equipment and a computer storage medium. Constructing a knowledge graph corresponding to the operation specification data, performing vectorization representation on each entity in the knowledge graph to obtain a feature vector of each entity in the knowledge graph, obtaining an initial query model, and training the initial query model based on the feature vectors of the plurality of entities in the knowledge graph and a logical relationship among the plurality of entities in the knowledge graph to obtain a query result; and identifying a query statement input by the user based on the target query model obtained by training, and outputting an operation specification query result. By training the query model based on the knowledge graph and the entity feature vectorization and answering the operation queried by the user based on the query model to output the query result of the operation specification, the user can quickly obtain the corresponding operation specification only by inputting any keyword without consuming a lot of time to search the operation specification; the query efficiency of the operation specification can be improved, and the accuracy of the query result can be ensured.

Description

Technical Field

[0001] This application relates to the field of data query, specifically to an operation specification query method, computer equipment, and computer storage medium. Background Technology

[0002] In the process of car repair, it is inevitable to consult the bolt tightening specifications for various system components of the vehicle and to tighten or remove the components accordingly. For example, vehicle repair technicians currently need to frequently consult the bolt torque specifications for different vehicle models and different components during the repair process.

[0003] Current mainstream methods for checking bolt tightening standards include consulting the manufacturer's manual, searching online, and contacting the vehicle manufacturer's customer service. However, these methods all suffer from problems such as low efficiency or inaccurate results. Summary of the Invention

[0004] This application provides an operation specification query method, computer device, and computer storage medium. By training a query model based on knowledge graphs and entity feature vectorization, and answering the user's query based on the query model, the query results of operation specifications are output. This eliminates the need to spend a lot of time searching for operation specifications and improves the query efficiency of operation specifications.

[0005] The first aspect of this application provides a method for querying operational specifications, including:

[0006] Obtain operational specification data;

[0007] Construct a knowledge graph corresponding to the operation specification data, the knowledge graph including multiple entities in the operation specification data and the logical relationships between the multiple entities;

[0008] Each entity in the knowledge graph is vectorized to obtain the feature vector of each entity in the knowledge graph;

[0009] Obtain an initial query model, and train the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between the multiple entities in the knowledge graph to obtain the target query model;

[0010] Based on the target query model, the system identifies the query statement entered by the user and outputs the query results with operational guidelines.

[0011] A second aspect of this application provides a computer device, including:

[0012] The acquisition unit is used to acquire operational specification data;

[0013] A construction unit is used to construct a knowledge graph corresponding to the operation specification data, wherein the knowledge graph includes multiple entities in the operation specification data and the logical relationships between the multiple entities.

[0014] A vectorization unit is used to perform vectorization representation on each entity in the knowledge graph to obtain the feature vector of each entity in the knowledge graph;

[0015] The model training unit is used to obtain an initial query model and train the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between the multiple entities in the knowledge graph to obtain the target query model.

[0016] The query unit is used to identify the query statement input by the user based on the target query model and output the operation specification query result.

[0017] A third aspect of this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method of the first aspect described above.

[0018] A fourth aspect of this application provides a computer storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described in the first aspect.

[0019] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0020] A knowledge graph corresponding to the operational specification data is constructed. Each entity in the knowledge graph is vectorized to obtain its feature vector. An initial query model is then obtained and trained based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between them. The trained target query model is then used to identify the user's input query and output the operational specification query results. Therefore, by training a query model based on a knowledge graph and entity feature vectorization, and then using this model to answer the user's query for operational specifications, users can quickly obtain the corresponding operational specifications simply by entering any keywords, without spending a lot of time searching for them. This improves the efficiency of operational specification queries and ensures the accuracy of the results. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the operation specification query method in this application embodiment;

[0022] Figure 2 This is another flowchart illustrating the operation specification query method in this application embodiment;

[0023] Figure 3 This is a schematic diagram of the structure of a computer device in an embodiment of this application;

[0024] Figure 4 This is another schematic diagram of the structure of the computer device in the embodiments of this application. Detailed Implementation

[0025] This application provides an operation specification query method, computer device, and computer storage medium. By training a query model based on knowledge graphs and entity feature vectorization, and answering the user's query based on the query model, the query results of operation specifications are output. This eliminates the need to spend a lot of time searching for operation specifications and improves the query efficiency of operation specifications.

[0026] The method for querying operational specifications in this application embodiment is described below:

[0027] Please see Figure 1 One embodiment of the operation specification query method in this application includes:

[0028] 101. Obtain operational specification data;

[0029] The method in this embodiment can be applied to computer devices, such as terminals and servers, which possess data computing and processing capabilities. The computer device can acquire operational specification data, which can be operational specifications that must be followed and adhered to in any field, such as operational specifications that must be followed and adhered to during vehicle maintenance.

[0030] This operational specification data may contain multiple data items, each containing information on at least one operation and the matters that need to be followed and based on for that operation. For example, when the operational specification data includes operational specifications for vehicle maintenance, each data item may contain information on bolt tightening operations and the matters that need to be followed and based on for that operation, such as the vehicle component on which the bolt acts, the torque parameters for tightening the bolt, and other similar information.

[0031] This operational specification data can be obtained through books, standards developed for industries or fields, standards developed for enterprises or other organizations, etc. This embodiment does not limit the method of obtaining this operational specification data.

[0032] 102. Construct a knowledge graph corresponding to the operation specification data, wherein the knowledge graph includes multiple entities in the operation specification data and the logical relationships between the multiple entities;

[0033] When operational specification data is obtained, a knowledge graph corresponding to this data can be constructed. This knowledge graph includes multiple entities from the operational specification data and the logical relationships between these entities. An entity can be any object within the operational specification data. For example, if the operational specification data pertains to vehicle repair operations, its entities could be objects such as bolts, the vehicle components they act upon, and the torque parameters for bolt tightening. The logical relationships between multiple entities represent the correspondence between them. For instance, if there is a correspondence between a bolt, the vehicle components it acts upon, and the torque parameters for bolt tightening, the logical relationship is that the bolt must be tightened based on these torque parameters.

[0034] 103. Vectorize each entity in the knowledge graph to obtain the feature vector of each entity in the knowledge graph;

[0035] Computer devices can also vectorize each entity in a knowledge graph, obtaining a feature vector for each entity. Vectorizing entities makes it easier for computers to recognize and understand their feature information, and also helps models learn the logical relationships between multiple entities in the knowledge graph based on the feature vectors.

[0036] 104. Obtain an initial query model, and train the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between multiple entities in the knowledge graph to obtain a target query model;

[0037] 105. Based on the target query model, identify the query statement input by the user and output the operation specification query result;

[0038] The computer device can also acquire an initial query model and train the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between multiple entities in the knowledge graph to obtain the target query model.

[0039] After training the target query model, it can be used to identify user-input queries. Because the target query model learns the logical relationships between multiple entities in the operational specification data during training, when a user inputs a query, the model can determine the other entities corresponding to those entities based on the learned logical relationships. For example, it can identify the bolt input by the user and the vehicle component it acts on, determine the logical relationship between the bolt and the torque parameter required for tightening, and then output the torque parameter. This allows the user to obtain the torque parameter required to tighten the bolt and tighten it accordingly.

[0040] In this embodiment, a knowledge graph corresponding to the operation specification data is constructed. Each entity in the knowledge graph is vectorized to obtain the feature vector of each entity. An initial query model is obtained, and the initial query model is trained based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between them. Based on the trained target query model, the query statement input by the user is identified, and the operation specification query result is output. Therefore, by training the query model based on the knowledge graph and entity feature vectorization, and answering the user's query for operation specifications based on the query model to output the operation specification query result, the user only needs to input any keywords to quickly obtain the corresponding operation specifications without spending a lot of time searching for operation specifications. This improves the query efficiency of operation specifications and ensures the accuracy of the query results.

[0041] The following will be discussed in the preceding text. Figure 1 Based on the illustrated embodiments, embodiments of this application will be described in further detail. Please refer to [link to relevant documentation]. Figure 2 Another embodiment of the operation specification query method in this application includes:

[0042] 201. Obtain operational specification data;

[0043] In this embodiment, the operating procedure data can be the operating procedure for vehicle maintenance, such as the operating procedure data for bolt tightening. When tightening bolts, the vehicle mechanic consults the bolt tightening procedure and tightens the bolts according to the instructions of the procedure to ensure the reliability of the connection between various vehicle components, prevent loosening or bolts from being overstretched and deformed or even broken, avoid damage to the connected components, ensure the sealing between the components, and ensure the stability of the various connection positions of the vehicle systems, so as to ensure the stable operation of the vehicle systems.

[0044] In some optional implementations, when acquiring operational specification data, bolt tightening specification data corresponding to vehicle components of various models under various vehicle brands can be collected, and bolt tightening specification data corresponding to vehicle components of various models under multiple different vehicle brands can be integrated.

[0045] Therefore, by collecting and integrating bolt tightening specification data from multiple vehicle brands and models, complete bolt tightening specification data in the vehicle repair field can be obtained. This helps the model learn complete bolt tightening specification data during training, thereby ensuring that any keywords entered by the user can yield corresponding query results and improving the user's experience in querying operating specifications.

[0046] 202. Construct a knowledge graph corresponding to the operation specification data, wherein the knowledge graph includes multiple entities in the operation specification data and the logical relationships between the multiple entities;

[0047] In some optional implementations of this embodiment, the operation specification data includes bolt tightening specification data. When constructing the knowledge graph corresponding to this operation specification data, a knowledge graph corresponding to the bolt tightening specification data can be constructed. The entities in this knowledge graph include the vehicle component on which the bolt acts, the vehicle brand, model, and their corresponding torque parameters. Furthermore, the logical relationships between multiple entities can also be obtained through the knowledge graph, such as the logical relationship between the bolt and the vehicle component on which it acts, and the torque parameters when the bolt is tightened.

[0048] Therefore, knowledge graphs can help models identify the logical relationships between multiple entities. Based on the knowledge graph, the model can quickly determine the torque parameters corresponding to the bolt and other tightening operation specifications, enabling a rapid response to user query requests and improving the efficiency of querying operation specifications.

[0049] 203. Vectorize each entity in the knowledge graph to obtain the feature vector of each entity in the knowledge graph;

[0050] After obtaining the knowledge graph corresponding to the operational specification data, the knowledge graph can be vectorized to obtain the feature vector of each entity in the knowledge graph. For example, the Trans series algorithms for knowledge graph vectorization can be used to vectorize multiple entities in the knowledge graph, such as the TransE algorithm, TransH algorithm, TransR algorithm, etc., to vectorize entities and obtain the feature vector of each entity.

[0051] 204. Obtain an initial query model, and train the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between multiple entities in the knowledge graph to obtain a target query model;

[0052] In some optional implementations of this embodiment, when training the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between the multiple entities in the knowledge graph, the feature vectors of multiple entities in the knowledge graph can be input into the initial query model so that the initial query model outputs the corresponding predicted entity according to the feature vector, and establishes a loss function based on the predicted entity and the real entities in the knowledge graph that have a logical relationship with the entity of the feature vector. The model parameters are adjusted according to the loss function until convergence is obtained to obtain the target query model.

[0053] For example, the operational specification data could be bolt tightening specifications. The knowledge graph could then consist of entities such as the various bolt components, the vehicle components they act upon, the torque parameters corresponding to bolt tightening, and the logical relationships between these entities. The feature vectors corresponding to these entities can be input into the initial query model. The initial query model identifies the feature vectors and predicts their corresponding entities, such as predicting the torque parameters corresponding to the bolts. Based on the prediction results and the torque parameters in the knowledge graph that have logical relationships with the bolts, a loss function is established. The model parameters are then adjusted based on the loss value until convergence, thus completing model training and obtaining the target query model.

[0054] Therefore, by training the query model, we can ensure that the model can quickly understand and identify each entity in the operational specification data, which helps the model to efficiently understand and identify the query statement entered by the user, respond quickly to the user's query request, and improve the feedback speed of query results.

[0055] 205. Obtain the query statement input by the user, the query statement includes at least one query entity, and perform vectorization representation on the query entity to obtain the query entity feature vector;

[0056] 206. Based on the target query model, identify the query statement input by the user and output the operation specification query result;

[0057] In this embodiment, when identifying the user-input query statement and outputting the operation specification query result based on the target query model, one optional implementation method is to obtain the user-input query statement, which includes at least one query entity, to vectorize the query entity to obtain the query entity feature vector, and to input the query entity feature vector into the target query model so that the target query model outputs the corresponding operation specification query result based on the query entity feature vector.

[0058] The operational specification data can be operational specification data for bolt tightening. Therefore, in some alternative embodiments, the query entity can include one or more of the vehicle parts affected by the bolt, vehicle brand, and vehicle model. When the target query model outputs the corresponding operational specification query results based on the query entity feature vector, the query entity feature vector can be input into the target query model. The target query model then outputs the corresponding torque parameters and / or bolt tightening precautions based on the query entity feature vector, achieving a rapid response to user query requests and feedback of query results.

[0059] For example, if a user inputs "cylinder head main bolt", the system can extract the corresponding feature vector for that keyword entity. The target query model then outputs the corresponding torque parameters and / or bolt tightening precautions based on these feature vectors. For instance, the output query result might be "The standard specification for the cylinder head main bolt is M16, with a torque range of 200–210 N·m. It requires two rotations of 90° ± 5° to achieve tightening." Thus, the user can tighten the bolts based on these torque parameters and the bolt tightening precautions.

[0060] Similarly, if a user inputs "flywheel and oil pump bolts", the corresponding feature vector can be extracted from the keyword entity. The target query model outputs the corresponding torque parameters and / or bolt tightening precautions based on the feature vector. For example, the corresponding query result might be "flywheel bolt specification M14×1.5, torque 60~80N·m, rotate two turns 90°±5°; oil pump bolt M10, torque 60~65N·m, use with anti-loosening adhesive". Thus, the user can tighten the bolts based on the torque parameters and bolt tightening precautions.

[0061] Therefore, this embodiment collects and integrates bolt tightening specifications for components from various manufacturers' vehicle models and incorporates them into a large model. This allows vehicle repair technicians to efficiently and accurately look up the bolt tightening specifications for each component, improving their repair efficiency. It enables vehicle repair technicians to quickly find the bolt tightening specifications for corresponding vehicle model components while repairing cars, thus increasing repair efficiency.

[0062] The target query model outputs the corresponding operation specification query results based on the query entity feature vector. One optional implementation involves inputting the query entity feature vector into the target query model, enabling the model to determine a target feature vector that matches the query entity feature vector, and then outputting target entities that have a logical relationship with the entity corresponding to the target feature vector based on the knowledge graph. The target entities include operation parameters.

[0063] For example, a computer device can deploy this target query model. A user can input a query command for bolt tightening specifications on the computer device's interface. This command carries multiple query entities, such as the vehicle component the bolt acts on, the vehicle brand, model, and bolt type, among other keywords. The computer device can then vectorize these keywords to obtain a query entity feature vector. This query entity feature vector can then be input into the target query model, which determines a target feature vector that matches the query entity feature vector. This target feature vector could be the feature vector of entities such as the vehicle component the bolt acts on, the vehicle brand, model, and bolt type from the operating specification data.

[0064] Furthermore, since the knowledge graph also records the logical relationships between multiple entities, the target entity that has a logical relationship with the target feature vector can be output based on the knowledge graph. For example, the torque parameter that has a logical relationship with the vehicle part that the bolt acts on, the vehicle brand, model, bolt type, etc., can be determined based on the knowledge graph, and then the torque parameter can be output so that the operator can tighten the bolt based on the torque parameter.

[0065] Therefore, by using the trained query model to identify and match the user-input query entity keywords, it is possible to quickly determine the operation parameters in the operation specification data that have a logical relationship with the user-input keyword entities, thereby enabling a rapid response to user query requests, improving the query efficiency of operation specifications, and enhancing the user's query experience of operation specifications.

[0066] The above describes the operation specification query method in the embodiments of this application. The following describes the computer equipment in the embodiments of this application. Please refer to [link / reference]. Figure 3 One embodiment of the computer device in this application includes:

[0067] The acquisition unit is used to acquire operational specification data;

[0068] A construction unit is used to construct a knowledge graph corresponding to the operation specification data, wherein the knowledge graph includes multiple entities in the operation specification data and the logical relationships between the multiple entities.

[0069] A vectorization unit is used to perform vectorization representation on each entity in the knowledge graph to obtain the feature vector of each entity in the knowledge graph;

[0070] The model training unit is used to obtain an initial query model and train the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between the multiple entities in the knowledge graph to obtain the target query model.

[0071] The query unit is used to identify the query statement input by the user based on the target query model and output the operation specification query result.

[0072] In a preferred embodiment of this invention, the query unit is specifically used for:

[0073] Obtain the query statement input by the user, the query statement including at least one query entity;

[0074] The query entity is vectorized to obtain the query entity feature vector;

[0075] The query entity feature vector is input into the target query model so that the target query model outputs the corresponding operation specification query result based on the query entity feature vector.

[0076] In a preferred embodiment of this invention, the operating specification data includes bolt tightening specification data;

[0077] Building units are specifically used for:

[0078] Construct a knowledge graph corresponding to the bolt tightening specification data. The entities in the knowledge graph include the vehicle parts to which the bolts act, the vehicle brand, the vehicle model, and their corresponding torque parameters.

[0079] In a preferred embodiment of this example, the query entity includes one or more of the following: the vehicle component acted upon by the bolt, the vehicle brand, and the vehicle model.

[0080] The query unit is specifically used for:

[0081] The query entity feature vector is input into the target query model so that the target query model outputs the corresponding torque parameters and / or bolt tightening precautions based on the query entity feature vector.

[0082] In a preferred embodiment of this invention, the acquisition unit is specifically used for:

[0083] Collect bolt tightening specifications for vehicle components of various models under various vehicle brands;

[0084] We have integrated the bolt tightening specifications for various vehicle models from multiple different vehicle brands.

[0085] In a preferred embodiment of this invention, the query unit is specifically used for:

[0086] The query entity feature vector is input into the target query model so that the target query model determines a target feature vector that matches the query entity feature vector, and outputs a target entity that has the logical relationship with the entity of the target feature vector according to the knowledge graph; wherein, the target entity includes operation parameters.

[0087] In a preferred embodiment of this invention, the model training unit is specifically used for:

[0088] The feature vectors of multiple entities in the knowledge graph are input into the initial query model, so that the initial query model outputs the corresponding predicted entity based on the feature vector. A loss function is established based on the predicted entity and the real entities in the knowledge graph that have the logical relationship with the entity of the feature vector. The model parameters are adjusted according to the loss function until convergence, and the target query model is obtained.

[0089] In this embodiment, the operations performed by each unit in the computer device are the same as described above. Figures 1 to 2 The embodiments shown are similar and will not be repeated here.

[0090] The computer device in the embodiments of this application is described below. Please refer to [link / reference]. Figure 4 One embodiment of the computer device in this application includes:

[0091] The computer device 400 may include one or more central processing units (CPUs) 401 and a memory 405, in which one or more applications or data are stored.

[0092] The memory 405 can be volatile or persistent storage. The program stored in the memory 405 can include one or more modules, each module including a series of instruction operations on the computer device. Furthermore, the central processing unit 401 can be configured to communicate with the memory 405 and execute the series of instruction operations stored in the memory 405 on the computer device 400.

[0093] Computer device 400 may also include one or more power supplies 402, one or more wired or wireless network interfaces 403, one or more input / output interfaces 404, and / or one or more operating systems, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0094] The central processing unit 401 can perform the aforementioned... Figures 1 to 2 The specific operations performed by the computer device in the illustrated embodiment will not be described in detail here.

[0095] This application also provides a computer storage medium, one embodiment of which includes: the computer storage medium storing instructions, which, when executed on a computer, cause the computer to perform the aforementioned... Figures 1 to 2 The operations performed by the computer device in the illustrated embodiment.

[0096] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0097] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0098] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0099] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0100] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the 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 cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for querying operational specifications, characterized in that, include: Obtain operational specification data; Construct a knowledge graph corresponding to the operation specification data, the knowledge graph including multiple entities in the operation specification data and the logical relationships between the multiple entities; Each entity in the knowledge graph is vectorized to obtain the feature vector of each entity in the knowledge graph; Obtain an initial query model, and train the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between the multiple entities in the knowledge graph to obtain the target query model; Based on the target query model, the system identifies the query statement entered by the user and outputs the query results with operational guidelines.

2. The method according to claim 1, characterized in that, The process of identifying the user's input query statement based on the target query model and outputting the operation specification query result includes: Obtain the query statement input by the user, the query statement including at least one query entity; The query entity is vectorized to obtain the query entity feature vector; The query entity feature vector is input into the target query model so that the target query model outputs the corresponding operation specification query result based on the query entity feature vector.

3. The method according to claim 2, characterized in that, The operational specification data includes bolt tightening specification data; Constructing the knowledge graph corresponding to the operational specification data includes: Construct a knowledge graph corresponding to the bolt tightening specification data. The entities in the knowledge graph include the vehicle parts to which the bolts act, the vehicle brand, the vehicle model, and their corresponding torque parameters.

4. The method according to claim 3, characterized in that, The query entity includes one or more of the following: the vehicle component acted upon by the bolt, the vehicle brand, and the vehicle model. The step of inputting the query entity feature vector into the target query model, so that the target query model outputs the corresponding operation specification query result based on the query entity feature vector, includes: The query entity feature vector is input into the target query model so that the target query model outputs the corresponding torque parameters and / or bolt tightening precautions based on the query entity feature vector.

5. The method according to claim 3, characterized in that, The acquisition of operation specification data includes: Collect bolt tightening specifications for vehicle components of various models under various vehicle brands; We have integrated the bolt tightening specifications for various vehicle models from multiple different vehicle brands.

6. The method according to claim 2, characterized in that, The step of inputting the query entity feature vector into the target query model, so that the target query model outputs the corresponding operation specification query result based on the query entity feature vector, includes: The query entity feature vector is input into the target query model so that the target query model determines a target feature vector that matches the query entity feature vector, and outputs a target entity that has the logical relationship with the entity of the target feature vector according to the knowledge graph; wherein, the target entity includes operation parameters.

7. The method according to any one of claims 1 to 6, characterized in that, The process of training the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between these entities to obtain the target query model includes: The feature vectors of multiple entities in the knowledge graph are input into the initial query model, so that the initial query model outputs the corresponding predicted entity based on the feature vector. A loss function is established based on the predicted entity and the real entities in the knowledge graph that have the logical relationship with the entity of the feature vector. The model parameters are adjusted according to the loss function until convergence, and the target query model is obtained.

8. A computer device, characterized in that, include: The acquisition unit is used to acquire operational specification data; A construction unit is used to construct a knowledge graph corresponding to the operation specification data, wherein the knowledge graph includes multiple entities in the operation specification data and the logical relationships between the multiple entities. A vectorization unit is used to perform vectorization representation on each entity in the knowledge graph to obtain the feature vector of each entity in the knowledge graph; The model training unit is used to obtain an initial query model and train the initial query model based on the feature vectors of multiple entities in the knowledge graph and the logical relationships between the multiple entities in the knowledge graph to obtain the target query model. The query unit is used to identify the query statement input by the user based on the target query model and output the operation specification query result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The computer storage medium stores instructions that, when executed on the computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images