System question query method, device and equipment based on natural language

Abstract

The embodiment of the invention provides a system question query method, device and equipment based on a natural language, and can be applied to the technical field of artificial intelligence. The method comprises the following steps: receiving a first query instruction for a target system; the first query instruction comprises an instruction for an operation user to describe an abnormal problem occurring in a target system based on a natural language; analyzing the first query instruction to obtain query information; encoding based on the query information to obtain a second query instruction; sending a second query instruction to the target system, so that the target system queries and feeds back a first query result according to the second query instruction; the second query instruction and the first query result correspond to a read-write format of a target system; sorting the first query result into a second query result and displaying the second query result to the operation user; the second query result is description content based on a natural language. According to the method, the threshold of checking system problems by non-professionals is reduced, and the development effect of the system is optimized.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of artificial intelligence technology, and in particular to a natural language-based system question query method, device, and equipment. Background Art

[0002] As computer technology matures, some technologies are optimized so that non-professionals can also effectively perform corresponding operations. For example, low-code platforms have reduced the difficulty and threshold of development in terms of building applications quickly and efficiently. General low-code platforms can create models by dragging and dropping, and then integrate components and use secondary development codes to complete the execution of some more complex logic, so that non-technical personnel can also effectively complete the construction of applications according to their own needs.

[0003] However, this open approach requires platform developers to encapsulate a large number of common functions into components in advance, and non-technical personnel do not understand the underlying logic during actual use. In actual use, if there is a problem with a component or the overall model, non-technical personnel cannot effectively troubleshoot the problem, that is, non-technical personnel are currently unable to find and fix problems that arise during the construction of a low-code platform by themselves. Even if the problem is fed back to the platform developer, the platform developer can only analyze the problem by querying the logs, which is not efficient. Therefore, there is an urgent need for a method that can conveniently and effectively troubleshoot system problems. Summary of the invention

[0004] The purpose of the embodiments of this specification is to provide a natural language-based system problem query method, device and equipment to solve the problem of how to conveniently and effectively troubleshoot system problems.

[0005] To solve the above technical problems, an embodiment of the present specification provides a system problem query method based on natural language, including: receiving a first query instruction for a target system; the first query instruction includes an instruction for an operating user to describe an abnormal problem occurring in the target system based on natural language; parsing the first query instruction to obtain query information; wherein, including: extracting keyword information from the first query instruction; the keyword information includes an operation type and / or a query object; encoding based on the query information to obtain a second query instruction; sending the second query instruction to the target system, so that the target system queries and feeds back a first query result according to the second query instruction; the second query instruction and the first query result correspond to the read-write format of the target system; organizing the first query result into a second query result and displaying it to the operating user; wherein, including: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

[0006] In some implementations, parsing the first query instruction to obtain query information includes: segmenting the first query instruction to obtain a segmentation array; and extracting query keywords from the segmentation array.

[0007] In some embodiments, encoding based on the query information to obtain a second query instruction includes: using a large model to encode the query information to obtain a second query instruction; arranging the first query result into a second query result and displaying it to the operating user includes: using a large model to arrange the first query result into a second query result.

[0008] Based on the above implementation, before using the big model to encode the query information to obtain the second query instruction, it also includes: marking the query keywords based on the big model specification to generate an input vector; the input vector conforms to the input format of the big model.

[0009] In some embodiments, organizing the first query result into a second query result and presenting it to the operating user includes: determining at least one question information and a corresponding information order in the first query result; organizing the question information according to a preset expression format based on the information order to obtain the second query result.

[0010] In some implementations, after arranging the first query result into a second query result and presenting it to the operating user, the method further includes: determining a preset repair suggestion that matches the first query result; and presenting the preset repair suggestion to the operating user.

[0011] In some embodiments, the target system includes a low-code platform; the low-code platform includes a platform built based on multiple existing components.

[0012] The embodiment of this specification also proposes a system problem query method based on natural language, which is applied to the platform end corresponding to the target system; the method includes: receiving a second query instruction; the second query instruction includes that after the operation end receives the first query instruction for the target system, the first query instruction is parsed to obtain query information, which includes: extracting keyword information from the first query instruction; the keyword information includes the operation type and / or query object; and then using the instruction encoded by the query information; the first query instruction includes an instruction for the operation user to describe the abnormal problem occurring in the target system based on natural language; querying based on the query information in the second query instruction to obtain a first query result; the second query instruction and the first query result correspond to the read-write format of the target system; sending the first query result to the operation end, so that the operation end organizes the first query result into a second query result and displays it to the operation user; which includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

[0013] In some implementations, querying based on the query information in the second query instruction to obtain the first query result includes: determining a target interface based on the query information; extracting an associated log of the target interface; and extracting the first query result from the associated log.

[0014] Based on the above implementation, the log in the target system has a log identifier based on the interface link; determining the target interface based on the query information includes: determining the interface link corresponding to the abnormal problem; finding at least one target interface corresponding to the interface link; extracting the associated log of the target interface includes: determining the target log identifier corresponding to the at least one target interface; and extracting the associated log using the target log identifier.

[0015] The embodiment of this specification also proposes a system problem query device based on natural language, which is arranged at the operation end; the device includes: a first query instruction receiving module, which is used to receive a first query instruction for a target system; the query instruction includes an instruction for an operating user to describe an abnormal problem occurring in the target system based on natural language; a first query instruction parsing module, which is used to parse the first query instruction to obtain query information; wherein, it includes: extracting keyword information from the first query instruction; the keyword information includes an operation type and / or a query object; an encoding module, which is used to encode based on the query information to obtain a second query instruction; a second query instruction sending module, which is used to send the second query instruction to the target system, so that the target system queries and feeds back a first query result according to the second query instruction; the second query instruction and the first query result correspond to the read-write format of the target system; a second query result display module, which is used to organize the first query result into a second query result and display it to the operating user; wherein, it includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

[0016] The embodiment of this specification also proposes a system problem query device based on natural language, which is set at the platform end corresponding to the target system; the device includes: a second query instruction receiving module, which is used to receive a second query instruction; the second query instruction includes that after the operation end receives the first query instruction for the target system, the first query instruction is parsed to obtain query information, which includes: extracting keyword information from the first query instruction; the keyword information includes the operation type and / or query object; and then using the query information to encode the instruction; the first query instruction includes an instruction for the operation user to describe the abnormal problem occurring in the target system based on natural language; a first query result query module, which is used to query based on the query information in the second query instruction to obtain a first query result; the second query instruction and the first query result correspond to the read-write format of the target system; a first query result sending module, which is used to send the first query result to the operation end, so that the operation end organizes the first query result into a second query result and displays it to the operation user; which includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

[0017] The embodiment of this specification also proposes an electronic device, including a memory and a processor; the memory is used to store computer programs / instructions; the processor is used to execute the computer programs / instructions to implement the steps of the above-mentioned natural language-based system question query method.

[0018] The embodiments of this specification also provide a computer-readable storage medium on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the steps of the above-mentioned natural language-based system question query method are implemented.

[0019] The embodiments of this specification also provide a computer program product, including a computer program / instruction, which implements the steps of the above-mentioned natural language-based system question query method when executed by a processor.

[0020] It can be seen from the technical solutions provided in the above embodiments of this specification that in the method of the embodiments of this specification, the user can directly input a first query instruction that describes the problem in natural language, then parse the first query instruction to obtain query information, and encode the query information to obtain a second query instruction, and send the second query instruction to the target system so that the target system can find the corresponding first query result according to the second query instruction, and then the operating end organizes the first query result into a second query result in natural language form and displays it to the operating user. Through the above method, not only is effective query of system problems achieved, but also the instructions entered by the user and the results directly viewed during the query process are in the form of natural language expressions, thereby greatly reducing the threshold for non-professionals to troubleshoot system problems, optimizing the development effect of the system, and ensuring the final application effect of the system, and having good practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 This is a flowchart of a natural language-based system question query method according to an embodiment of this specification;

[0023] Figure 2 This is a flowchart of a natural language-based system question query method according to an embodiment of this specification;

[0024] Figure 3 This is a flowchart of a natural language-based system question query method according to an embodiment of this specification;

[0025] Figure 4 This is a module diagram of a natural language-based system problem query device according to an embodiment of this specification;

[0026] Figure 5 This is a module diagram of a natural language-based system problem query device according to an embodiment of this specification. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0028] In order to solve the above technical problems, the embodiment of this specification proposes a system problem query method based on natural language. The execution subject of the system problem query method based on natural language is a corresponding system problem query system, which mainly involves an operation end and a platform end corresponding to a target system.

[0029] The operation end is the device end that receives user instructions for processing, and can also convert the instructions sent by the platform end into natural language for display to the user for viewing. The operation end can be a terminal that directly interacts with the user, or a device end that can communicate with the user terminal, such as a cloud server or host device.

[0030] The platform end may be the device end where the target system is located. The target system may be a system under development or a system in application. The target system has specific application functions, but the target system may have some problems in actual application, so it is necessary to troubleshoot and repair the problem for the target system. In order to enable non-professionals to also troubleshoot and repair the problem for the target system, the solution in the embodiment of this specification proposes corresponding problem-solving ideas.

[0031] In some embodiments, the target system includes a low-code platform. The low-code platform includes a platform built based on multiple existing components. The low-code platform can provide modeling, process, page creation, second opening and other functions.

[0032] It should be noted that in actual applications, the operation end and the platform end can be the same device. In the embodiments of this specification, for the convenience of description, the two are distinguished as different execution entities for description, but it is not limited to the two being the same device. If the operation end and the platform end are the same device in actual applications, the interaction between the operation end and the platform end in the following process can be understood as the interactive relationship between the device content.

[0033] like Figure 1 As shown, the natural language-based system question query method may include the following specific implementation steps.

[0034] S110: The operation end receives a first query instruction for the target system.

[0035] The first query instruction is an instruction directly input by the user. In the embodiment of the present specification, for the above-mentioned technical problem, the first query instruction can be in the form of natural language to describe the corresponding problem, that is, the instruction type used by the user for daily expression. For example, it can be a problem described in natural language such as "XX function cannot be applied" or "Type A error message appears after execution". These problems cannot be directly analyzed by the system to obtain specific meanings, and corresponding conversions are required to complete the problem query process.

[0036] In the application scenario of the embodiments of this specification, the first query instruction is used to query abnormal problems that occur in the target system. When developing and designing the target system or debugging the target system, the developers may not have strong professional knowledge, that is, they cannot directly write code or instructions to query the problems. Therefore, by entering natural language to express their own needs, the threshold for developing and debugging the target system is lowered.

[0037] The first query instruction is input in the form of natural language, which lowers the threshold for user operation and enables system problems to be identified and query operations to be completed without mastering professional techniques or underlying system logic.

[0038] Based on the system architecture in actual applications, the first query instruction can be an instruction directly input into the operation terminal. For example, when the operation terminal has a corresponding operation interface, the first query instruction input by the user can be directly received and forwarded to the corresponding processing module for processing; the first query instruction can also be input into the user terminal by the user and forwarded by the user terminal, that is, the operation terminal can be a cloud device or an independent computing device, and the user inputs the instruction through the corresponding terminal, and then forwarded to the operation terminal for processing. In actual applications, it can be set according to specific circumstances, and there is no restriction on this.

[0039] S120: The operating end parses the first query instruction to obtain query information.

[0040] The query information is used to reflect the specific query content. Since the first query instruction is an instruction described in natural language and may contain meaningless words, the corresponding electronic device cannot directly analyze and utilize the first query instruction, so the first query instruction can be cleaned and analyzed to obtain the query information.

[0041] In some embodiments, the query information includes a query object and an operation type. The query object is the object targeted by the first query instruction, and specifically may be the object referred to by the user's question. The query object may be directly reflected in the first query instruction, or may be found by searching for a specific question. The operation type may be used to reflect the type of question, thereby facilitating the corresponding search for execution strategies and other contents according to the type of question.

[0042] In some implementations, in order to ensure the parsing effect of the first query instruction, the first query instruction may be segmented to obtain a segmentation array, and then the query keyword may be extracted from the segmentation array.

[0043] The word segmentation process can be executed based on the corresponding word segmentation operation process, for example, it can be operated according to preset word segmentation rules, or directly processed using the corresponding neural network model. The specific word segmentation process can be set according to the actual application situation, and there is no restriction on this.

[0044] After word segmentation, meaningless words can be removed, and query keywords can be extracted through semantic analysis and other operations. Query keywords are keywords that reflect actual meanings, such as specific issues or problem objects involved. In actual applications, standards and methods for filtering query keywords can be set according to needs, which will not be repeated here.

[0045] S130: The operating end encodes the query information to obtain a second query instruction.

[0046] The second query instruction is a format for the platform or target system to read and write information. By converting the query information into the second query instruction, the second query instruction can be effectively read by the platform, thereby realizing the corresponding problem finding process.

[0047] The encoding process of the query information may be to convert the extracted query information into an encoding format according to a preset conversion rule. For example, a corresponding format is preset, and the corresponding second query instruction can be obtained by filling in the keywords in the query information.

[0048] In some embodiments, the query information can be encoded using a large model to obtain a second query instruction. A large model refers to a machine learning model with large-scale parameters and complex computing structures. These models are usually built from deep neural networks and have billions or even hundreds of billions of parameters. The purpose of designing a large model is to improve the expressiveness and predictive performance of the model and to be able to handle more complex tasks and data. Large models are widely used in various fields, including natural language processing, computer vision, speech recognition, and recommendation systems. Large models learn complex patterns and features by training massive amounts of data, have more powerful generalization capabilities, and can make accurate predictions on unseen data.

[0049] Specifically, corresponding sample data can be constructed in advance, and the large model can be trained based on these sample data, so that the trained large model can determine the corresponding encoding method according to natural language or keyword information, and then encode it into a format that can be directly read by the platform end, and complete the entry of corresponding instructions without the need for additional development of the platform end.

[0050] Preferably, when a large model is applied, the query keywords may be marked in advance based on the large model specification to generate an input vector. The input vector conforms to the input format of the large model. Since the large model may be a separately developed model, in order to ensure that the large model can be effectively applied without additional development, the above processing operation may be performed to ensure that the large model can effectively analyze the query information.

[0051] The specific model structure and training process can be set according to the actual application situation and will not be described in detail here.

[0052] S140: The operation end sends the second query instruction to the platform end.

[0053] After generating the second query instruction, the operation end may send the second query instruction to the platform end.

[0054] S150: The platform performs a query based on the query information in the second query instruction to obtain a first query result.

[0055] Since the second query instruction is an instruction that can be directly read by the platform side and already contains corresponding query information, a query can be performed based on the query information in the second query instruction to obtain the corresponding first query result.

[0056] For example, based on the format of the second query instruction, the platform can effectively read the operation content and the target object of the second query instruction, and then target the corresponding object.

[0057] In some implementations, a target interface may be determined based on the query information, and then a log associated with the target interface may be extracted, and the first query result may be extracted from the log associated with the target interface.

[0058] Since the query information can reflect the specific query content, such as the query object or problem type, the specific interface or component where the problem occurs can be located based on the query information. For different interfaces, corresponding execution logs are recorded, and the specific content of the problem can be extracted by analyzing the logs.

[0059] Specifically, when searching for logs, you can first determine the interface link corresponding to the abnormal problem, and then search for at least one target interface corresponding to the interface link. Since the target system is composed of multiple components, the communication relationship and interaction relationship between different components are fixed, and the interface link can reflect the interface communication relationship corresponding to the component relationship in the system. Therefore, you can determine the corresponding interface link based on the abnormal problem, and then determine the interface contained in the link, so as to conveniently and effectively extract and parse the context log.

[0060] Correspondingly, the log is uniquely identified according to the interface link, and each log information carries the application code, application version, and detailed input and return information. Therefore, the target log identifier corresponding to the at least one target interface can be determined to complete the corresponding search of the log.

[0061] After obtaining the query information, operations such as cleaning and keyword extraction can be performed on the query information, which makes the processing flow in the subsequent process more convenient.

[0062] S160: The platform sends the first query result to the operation end.

[0063] After generating the first query result, the platform end may send the first query result to the operation end.

[0064] S170: The operation end organizes the first query result into a second query result and displays it to the operation user.

[0065] After obtaining the first query result, since the first query result is directly fed back by the platform, the content format is generally more professional, and the operating user may not be able to directly understand it if he is not a professional technician or developer. Therefore, the first query result can be organized into a second query result. The second query result is a description in natural language, so that the operating user can directly understand the meaning of the second query result.

[0066] The second query result may be constructed based on the query result keywords after extracting the query result keywords from the first query result. The query result keywords may be key information in the first query result. Since the first query result is content generated by the target system, it contains some format descriptions that are not convenient for users to understand directly, so these contents can be removed to complete the extraction of query result keywords.

[0067] The specific extraction process can be performed by pre-setting corresponding rules to directly extract the keywords, for example, by identifying them through a neural network model. The format of the first query result can also be predetermined, and then the first query result is analyzed based on this format to determine the query result keywords. The specific acquisition process can be set according to the needs of the actual application, and will not be repeated here.

[0068] In some implementations, at least one piece of question information and a corresponding information sequence in the first query result may be determined first, and then the question information may be organized according to a preset expression format based on the information sequence to obtain a second query result.

[0069] The order of information is used to reflect the logic of the question, making it easier to integrate the display logic of the question based on the order of information in the subsequent process.

[0070] The preset expression format may be a preset natural language format content, wherein corresponding filling positions are set for specific questions. By analyzing and extracting the question information, corresponding keywords can be obtained, and then the keywords are added to the filling positions to complete the construction of the second query result.

[0071] Preferably, a conversion format may be pre-set to convert the question information into corresponding natural language description content. For example, natural language description content may be pre-annotated for different questions, and the corresponding natural language description content may be displayed simultaneously with the question information, so that the operating user can better understand the specific question based on the natural language description content.

[0072] Correspondingly, the above-mentioned sorting process can also be implemented by a large model. For example, the large model can be trained in advance so that the large model also has the function of converting instruction-type content into natural language-type content. The specific training process and application method can be set based on the actual application situation and will not be repeated here.

[0073] In some implementations, in order to further ensure that the operating user handles the anomaly in the system, a preset repair suggestion matching the first query result may be determined, and the preset repair suggestion may be displayed to the operating user.

[0074] Specifically, developers can pre-set corresponding repair suggestions in the system, and the repair suggestions are matched with specific problems. For example, when there is a problem type corresponding to a system abnormality, the repair suggestion can correspond to the problem type. When it is necessary to determine the repair suggestion, the corresponding repair suggestion can be directly matched according to the problem type reported for the abnormal problem in the first query result, and displayed to the operating user. Preferably, the repair suggestion can have corresponding operation instructions to facilitate the operating user to complete the repair of the target system based on the repair suggestion.

[0075] The specific content of the repair suggestion and the specific matching process can be set according to the actual application situation, and will not be repeated here.

[0076] Based on the above process, when developing for the target system, since the target system receives and outputs instructions or data corresponding to its own read and write format, there is no need for additional configuration of the target system. Developers can apply the above method to different target systems, ensuring that non-professionals can effectively apply it when developing for low-code platforms, which is more in line with the application scenarios corresponding to the embodiments of this specification.

[0077] Based on the introduction of the above embodiments and specific examples, in the above natural language-based system problem query method, the user can directly input a first query instruction that describes the problem in natural language, then parse the first query instruction to obtain query information, and encode the query information to obtain a second query instruction, and send the second query instruction to the target system so that the target system can find the corresponding first query result according to the second query instruction, and then the operation end organizes the first query result into a second query result in natural language form and displays it to the operating user. Through the above method, not only is the effective query of system problems achieved, but also the instructions entered by the user and the results directly viewed during the query process are in the form of natural language expressions, which greatly reduces the threshold for non-professionals to troubleshoot system problems, optimizes the system development effect, and ensures the final application effect of the system, and has good practical application value.

[0078] The following further introduces the system problem query method with the operation end and the platform end as the execution subjects respectively.

[0079] based on Figure 1 Corresponding to the natural language-based system problem query method, the embodiment of this specification also proposes a natural language-based system problem query method. The execution subject of the natural language-based system problem query method is the operation end. Figure 2 As shown, the method includes the following specific implementation steps.

[0080] S210: Receive a first query instruction for a target system; the first query instruction includes an instruction for an operating user to describe an abnormal problem occurring in the target system based on natural language.

[0081] For the specific description of this step, please refer to the introduction in step S110, which will not be repeated here.

[0082] S220: Parse the first query instruction to obtain query information; which includes: extracting keyword information from the first query instruction; the keyword information includes an operation type and / or a query object.

[0083] For the specific description of this step, please refer to the introduction in step S120, which will not be repeated here.

[0084] S230: Encode based on the query information to obtain a second query instruction.

[0085] For the specific description of this step, please refer to the introduction in step S130, which will not be repeated here.

[0086] S240: Sending the second query instruction to the target system, so that the target system queries and feeds back the first query result according to the second query instruction; the second query instruction and the first query result correspond to the read-write format of the target system.

[0087] For the specific description of this step, please refer to the introduction in steps S140, S150, and S160, which will not be repeated here.

[0088] S250: Arrange the first query result into a second query result and display it to the operating user; which includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

[0089] For the specific description of this step, please refer to the introduction in step S170, which will not be repeated here.

[0090] based on Figure 1 The corresponding natural language-based system problem query method, this specification embodiment also proposes a natural language-based system problem query method. The execution subject of the natural language-based system problem query method is the platform end. Figure 3 As shown, the method includes the following specific implementation steps.

[0091] S310: Receive a second query instruction; the second query instruction includes that after the operation end receives the first query instruction for the target system, the first query instruction is parsed to obtain query information, which includes: extracting keyword information from the first query instruction; the keyword information includes the operation type and / or query object; and then using the query information to encode the instruction; the first query instruction includes the instruction for the operation user to describe the abnormal problem occurring in the target system based on natural language.

[0092] For the specific description of this step, please refer to the introduction in steps S110, S120, S130, and S140, which will not be repeated here.

[0093] S320: Perform a query based on the query information in the second query instruction to obtain a first query result; the second query instruction and the first query result correspond to a read-write format of the target system.

[0094] For the specific description of this step, please refer to the introduction in step S150, which will not be repeated here.

[0095] S330: Send the first query result to the operation end, so that the operation end organizes the first query result into a second query result and displays it to the operation user; which includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

[0096] For the specific description of this step, please refer to the introduction in steps S160 and S170, which will not be repeated here.

[0097] based on Figure 2 The corresponding natural language-based system problem query method introduces a natural language-based system problem query device in the embodiment of this specification. The natural language-based system problem query device can be set on the operating end. Figure 4 As shown, the natural language-based system problem query device includes the following modules.

[0098] The first query instruction receiving module 410 is used to receive a first query instruction for a target system; the query instruction includes an instruction for an operating user to describe an abnormal problem occurring in the target system based on natural language.

[0099] The first query instruction parsing module 420 is used to parse the first query instruction to obtain query information; which includes: extracting keyword information from the first query instruction; the keyword information includes an operation type and / or a query object.

[0100] The encoding module 430 is used to perform encoding based on the query information to obtain a second query instruction.

[0101] The second query instruction sending module 440 is used to send the second query instruction to the target system, so that the target system queries and feeds back the first query result according to the second query instruction; the second query instruction and the first query result correspond to the read-write format of the target system.

[0102] The second query result display module 450 is used to organize the first query result into a second query result and display it to the operating user; which includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

[0103] based on Figure 3The corresponding natural language-based system problem query method introduces a natural language-based system problem query device in the embodiment of this specification. The natural language-based system problem query device can be set on the operating end. Figure 5 As shown, the natural language-based system problem query device includes the following modules.

[0104] The second query instruction receiving module 510 is used to receive a second query instruction; the second query instruction includes that after the operation end receives the first query instruction for the target system, the first query instruction is parsed to obtain query information, which includes: extracting keyword information from the first query instruction; the keyword information includes the operation type and / or query object; and then using the query information to encode the instruction; the first query instruction includes the instruction for the operation user to describe the abnormal problem that occurs in the target system based on natural language.

[0105] The first query result query module 520 is used to query based on the query information in the second query instruction to obtain a first query result; the second query instruction and the first query result correspond to the read-write format of the target system.

[0106] The first query result sending module 530 is used to send the first query result to the operation end, so that the operation end organizes the first query result into a second query result and displays it to the operation user; which includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

[0107] based on Figure 1 Corresponding to the natural language-based system problem query method, the embodiment of this specification provides an electronic device. The electronic device may include a memory and a processor.

[0108] In this embodiment, the memory may be implemented in any appropriate manner. For example, the memory may be a read-only memory, a mechanical hard disk, a solid-state hard disk, or a USB flash drive, etc. The memory may be used to store computer programs / instructions.

[0109] In this embodiment, the processor may be implemented in any suitable manner. For example, the processor may take the form of a microprocessor or a processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) executable by the (micro)processor, a logic gate, a switch, an application specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, etc. The processor may execute the computer program instructions to implement the following. Figure 2 or Figure 3 The corresponding natural language-based system question query method.

[0110] The embodiments of this specification provide a computer-readable storage medium on which a computer program / instruction is stored. The computer-readable storage medium can be read by a processor based on an internal bus of a device, and the program instructions in the computer-readable storage medium are then implemented by the processor.

[0111] In this embodiment, the computer-readable storage medium can be implemented in any appropriate manner. The computer-readable storage medium includes, but is not limited to, random access memory (RAM), read-only memory (ROM), cache, hard disk drive (HDD), memory card, etc. The computer storage medium stores computer program instructions. When the computer program instructions are executed, the present specification is implemented. Figure 2 or Figure 3 The program instructions or modules of the corresponding embodiment.

[0112] The present specification also provides a computer program product, including a computer program / instruction. The computer program product may be a program written in a corresponding computer program language, stored in a corresponding storage device in a program manner, and may be transmitted via a computer network. The computer program product may be executed by a processor. In the present specification, the computer program product may be executed to implement the following: Figure 2 or Figure 3 The corresponding embodiment is a program instruction or module of a natural language-based system question query method.

[0113] It should be noted that the above-mentioned natural language-based system question query method, device and equipment can be applied to the field of artificial intelligence technology, and can also be applied to other technical fields except the field of artificial intelligence technology, without any limitation.

[0114] In addition, all data acquisition, processing, storage, forwarding and other operations in the above-mentioned natural language-based system problem query method, device and equipment comply with the relevant provisions of relevant laws and regulations.

[0115] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0116] Although the process flows described above include multiple operations that occur in a particular order, it should be clearly understood that the processes may include more or fewer operations, which may be performed sequentially or in parallel (eg, using parallel processors or a multi-threaded environment).

[0117] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0118] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0119] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0120] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0121] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0122] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, tape storage, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0123] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, the embodiments of this specification may take the form of complete hardware embodiments, complete software embodiments or embodiments combining software and hardware. Moreover, the embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0124] The present specification embodiments may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present specification embodiments may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0125] Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. In the description of this specification, the description of the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of this specification. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, in the absence of contradiction, a person skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0126] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A system problem query method based on natural language, characterized in that: Applied to the operating end; the method comprises: Receiving a first query instruction for a target system; the first query instruction includes an instruction for an operating user to describe an abnormal problem occurring in the target system based on natural language; Parsing the first query instruction to obtain query information; wherein, the method includes: extracting keyword information from the first query instruction; the keyword information includes an operation type and / or a query object; Encoding based on the query information to obtain a second query instruction; Sending the second query instruction to the target system, so that the target system queries and feeds back the first query result according to the second query instruction; the second query instruction and the first query result correspond to the read-write format of the target system; The first query result is organized into a second query result and displayed to the operating user; wherein, it includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

2. The method according to claim 1, characterized in that The step of parsing the first query instruction to obtain query information includes: Segmenting the first query instruction to obtain a segmentation array; Extract query keywords from the word segmentation array.

3. The method according to claim 2, characterized in that The step of encoding based on the query information to obtain a second query instruction includes: Encoding the query information using the large model to obtain a second query instruction; The arranging the first query result into a second query result and displaying the second query result to the operating user includes: The first query result is organized into a second query result using a large model.

4. The method according to claim 3, characterized in that Before encoding the query information by using the large model to obtain the second query instruction, the method further includes: The query keywords are marked based on the big model specification to generate an input vector; the input vector conforms to the input format of the big model.

5. The method according to claim 1, characterized in that The arranging the first query result into a second query result and displaying the second query result to the operating user includes: Determine at least one question information and a corresponding information sequence in the first query result; The question information is organized according to a preset expression format based on the information sequence to obtain a second query result.

6. The method according to claim 1, characterized in that After arranging the first query result into a second query result and displaying it to the operating user, the method further includes: Determining a preset repair suggestion that matches the first query result; The preset repair suggestion is presented to the operating user.

7. The method according to claim 1, characterized in that The target system includes a low-code platform; the low-code platform includes a platform built based on multiple existing components.

8. A system problem query method based on natural language, characterized in that: Applied to the platform end corresponding to the target system; the method includes: receiving a second query instruction; the second query instruction includes that after the operation end receives the first query instruction for the target system, the first query instruction is parsed to obtain query information, which includes: extracting keyword information from the first query instruction; the keyword information includes the operation type and / or the query object; and then encoding the instruction using the query information; the first query instruction includes an instruction for the operation user to describe the abnormal problem occurring in the target system based on natural language; Performing a query based on the query information in the second query instruction to obtain a first query result; the second query instruction and the first query result correspond to a read-write format of the target system; The first query result is sent to the operation end, so that the operation end organizes the first query result into a second query result and displays it to the operation user; wherein, it includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

9. The method according to claim 8, characterized in that The querying based on the query information in the second query instruction to obtain the first query result includes: Determine a target interface based on the query information; Extracting the associated logs of the target interface; A first query result is extracted from the associated log.

10. The method according to claim 9, characterized in that The log in the target system has a log identifier based on the interface link; and determining the target interface based on the query information includes: Determine the interface link corresponding to the abnormal problem; Find at least one target interface corresponding to the interface link; The extracting the associated log of the target interface includes: Determine a target log identifier corresponding to the at least one target interface; The target log identifier is used to extract the associated log.

11. A system problem query device based on natural language, characterized in that: Set at the operating end; the device includes: A first query instruction receiving module is used to receive a first query instruction for a target system; the query instruction includes an instruction for an operating user to describe an abnormal problem occurring in the target system based on natural language; A first query instruction parsing module is used to parse the first query instruction to obtain query information; wherein, it includes: extracting keyword information from the first query instruction; the keyword information includes an operation type and / or a query object; An encoding module, used for encoding based on the query information to obtain a second query instruction; A second query instruction sending module is used to send the second query instruction to the target system, so that the target system queries and feeds back the first query result according to the second query instruction; the second query instruction and the first query result correspond to the read-write format of the target system; The second query result display module is used to organize the first query result into a second query result and display it to the operating user; which includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

12. A system problem query device based on natural language, characterized in that: The device is arranged on a platform corresponding to the target system; the device comprises: A second query instruction receiving module is used to receive a second query instruction; the second query instruction includes that after the operation end receives the first query instruction for the target system, the first query instruction is parsed to obtain query information, which includes: extracting keyword information from the first query instruction; the keyword information includes an operation type and / or a query object; and then using the query information to encode the instruction; the first query instruction includes an instruction for the operation user to describe the abnormal problem occurring in the target system based on natural language; A first query result query module, used to query based on the query information in the second query instruction to obtain a first query result; the second query instruction and the first query result correspond to the read-write format of the target system; The first query result sending module is used to send the first query result to the operation end, so that the operation end organizes the first query result into a second query result and displays it to the operation user; wherein, it includes: extracting query result keywords from the first query result, and constructing the second query result based on the query result keywords; the second query result is a description content based on natural language.

13. An electronic device comprising a memory and a processor; characterized in that: The memory is used to store computer programs / instructions; the processor is used to execute the computer programs / instructions to implement the steps of the method according to any one of claims 1 to 10.

14. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: The computer program / instructions, when executed by a processor, implement the steps of the method according to any one of claims 1 to 10.

15. A computer program product comprising a computer program / instructions, characterized in that The computer program / instructions, when executed by a processor, implement the steps of the method according to any one of claims 1 to 10.

Images