Operation and maintenance question and answer rule generation method and device, low-code platform and medium

Abstract

The invention relates to an operation and maintenance question and answer rule generation method and device, a low-code platform and a medium. In response to a dragging operation on the basic function component of the component providing region, displaying a connection relationship between the basic function component and different basic function components in the rule generation region, the connection relationship between the different basic function components representing a logic relationship between the different basic function components; in response to a triggering operation on a basic function component of the rule generation region, displaying a to-be-configured item corresponding to the basic function component in the component attribute configuration region; based on the configuration operation of the to-be-configured item corresponding to the basic function component, obtaining component configuration information of the basic function component; and according to each basic function component, the logic relationship between different basic function components and the component configuration information of each basic function component, generating an operation and maintenance question and answer rule. In this way, the operation and maintenance logs can be automatically checked, and the answers matched with the after-sales operation and maintenance scenes can be searched.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent question-answering technology, and in particular to a method, device, low-code platform and medium for generating operation and maintenance question-answering rules. Background Art

[0002] In the current after-sales operation and maintenance scenarios, the operation and maintenance logs are manually checked, and the answers to the operation and maintenance questions are searched from the manually checked operation and maintenance logs in a question-and-answer manner. However, manually checking the operation and maintenance logs takes a lot of time and is inefficient. In addition, if the large language model is directly applied to the after-sales operation and maintenance scenarios to search for answers to operation and maintenance questions, the results will be generalized and cannot be adapted to personalized or customized after-sales operation and maintenance scenarios.

[0003] Therefore, in the face of different after-sales operation and maintenance scenarios, it is urgent to provide an appropriate operation and maintenance question and answer rule generation method to realize automatic troubleshooting of operation and maintenance logs and search for answers that are suitable for after-sales operation and maintenance scenarios. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a method, device, low-code platform and medium for generating operation and maintenance question and answer rules.

[0005] In a first aspect, the present disclosure provides a method for generating operation and maintenance question and answer rules, which is applied to a low-code platform, wherein the configuration interface of the low-code platform includes a component provision area, a rule generation area, and a component attribute configuration area, and the method includes:

[0006] In response to a drag operation on a basic functional component in the component providing area, displaying the basic functional component and connection relationships between different basic functional components in the rule generating area, wherein the connection relationships between different basic functional components represent logical relationships between different basic functional components;

[0007] In response to a triggering operation on a basic functional component in the rule generation area, displaying a to-be-configured item corresponding to the basic functional component in the component attribute configuration area;

[0008] Based on the configuration operation of the to-be-configured item corresponding to the basic functional component, obtaining component configuration information of the basic functional component;

[0009] An operation and maintenance question and answer rule is generated according to each of the basic functional components, the logical relationship between different basic functional components, and the component configuration information of each basic functional component.

[0010] In a second aspect, the present disclosure provides an operation and maintenance question and answer rule generation device, which is configured on a low-code platform, wherein the configuration interface of the low-code platform includes a component provision area, a rule generation area, and a component attribute configuration area, and the device includes:

[0011] A component display module, configured to, in response to a drag operation on a basic function component in the component providing area, display the basic function component and a connection relationship between different basic function components in the rule generating area, wherein the connection relationship between different basic function components represents a logical relationship between different basic function components;

[0012] A component triggering module, configured to respond to a triggering operation on a basic functional component in the rule generation area and display a to-be-configured item corresponding to the basic functional component in the component attribute configuration area;

[0013] A configuration information acquisition module, used to acquire component configuration information of the basic functional component based on a configuration operation on a to-be-configured item corresponding to the basic functional component;

[0014] The rule generation module is used to generate operation and maintenance question and answer rules according to each of the basic functional components, the logical relationship between different basic functional components, and the component configuration information of each basic functional component.

[0015] In a third aspect, the present disclosure also provides a low-code platform, including:

[0016] processor;

[0017] A memory for storing executable instructions;

[0018] The processor is used to read the executable instructions from the memory and execute the executable instructions to implement the method provided in the first aspect.

[0019] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method provided in the first aspect is implemented.

[0020] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0021] The present invention discloses an operation and maintenance question and answer rule generation method, device, low-code platform and medium. The configuration interface of the low-code platform includes a component providing area, a rule generating area and a component property configuration area. The method includes: in response to a drag operation on a basic function component in the component providing area, displaying the connection relationship between the basic function component and different basic function components in the rule generating area, wherein the connection relationship between different basic function components represents the logical relationship between different basic function components; in response to a trigger operation on the basic function component in the rule generating area, displaying the to-be-configured items corresponding to the basic function component in the component property configuration area; based on the configuration operation on the to-be-configured items corresponding to the basic function component, obtaining the component configuration information of the basic function component; generating an operation and maintenance question and answer rule according to each basic function component, the logical relationship between different basic function components and the component configuration information of each basic function component. In this way, facing different after-sales operation and maintenance scenarios, the component configuration information of different basic functional components and the logical relationship between different basic functional components are determined on the low-code platform, so as to obtain operation and maintenance question and answer rules suitable for the after-sales operation and maintenance scenarios in which they are located. Based on the operation and maintenance question and answer rules, it is possible to automatically check the operation and maintenance logs and search for answers suitable for the after-sales operation and maintenance scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0023] In order to more clearly illustrate the embodiments of the present invention 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, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0024] Figure 1 A flowchart of a method for generating operation and maintenance question and answer rules provided in an embodiment of the present disclosure;

[0025] Figure 2 A schematic diagram of a configuration interface provided by an embodiment of the present disclosure;

[0026] Figure 3 A schematic diagram of the structure of an operation and maintenance question and answer rule generation device provided in an embodiment of the present disclosure;

[0027] Figure 4 A schematic diagram of the structure of a low-code platform provided for an embodiment of the present disclosure. DETAILED DESCRIPTION

[0028] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0029] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0030] In order to automatically check the operation and maintenance logs and search for answers that are suitable for after-sales operation and maintenance scenarios, the following is combined Figure 1 to Figure 2 The operation and maintenance question and answer rule generation method provided in the embodiment of the present disclosure is described. In the embodiment of the present disclosure, the operation and maintenance question and answer rule generation method can be executed by a low-code platform. The low-code platform is a software development tool that allows users to quickly create applications through a graphical interface and drag-and-drop functions without in-depth code knowledge. Among them, the configuration interface of the low-code platform includes a component provision area, a rule generation area, and a component attribute configuration area. The component provision area is used to provide basic functional components for generating operation and maintenance question and answer rules; the rule generation area is used to generate operation and maintenance question and answer rules based on the basic functional components; the component attribute configuration area is used to configure attribute information for the basic functional components.

[0031] Figure 1 A flow chart of a method for generating operation and maintenance question and answer rules provided in an embodiment of the present disclosure is shown.

[0032] like Figure 1 As shown, the operation and maintenance question and answer rule generation method may include the following steps.

[0033] S110. In response to a drag operation on a basic functional component in a component providing area, a basic functional component and a connection relationship between different basic functional components are displayed in a rule generating area, wherein the connection relationship between different basic functional components represents a logical relationship between different basic functional components.

[0034] In this embodiment, since the component providing area provides basic functional components for generating operation and maintenance question and answer rules, the user can drag the basic functional components from the component providing area to the rule generation area according to the project requirements of the after-sales operation and maintenance scenario. The low-code platform responds to the dragging operation, displays the dragged basic functional components in the rule generation area, and displays the connection relationship between different basic functional components.

[0035] Since the connection relationship between different basic functional components represents the logical relationship between different basic functional components, the low-code platform can determine the code execution logic of the operation and maintenance question and answer rules based on the connection relationship.

[0036] For easier understanding, see Figure 2 The interface diagram of the configuration interface shown in the figure is as follows: Figure 2 The component providing area 210 of the configuration interface shown includes basic functional components such as operation and maintenance problem input component, log component and knowledge base component. Specifically, the user's first dragging operation is to move the operation and maintenance problem input component of the component providing area 210 to the rule generating area 220, the user's second dragging operation is to move the log component to the rule generating area 220, and the user's third dragging operation is to move the knowledge base component of the component providing area 210 to the rule generating area 220. Then the rule generating area 220 displays the operation and maintenance problem input component, log component and knowledge base component in the order of dragging, and displays the connection relationship between the three components, and determines the logical relationship between the three components based on the connection relationship between the three components.

[0037] S120 . In response to a triggering operation on a basic functional component in a rule generation area, displaying to-be-configured items corresponding to the basic functional component in a component property configuration area.

[0038] In this embodiment, in order to generate more functional operation and maintenance question-and-answer rules, the low-code platform also allows users to configure the attribute information of basic functional components. Specifically, after the user moves one or all basic functional components to the rule generation area, clicks on the basic functional component, and the attribute configuration area of ​​the low-code platform displays the to-be-configured items corresponding to the clicked basic functional component, allowing the user to configure the attribute information for the basic functional component based on the to-be-configured items.

[0039] The item to be configured refers to a configuration window for configuring attribute information of a basic functional component, which is used to remind or guide the user to input corresponding component configuration information.

[0040] For ease of understanding, see Figure 2 If the user clicks on the operation and maintenance problem input component in the rule generation area 220, the low-code platform responds to the trigger operation and displays the to-be-configured items corresponding to the operation and maintenance problem input component in the component property configuration area 230. The to-be-configured items can be used to remind the user to input the operation and maintenance problem according to the operation and maintenance problem input template; similarly, if the user clicks on the log component in the rule generation area 220, the low-code platform responds to the trigger operation and displays the to-be-configured items corresponding to the log component in the component property configuration area 230, which is used to guide the user to import the operation and maintenance log; similarly, if the user clicks on the knowledge base component in the rule generation area 220, the low-code platform responds to the trigger operation and displays the to-be-configured items corresponding to the knowledge base component in the component property configuration area 230, which is used to guide the user to import the knowledge base.

[0041] S130. Obtain component configuration information of the basic functional component based on the configuration operation of the to-be-configured item corresponding to the basic functional component.

[0042] In this embodiment, for the to-be-configured items corresponding to the basic function components displayed in the component property configuration area, the user can configure the basic function components according to the specific functions to be implemented by the basic function components in the after-sales operation and maintenance scenarios. The low-code platform can obtain the user's configuration operations such as clicking operations and input operations on the to-be-configured items corresponding to the basic function components to obtain the component configuration information, so that the basic function components have corresponding functions.

[0043] Optionally, the component configuration information can be understood as the attribute information of the basic functional component. Optionally, the component configuration information includes but is not limited to one or more of the following combinations: input rules, execution rules, output rules, format rules, and archiving rules.

[0044] S140: Generate operation and maintenance question and answer rules according to each basic functional component, the logical relationship between different basic functional components, and the component configuration information of each basic functional component.

[0045] In this embodiment, after the low-code platform determines the component configuration information of each basic functional component, it encapsulates the component configuration information of each basic functional component, and determines the code execution logic of different basic functional components based on the logical relationship between different basic functional components. Thus, the low-code platform generates operation and maintenance question and answer rules suitable for after-sales operation and maintenance scenarios by assembling each basic functional component, the logical relationship between different basic functional components, and the component configuration information of each basic functional component.

[0046] Among them, the operation and maintenance question and answer rules refer to the execution rules or algorithms that automatically check the operation and maintenance logs and search for answers that are suitable for the after-sales operation and maintenance scenarios in the after-sales operation and maintenance scenarios.

[0047] A method for generating operation and maintenance question and answer rules in an embodiment of the present disclosure includes a component providing area, a rule generating area and a component attribute configuration area on the configuration interface of a low-code platform, and the method includes: in response to a drag operation on a basic function component in the component providing area, displaying a connection relationship between a basic function component and different basic function components in the rule generating area, wherein the connection relationship between different basic function components represents a logical relationship between different basic function components; in response to a trigger operation on a basic function component in the rule generating area, displaying a to-be-configured item corresponding to the basic function component in the component attribute configuration area; based on a configuration operation on the to-be-configured item corresponding to the basic function component, obtaining component configuration information of the basic function component; and generating an operation and maintenance question and answer rule according to each basic function component, the logical relationship between different basic function components and the component configuration information of each basic function component. In this way, in the face of different after-sales operation and maintenance scenarios, the component configuration information of different basic function components and the logical relationship between different basic function components are determined on the low-code platform, thereby obtaining an operation and maintenance question and answer rule adapted to the after-sales operation and maintenance scenario in which it is located, so that based on the operation and maintenance question and answer rule, it is possible to automatically check the operation and maintenance log and search for answers adapted to the after-sales operation and maintenance scenario.

[0048] In another embodiment of the present disclosure, S130 in the operation and maintenance question and answer rule generation method is explained in detail.

[0049] In some embodiments, when the basic functional components include: an operation and maintenance problem input component, a log component, and a knowledge base component, S130 specifically includes the following steps:

[0050] Based on the configuration operation of the to-be-configured items corresponding to the operation and maintenance problem input component, an operation and maintenance problem input template is obtained, and the operation and maintenance problem input template is used as the component configuration information of the operation and maintenance problem input component; based on the configuration operation of the to-be-configured items corresponding to the log component, multiple candidate operation and maintenance logs are obtained, and the multiple candidate operation and maintenance logs are used as the component configuration information corresponding to the log component; based on the configuration operation of the to-be-configured items corresponding to the knowledge base component, a target knowledge base is obtained, and the target knowledge base is used as the component configuration information of the knowledge base component.

[0051] Among them, the operation and maintenance problem input template can be used to prompt the user to input the operation and maintenance problem. For example, the operation and maintenance problem input template: please enter the frame number, fault time range, and fault phenomenon.

[0052] The candidate operation and maintenance logs refer to operation and maintenance logs related to the operation and maintenance problems input by the user.

[0053] In some embodiments, the specific implementation method of "obtaining multiple candidate operation and maintenance logs based on the configuration operation of the to-be-configured items corresponding to the log component" includes but is not limited to the following method: in response to the configuration operation, obtaining the operation and maintenance log file name from the operation and maintenance log list stored in the operation and maintenance database, and generating an operation and maintenance log acquisition request carrying the operation and maintenance log file name; sending the operation and maintenance log acquisition request to the cloud, so that the cloud performs log parsing based on the operation and maintenance log file name carried in the operation and maintenance log acquisition request to obtain multiple candidate operation and maintenance logs; receiving multiple candidate operation and maintenance logs returned by the cloud.

[0054] Specifically, based on the configuration operation, the low-code platform retrieves the operation and maintenance log list from the operation and maintenance database, determines the operation and maintenance log file name corresponding to the configuration operation, and then, through the distributed real-time computing engine on the cloud, parses the log corresponding to the operation and maintenance log file name based on the pre-generated graph database to obtain multiple candidate operation and maintenance logs, and obtains multiple candidate operation and maintenance logs returned by the distributed real-time computing engine on the cloud.

[0055] In some embodiments, the specific implementation method of "obtaining multiple candidate operation and maintenance logs based on the configuration operation of the to-be-configured items corresponding to the log component" includes but is not limited to the following methods: obtaining the log search time range from the configuration operation; based on the log search time range, obtaining multiple candidate operation and maintenance logs from the operation and maintenance log list stored in the operation and maintenance database.

[0056] Specifically, if the configuration operation carries a log search time range, the low-code platform uses regular expressions to search for logs from the operation and maintenance log list stored in the operation and maintenance database based on the log search time range, and obtains multiple candidate operation and maintenance logs.

[0057] Among them, the specific implementation method of "obtaining the target knowledge base based on the configuration operation of the to-be-configured item corresponding to the knowledge base component" includes but is not limited to the following method: based on the configuration operation, obtaining the target knowledge base from the knowledge base list stored in the operation and maintenance database.

[0058] Specifically, the low-code platform selects the operation and maintenance knowledge base related to the configuration operation as the target knowledge base from the list of knowledge bases stored in the operation and maintenance database.

[0059] Therefore, by dragging and configuring components, the specialized knowledge base and operation and maintenance logs suitable for the after-sales operation and maintenance scenarios can be automatically checked for subsequent searches for answers suitable for the after-sales operation and maintenance scenarios.

[0060] Furthermore, the basic functional component also includes: a rule setting component; accordingly, S130 specifically includes the following steps: determining the current configuration path where the rule setting component is located; if the rule setting component is the last rule setting component on the current configuration path, then based on the configuration operation, obtaining the component configuration information of the rule setting component.

[0061] Specifically, different rule setting components may be located on different configuration paths, and a configuration path may include multiple different basic functional components. In order to simplify the operation and maintenance Q&A rules, if a configuration path contains multiple identical rule setting components, the low-code platform only sets the component configuration information for the last rule setting component among the multiple identical rule setting components on the configuration path.

[0062] In some embodiments, the rule setting component includes: a rule engine setting component; accordingly, the specific implementation method of "obtaining component configuration information of the rule setting component based on the configuration operation" includes but is not limited to the following methods: obtaining the log name and fault parameters from the configuration operation; based on the log name, obtaining multiple target operation and maintenance logs within a target time range from multiple candidate operation and maintenance logs; according to the fault parameters, obtaining the first fault description information from the multiple target operation and maintenance logs, and using the first fault description information as the component configuration information of the rule engine setting component.

[0063] The log name and fault parameters are business rule fields used to fuzzily search operation and maintenance logs and identify operation and maintenance problems.

[0064] Specifically, the low-code platform can call regular expressions to perform fuzzy searches on multiple target operation and maintenance logs based on the log name and target time range. If the target time range is not obtained, multiple target operation and maintenance logs are searched within a recent period of time based on the log name, thereby narrowing the log search scope. The multiple target operation and maintenance logs are then searched for related rows where fault parameters occur, and the first fault description information is obtained from these related rows.

[0065] Optionally, the fault parameters include but are not limited to key, value and other parameters.

[0066] For example, a search rule for fault description information is: in the log file of module X, a line related to the PBOX keyword is retrieved, and EOL Config PBOX:1 appears, which means that the vehicle has a pbox and is a high-configuration vehicle; accordingly, a process for determining the first fault description information includes: finding logs that meet the date range according to the log name, then finding the log lines that meet the date range and the keyword PBOX, and judging whether there is a result that satisfies EOL Config PBOX=1 in the text block. If so, it means that the vehicle is a high-configuration vehicle, thereby obtaining the first fault description information.

[0067] Therefore, by configuring components, the rule engine setting components suitable for the after-sales operation and maintenance scenarios are automatically configured for subsequent searches for answers suitable for the after-sales operation and maintenance scenarios.

[0068] Furthermore, the rule setting component also includes: a prompt word-inference chain component that combines prompt word optimization and reasoning chain optimization algorithms; accordingly, the specific implementation method of "obtaining component configuration information of the rule setting component based on the configuration operation" includes but is not limited to the following method: based on the configuration operation, obtaining the prompt word-inference chain algorithm from the rule optimization algorithm stored in the operation and maintenance database; using the prompt word optimization-inference chain optimization algorithm, obtaining the second fault description information from multiple candidate operation and maintenance logs, and using the second fault description information as the component configuration information of the prompt word-inference chain component.

[0069] Specifically, the operation and maintenance database pre-stores multiple rule optimization algorithms. If the user wants to configure the prompt word optimization-inference chain optimization algorithm (Prompt-Auto-CoT), the low-code platform responds to the configuration operation and directly obtains the algorithm from the rule optimization algorithm stored in the operation and maintenance database. The algorithm is used to improve the reasoning ability of large language models and the ability to generate outputs that are more in line with expectations, and obtain the second fault description information from multiple candidate operation and maintenance logs.

[0070] Among them, prompt refers to the process of guiding the model to generate more expected output by designing and optimizing input prompts when using a large language model. The Auto-CoT algorithm is a method for improving the capabilities of large language models by combining prompts with step-by-step thinking and increasing demonstration diversity.

[0071] Therefore, by configuring components, on the basis of automatically configuring the rule engine setting components that are suitable for the after-sales operation and maintenance scenarios, the prompt word-reasoning chain components that combine the prompt word optimization and the reasoning chain optimization algorithm are also automatically configured for subsequent searches for answers that are suitable for the after-sales operation and maintenance scenarios.

[0072] Further, the rule setting component only includes one of the rule engine setting component and the prompt word optimization and reasoning chain optimization prompt word-reasoning chain component, or when the rule setting component does not include the two types of components: the rule engine setting component and the prompt word optimization and reasoning chain optimization prompt word-reasoning chain component, the rule setting component also includes: a retrieval enhancement-search component that combines retrieval enhancement optimization and search engine algorithm; accordingly, the specific implementation method of "obtaining component configuration information of the rule setting component based on the configuration operation" includes but is not limited to the following method: based on the configuration operation, obtaining the retrieval enhancement optimization-search engine algorithm from the rule optimization algorithm stored in the operation and maintenance database; using the retrieval enhancement optimization-search engine algorithm, obtaining the third fault description information from multiple candidate operation and maintenance logs, and using the third fault description information as the component configuration information of the retrieval enhancement-search component.

[0073] Specifically, the operation and maintenance database pre-stores multiple rule optimization algorithms. If the user wants to configure the retrieval enhancement optimization-search engine algorithm (Rag+ search engine), the low-code platform responds to the configuration operation and directly obtains the algorithm from the rule optimization algorithm stored in the operation and maintenance database. It uses the algorithm's ability to retrieve relevant documents and obtains the third fault description information from multiple candidate operation and maintenance logs.

[0074] Among them, the retrieval enhancement optimization algorithm (Rag) accepts input and retrieves a set of related / supporting documents, and gives the source of the documents. These documents are combined as context and the original prompt words of the input and sent to the text generator to obtain the final output.

[0075] Therefore, by configuring components, on the basis of automatically configuring one of the rule engine setting components suitable for the after-sales operation and maintenance scenarios and the prompt word-reasoning chain components combining prompt word optimization and reasoning chain optimization algorithms, or without configuring the rule engine setting components suitable for the after-sales operation and maintenance scenarios and the prompt word-reasoning chain components combining prompt word optimization and reasoning chain optimization algorithms, a retrieval enhancement-search component combining retrieval enhancement optimization and search engine algorithm is automatically configured, and the retrieval enhancement-search component combining retrieval enhancement optimization and search engine algorithm is used as a fallback component for subsequent searches for answers suitable for the after-sales operation and maintenance scenarios.

[0076] Furthermore, the basic functional components also include: a preset question and answer model component; accordingly, S130 specifically includes the following steps: based on the configuration operation of the to-be-configured items corresponding to the preset question and answer model component, obtaining the prompt word optimization parameters and self-consistency parameters of the preset question and answer model; using the prompt word optimization parameters and self-consistency parameters as component configuration information of the preset question and answer model component.

[0077] Specifically, the low-code platform configures a large language model in the preset question-and-answer model component. In order to enable the large language model to more accurately answer questions that are suitable for after-sales operation and maintenance scenarios, the low-code platform configures prompt word optimization parameters and self-consistency parameters for the large language model in the preset question-and-answer model component, so that the large language model can select the most consistent answer from multiple answers to output when searching for answers.

[0078] Among them, the prompt word optimization parameter is used to guide the large language model to generate more expected search answers. The self-consistency parameter is used to guide the large language model to select the most consistent result from multiple results. The large language model is a deep learning model trained with a large amount of text data that can understand and generate natural language.

[0079] Therefore, by configuring components, the preset question-and-answer model components are automatically configured so that the output answers are adapted to the after-sales operation and maintenance scenarios.

[0080] Furthermore, the basic functional components also include: an answer output setting component; accordingly, S130 specifically includes the following steps: based on the configuration operation of the to-be-configured items corresponding to the answer output setting component, obtaining the output format parameters of the preset question and answer model, and using the output format parameters as the component configuration information of the answer output setting component.

[0081] The output format is used to optimize the output of the output results so that the searched answers are output in a standardized format, which is convenient for analyzing the searched answers. Optionally, the output format includes: problem cause, analysis process, log, and log path.

[0082] Therefore, by configuring the components, the answer output setting component is automatically configured to optimize the output format of the answer.

[0083] Furthermore, the basic functional components also include: an operation and maintenance result output component; accordingly, S130 specifically includes the following steps: based on the configuration operation of the to-be-configured item corresponding to the operation and maintenance result output component, obtaining the threshold and priority of the output answer of the rule setting component; using the threshold and priority of the output answer of the rule setting component as the component configuration information of the operation and maintenance result output component.

[0084] In order to further optimize the operation and maintenance question and answer rules, the low-code platform can also set the threshold and priority of the output answer for one or more of the above three rule setting components, so that after determining the output answer using one or more of the above three rule setting components, the matching value between the operation and maintenance question and the output answer is calculated, and the matching value is compared with the threshold, and the final answer is output based on the comparison result and priority.

[0085] For example, the threshold of the rule engine setting component is set to 0.7 and the priority is 0; the threshold of the prompt word-inference chain component that combines the prompt word optimization and the reasoning chain optimization algorithm is set to 0.6 and the priority is 1; the threshold of the retrieval enhancement-search component that combines the retrieval enhancement optimization and the search engine is set to 0 and the priority is 3.

[0086] In the specific implementation, one implementation method is: if the matching degree value between the operation and maintenance problem input this time and the output answer is greater than 0.7, then the output answer determined by the rule engine setting component is directly output; if the matching value between the operation and maintenance problem input this time and the output answer is greater than 0.6 and less than 0.7, then the output answer determined by the prompt word-reasoning chain component based on the combination of prompt word optimization and reasoning chain optimization algorithm is directly output; if the matching value between the operation and maintenance problem input this time and the output answer is less than 0.6, then the output answer determined by the retrieval enhancement-search component based on the combination of retrieval enhancement optimization and search engine algorithm is directly output. Another implementation method is: in terms of execution, it can be selected to use the above rule setting components to determine the output answer in parallel, and compare the matching degree between the output answer determined by the three rule setting components and the operation and maintenance problem with the thresholds and priorities of the above three rules, or execute in the order of high priority. If the priority is high and the threshold meets the requirements, then the result is directly output, and other low priority solutions are not executed to save computing resources. Another implementation method is: the output answers determined based on the three rule setting components will be output, and the output of each solution will be compared with the operation and maintenance problem input this time for the degree of matching. However, the priorities and thresholds of the three rule setting components are not set, or the priorities and thresholds of the three rule setting components are set to the same, then the output result with the highest degree of matching is directly output.

[0087] Therefore, by configuring components and utilizing the operation and maintenance result output component, the output answers are output according to specific classifications to ensure the orderliness of the output answers while saving computing resources.

[0088] Furthermore, the basic functional components also include: an archiving component; accordingly, S130 specifically includes the following steps: based on the configuration operation of the to-be-configured items corresponding to the archiving component, obtaining the archiving supplementary information of the output answer of the preset question and answer model, and using the archiving supplementary information as the component configuration information of the archiving component.

[0089] Specifically, the low-code platform can also configure archiving components according to the needs of after-sales operation and maintenance scenarios. Specifically, it can configure the archived supplementary information of the output answers for the preset question and answer model, and finally collect it into the historical operation and maintenance troubleshooting result cases. The archived supplementary information of the output answers can also be used as part of the target knowledge base to parameterize subsequent model training.

[0090] Therefore, by configuring components, it is automatically determined whether to archive the output answer and to configure archiving supplementary information for the output answer, thereby expanding the amount of information in the knowledge base to provide richer model training data.

[0091] Figure 3A schematic diagram of the structure of an operation and maintenance question and answer rule generation device provided by an embodiment of the present disclosure is shown. The operation and maintenance question and answer rule generation device shown in 3 is configured on a low-code platform, and the configuration interface of the low-code platform includes a component provision area, a rule generation area, and a component attribute configuration area.

[0092] like Figure 3 As shown, the operation and maintenance question and answer rule generating device 300 may include:

[0093] A component display module 310, configured to display the basic functional component and the connection relationship between different basic functional components in the rule generation area in response to a drag operation on the basic functional component in the component providing area, wherein the connection relationship between different basic functional components represents the logical relationship between different basic functional components;

[0094] A component triggering module 320, configured to respond to a triggering operation on a basic functional component in the rule generation area and display a to-be-configured item corresponding to the basic functional component in the component attribute configuration area;

[0095] A configuration information acquisition module 330, configured to acquire component configuration information of the basic functional component based on a configuration operation on a to-be-configured item corresponding to the basic functional component;

[0096] The rule generation module 340 is used to generate operation and maintenance question and answer rules according to each of the basic functional components, the logical relationship between different basic functional components, and the component configuration information of each of the basic functional components.

[0097] An operation and maintenance question and answer rule generation device of an embodiment of the present disclosure, in response to the drag operation of the basic function component in the component providing area, displays the connection relationship between the basic function component and different basic function components in the rule generation area, wherein the connection relationship between different basic function components represents the logical relationship between different basic function components; in response to the trigger operation of the basic function component in the rule generation area, displays the to-be-configured items corresponding to the basic function component in the component attribute configuration area; based on the configuration operation of the to-be-configured items corresponding to the basic function component, obtains the component configuration information of the basic function component; generates operation and maintenance question and answer rules according to each basic function component, the logical relationship between different basic function components, and the component configuration information of each basic function component. In this way, in the face of different after-sales operation and maintenance scenarios, the component configuration information of different basic function components and the logical relationship between different basic function components are determined on the low-code platform, thereby obtaining operation and maintenance question and answer rules adapted to the after-sales operation and maintenance scenarios in which they are located, so that based on the operation and maintenance question and answer rules, it is possible to realize automatic troubleshooting of operation and maintenance logs and search for answers adapted to the after-sales operation and maintenance scenarios.

[0098] In some embodiments of the present disclosure, the basic functional components include: an operation and maintenance problem input component, a log component, and a knowledge base component;

[0099] Accordingly, the configuration information acquisition module 330 includes:

[0100] A first acquisition unit, configured to acquire an operation and maintenance problem input template based on a configuration operation of a to-be-configured item corresponding to the operation and maintenance problem input component, and use the operation and maintenance problem input template as component configuration information of the operation and maintenance problem input component;

[0101] A second acquisition unit is used to acquire a plurality of candidate operation and maintenance logs based on a configuration operation on the to-be-configured item corresponding to the log component, and use the plurality of candidate operation and maintenance logs as component configuration information corresponding to the log component;

[0102] The third acquisition unit is used to acquire a target knowledge base based on a configuration operation on the to-be-configured item corresponding to the knowledge base component, and use the target knowledge base as component configuration information of the knowledge base component.

[0103] In some embodiments of the present disclosure, the second acquisition unit is specifically configured to:

[0104] In response to the configuration operation, obtaining an operation and maintenance log file name from an operation and maintenance log list stored in an operation and maintenance database, and generating an operation and maintenance log acquisition request carrying the operation and maintenance log file name;

[0105] Sending the operation and maintenance log acquisition request to the cloud, so that the cloud performs log parsing based on the operation and maintenance log file name carried in the operation and maintenance log acquisition request to obtain the multiple candidate operation and maintenance logs;

[0106] Receive the multiple candidate operation and maintenance logs returned by the cloud.

[0107] In some embodiments of the present disclosure, the second acquisition unit is specifically configured to:

[0108] Obtaining a log search time range from the configuration operation;

[0109] Based on the log search time range, the multiple candidate operation and maintenance logs are obtained from an operation and maintenance log list stored in an operation and maintenance database.

[0110] In some embodiments of the present disclosure, the basic functional component further includes: a rule setting component;

[0111] Accordingly, the configuration information acquisition module 330 includes:

[0112] A first determining unit, configured to determine a current configuration path where the rule setting component is located;

[0113] A fourth acquisition unit is configured to acquire component configuration information of the rule setting component based on the configuration operation if the rule setting component is the last rule setting component on the current configuration path.

[0114] In some embodiments of the present disclosure, the basic functional component further includes: a rule engine setting component;

[0115] Accordingly, the configuration information acquisition module 330 includes:

[0116] A fifth acquisition unit, configured to acquire a log name and a fault parameter from the configuration operation;

[0117] A sixth acquisition unit, configured to acquire, based on the log name, a plurality of target operation and maintenance logs within a target time range from the plurality of candidate operation and maintenance logs;

[0118] A seventh acquisition unit is used to acquire first fault description information from the multiple target operation and maintenance logs according to the fault parameters, and use the first fault description information as component configuration information of the rule engine setting component.

[0119] In some embodiments of the present disclosure, the basic functional component further includes: a prompt word-reasoning chain component that combines prompt word optimization and reasoning chain optimization algorithms;

[0120] Accordingly, the configuration information acquisition module 330 includes:

[0121] An eighth acquisition unit, configured to acquire a prompt word-reasoning chain algorithm from a rule optimization algorithm stored in an operation and maintenance database based on the configuration operation;

[0122] The ninth acquisition unit is used to use the prompt word optimization-inference chain optimization algorithm to acquire second fault description information from the multiple candidate operation and maintenance logs, and use the second fault description information as component configuration information of the prompt word-inference chain component.

[0123] In some embodiments of the present disclosure, the basic functional components further include: a retrieval enhancement-search component that combines retrieval enhancement optimization and a search engine algorithm;

[0124] Accordingly, the configuration information acquisition module 330 includes:

[0125] A tenth acquisition unit, configured to acquire a retrieval enhancement optimization-search engine algorithm from a rule optimization algorithm stored in an operation and maintenance database based on the configuration operation;

[0126] The eleventh acquisition unit is used to use the retrieval enhancement optimization-search engine algorithm to acquire third fault description information from the multiple candidate operation and maintenance logs, and use the third fault description information as component configuration information of the retrieval enhancement-search component.

[0127] In some embodiments of the present disclosure, the basic functional component further includes: a preset question-answering model component;

[0128] Accordingly, the configuration information acquisition module 330 includes:

[0129] a twelfth acquisition unit, configured to acquire a prompt word optimization parameter and a self-consistency parameter of a preset question-answering model based on a configuration operation of the to-be-configured item corresponding to the preset question-answering model component;

[0130] The second determination unit is used to use the prompt word optimization parameter and the self-consistency parameter as component configuration information of the preset question and answer model component.

[0131] In some embodiments of the present disclosure, the basic functional component further includes: an operation and maintenance result output component;

[0132] Accordingly, the configuration information acquisition module 330 includes:

[0133] A fourteenth obtaining unit, configured to obtain a threshold and a priority of an output answer of the rule setting component based on a configuration operation on the to-be-configured item corresponding to the operation and maintenance result output component;

[0134] The third determining unit is used to set the threshold and priority of the output answer of the rule setting component as the component configuration information of the operation and maintenance result output component.

[0135] It should be noted that Figure 3 The operation and maintenance question and answer rule generation device 300 shown can execute Figure 1 to Figure 2 The various steps in the method embodiment shown in the figure are implemented Figure 1 to Figure 2 The various processes and effects in the method embodiment shown are not described in detail here.

[0136] Figure 4 A structural diagram of a low-code platform provided by an embodiment of the present disclosure is shown.

[0137] Among them, the configuration interface of the low-code platform includes the component provision area, the rule generation area and the component property configuration area.

[0138] like Figure 4 As shown, the low-code platform may include a processor 401 and a memory 402 storing computer program instructions.

[0139] Specifically, the processor 401 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0140] The memory 402 may include a large capacity memory for information or instructions. By way of example and not limitation, the memory 402 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive or a combination of two or more of these. Where appropriate, the memory 402 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 402 may be inside or outside the integrated gateway device. In a particular embodiment, the memory 402 is a non-volatile solid-state memory. In a particular embodiment, the memory 402 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (Electrically Erasable Programmable ROM, EEPROM), an electrically rewritable ROM (EAROM) or a flash memory, or a combination of two or more of these.

[0141] The processor 401 reads and executes the computer program instructions stored in the memory 402 to perform the steps of the method for generating operation and maintenance question and answer rules provided in the embodiment of the present disclosure.

[0142] In one example, the low-code platform may further include a transceiver 403 and a bus 404. Figure 4 As shown, the processor 401, the memory 402 and the transceiver 403 are connected via a bus 404 and communicate with each other.

[0143] The bus 404 includes hardware, software, or both. For example, but not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a Memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, the bus 404 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0144] The following is an embodiment of a computer-readable storage medium provided in an embodiment of the present disclosure. The computer-readable storage medium and the operation and maintenance question and answer rule generation method of the above-mentioned embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the computer-readable storage medium, please refer to the embodiment of the above-mentioned operation and maintenance question and answer rule generation method.

[0145] This embodiment provides a storage medium including computer executable instructions. When the computer executable instructions are executed by a computer processor, they are used to execute a method for generating operation and maintenance question and answer rules. The method is applied to a low-code platform. The configuration interface of the low-code platform includes a component provision area, a rule generation area, and a component attribute configuration area. The method includes:

[0146] In response to a drag operation on a basic functional component in the component providing area, displaying the basic functional component and connection relationships between different basic functional components in the rule generating area, wherein the connection relationships between different basic functional components represent logical relationships between different basic functional components;

[0147] In response to a triggering operation on a basic functional component in the rule generation area, displaying a to-be-configured item corresponding to the basic functional component in the component attribute configuration area;

[0148] Based on the configuration operation of the to-be-configured item corresponding to the basic functional component, obtaining component configuration information of the basic functional component;

[0149] An operation and maintenance question and answer rule is generated according to each of the basic functional components, the logical relationship between different basic functional components, and the component configuration information of each basic functional component.

[0150] Of course, the storage medium containing computer executable instructions provided by the embodiment of the present disclosure is not limited to the above method operations, and the computer executable instructions can also execute related operations in the operation and maintenance question and answer rule generation method provided by any embodiment of the present disclosure.

[0151] Through the above description of the implementation method, the technical personnel in the relevant field can clearly understand that the present disclosure can be implemented with the help of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present disclosure is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions to enable a computer cloud platform (which can be a personal computer, server, or network cloud platform, etc.) to execute the operation and maintenance question and answer rule generation method provided by each embodiment of the present disclosure.

[0152] Note that the above are only preferred embodiments of the present disclosure and the technical principles used. Those skilled in the art will understand that the present disclosure is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present disclosure. Therefore, although the present disclosure is described in more detail through the above embodiments, the present disclosure is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present disclosure, and the scope of the present disclosure is determined by the scope of the appended claims.

Claims

1. A method for generating operation and maintenance question and answer rules, characterized in that: Applied to a low-code platform, the configuration interface of the low-code platform includes a component provision area, a rule generation area, and a component attribute configuration area, and the method includes: In response to a drag operation on a basic functional component in the component providing area, displaying the basic functional component and connection relationships between different basic functional components in the rule generating area, wherein the connection relationships between different basic functional components represent logical relationships between different basic functional components; In response to a triggering operation on a basic functional component in the rule generation area, displaying a to-be-configured item corresponding to the basic functional component in the component attribute configuration area; Based on the configuration operation of the to-be-configured item corresponding to the basic functional component, obtaining component configuration information of the basic functional component; An operation and maintenance question and answer rule is generated according to each of the basic functional components, the logical relationship between different basic functional components, and the component configuration information of each basic functional component.

2. The method according to claim 1, characterized in that The basic functional components include: operation and maintenance problem input component, log component and knowledge base component; Correspondingly, the obtaining of component configuration information of the basic functional component based on the configuration operation of the to-be-configured item corresponding to the basic functional component includes: Based on the configuration operation of the to-be-configured item corresponding to the operation and maintenance problem input component, an operation and maintenance problem input template is obtained, and the operation and maintenance problem input template is used as component configuration information of the operation and maintenance problem input component; Based on the configuration operation of the to-be-configured item corresponding to the log component, a plurality of candidate operation and maintenance logs are obtained, and the plurality of candidate operation and maintenance logs are used as component configuration information corresponding to the log component; Based on the configuration operation of the to-be-configured item corresponding to the knowledge base component, a target knowledge base is acquired, and the target knowledge base is used as component configuration information of the knowledge base component.

3. The method according to claim 2, characterized in that The obtaining of a plurality of candidate operation and maintenance logs based on the configuration operation of the to-be-configured item corresponding to the log component includes: In response to the configuration operation, obtaining an operation and maintenance log file name from an operation and maintenance log list stored in an operation and maintenance database, and generating an operation and maintenance log acquisition request carrying the operation and maintenance log file name; Sending the operation and maintenance log acquisition request to the cloud, so that the cloud performs log parsing based on the operation and maintenance log file name carried in the operation and maintenance log acquisition request to obtain the multiple candidate operation and maintenance logs; Receive the multiple candidate operation and maintenance logs returned by the cloud.

4. The method according to claim 2, characterized in that: The obtaining of a plurality of candidate operation and maintenance logs based on the configuration operation of the to-be-configured item corresponding to the log component includes: Obtaining a log search time range from the configuration operation; Based on the log search time range, the multiple candidate operation and maintenance logs are obtained from an operation and maintenance log list stored in an operation and maintenance database.

5. The method according to claim 2, characterized in that: The basic functional components also include: a rule setting component; Correspondingly, the obtaining of component configuration information of the basic functional component based on the configuration operation of the to-be-configured item corresponding to the basic functional component includes: Determine the current configuration path where the rule setting component is located; If the rule setting component is the last rule setting component on the current configuration path, then based on the configuration operation, component configuration information of the rule setting component is obtained.

6. The method according to claim 5, characterized in that The rule setting component includes: a rule engine setting component; Correspondingly, the acquiring component configuration information of the rule setting component based on the configuration operation includes: Obtaining a log name and fault parameters from the configuration operation; Based on the log name, obtaining multiple target operation and maintenance logs within a target time range from the multiple candidate operation and maintenance logs; According to the fault parameters, first fault description information is obtained from the multiple target operation and maintenance logs, and the first fault description information is used as component configuration information of the rule engine setting component.

7. The method according to claim 6, characterized in that The rule setting component also includes: a prompt word-reasoning chain component that combines prompt word optimization and reasoning chain optimization algorithms; Correspondingly, the acquiring component configuration information of the rule setting component based on the configuration operation includes: Based on the configuration operation, a prompt word-reasoning chain algorithm is obtained from the rule optimization algorithm stored in the operation and maintenance database; The prompt word optimization-inference chain optimization algorithm is used to obtain second fault description information from the multiple candidate operation and maintenance logs, and the second fault description information is used as component configuration information of the prompt word-inference chain component.

8. The method according to claim 5, 6 or 7, characterized in that: The rule setting component also includes: a retrieval enhancement-search component that combines retrieval enhancement optimization and search engine algorithms; Correspondingly, the acquiring component configuration information of the rule setting component based on the configuration operation includes: Based on the configuration operation, a retrieval enhancement optimization-search engine algorithm is obtained from the rule optimization algorithm stored in the operation and maintenance database; The retrieval enhancement optimization-search engine algorithm is used to obtain third fault description information from the multiple candidate operation and maintenance logs, and the third fault description information is used as component configuration information of the retrieval enhancement-search component.

9. The method according to claim 5, characterized in that The basic functional components also include: a preset question-answering model component; Correspondingly, the obtaining of component configuration information of the basic functional component based on the configuration operation of the to-be-configured item corresponding to the basic functional component includes: Based on the configuration operation of the to-be-configured item corresponding to the preset question-answering model component, obtaining the prompt word optimization parameter and the self-consistency parameter of the preset question-answering model; The prompt word optimization parameter and the self-consistency parameter are used as component configuration information of the preset question-answering model component.

10. The method according to claim 9, characterized in that The basic functional components also include: an operation and maintenance result output component; Correspondingly, the obtaining of component configuration information of the basic functional component based on the configuration operation of the to-be-configured item corresponding to the basic functional component includes: Based on the configuration operation of the to-be-configured item corresponding to the operation and maintenance result output component, obtaining the threshold and priority of the output answer of the rule setting component; The threshold and priority of the output answer of the rule setting component are used as component configuration information of the operation and maintenance result output component.

11. A device for generating operation and maintenance question and answer rules, characterized in that: Configured on a low-code platform, the configuration interface of the low-code platform includes a component provision area, a rule generation area, and a component attribute configuration area, and the device includes: A component display module, configured to, in response to a drag operation on a basic function component in the component providing area, display the basic function component and a connection relationship between different basic function components in the rule generating area, wherein the connection relationship between different basic function components represents a logical relationship between different basic function components; A component triggering module, configured to respond to a triggering operation on a basic functional component in the rule generation area and display a to-be-configured item corresponding to the basic functional component in the component attribute configuration area; A configuration information acquisition module, used to acquire component configuration information of the basic functional component based on a configuration operation on a to-be-configured item corresponding to the basic functional component; The rule generation module is used to generate operation and maintenance question and answer rules according to each of the basic functional components, the logical relationship between different basic functional components, and the component configuration information of each basic functional component.

12. A low-code platform, characterized in that: include: processor; A memory for storing executable instructions; The processor is used to read the executable instructions from the memory and execute the executable instructions to implement the method described in any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the method according to any one of claims 1 to 10.

Images