Operation and maintenance method and system

Abstract

The present application belongs to the technical field of operation and maintenance. Disclosed are an operation and maintenance method and system. The method comprises: acquiring a task request, wherein the task request is a user intent described in natural language, and the user intent comprises task parameters and a task objective; generating a target task template on the basis of the task objective, wherein the target task template comprises task steps for achieving the task objective; and executing, according to the task parameters, the task steps comprised in the target task template, so as to obtain a task result matching the task request. By means of the method, even if no task template matching a user intent is preset in an operation and maintenance system, the operation and maintenance system can also automatically generate a task template matching the user intent and execute the task template, thereby completing an operation and maintenance task. That is, the method does not require a developer to develop task templates one by one for different operation and maintenance tasks, and thus the method is more intelligent and has a better user experience.

Description

Operation and maintenance method and system

[0001] The present application claims priority to the Chinese patent application No. 202411255218.1, filed on September 6, 2024, and entitled "Operation and maintenance method and system", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the technical field of operation and maintenance, in particular to an operation and maintenance method and system. BACKGROUND

[0003] With the increasing maturity of large model technology, applying large model technology in operation and maintenance systems can significantly improve operation and maintenance efficiency. Among them, a large model is a tool that can handle complex tasks and understand natural language.

[0004] In current operation and maintenance products based on large models, it is usually necessary to develop use cases one by one for operation and maintenance tasks in specific scenarios to obtain task templates that can perform different operation and maintenance tasks. However, this case-by-case development of use cases to obtain task templates not only consumes time and effort, but also is difficult to meet the diversified requirements in the operation and maintenance environment. SUMMARY

[0005] The present application provides an operation and maintenance method and system. The method does not require developers to develop task templates one by one for different operation and maintenance tasks when operating and maintaining network equipment, thus being more intelligent and providing better user experience.

[0006] The technical solutions provided by the present application are as follows:

[0007] In a first aspect, the present application provides an operation and maintenance method, which comprises: obtaining a task request, the task request being a user intent described in natural language, the user intent including a task parameter and a task target; generating a target task template according to the task target; and executing a task step included in the target task template according to the task parameter to obtain a task result matched with the task request. The target task template includes a task step for achieving the task target.

[0008] Through the method provided by the present application, even if there is no pre-installed task template matched with the user intent in the operation and maintenance system, the operation and maintenance system can automatically generate a task template matched with the user intent and execute the task template to complete the operation and maintenance task. It can be seen that the method provided by the present application does not require developers to develop task templates one by one for different operation and maintenance tasks, thus being more intelligent and providing better user experience.

[0009] In a possible design, the method further includes: determining whether a target task template exists in the task template set according to the task target. The target task template is generated according to the task target, including: in a case where the target task template does not exist in the task template set, generating the target task template according to the task target.

[0010] Through the possible design, when the target task template for achieving the task target is configured in the operation and maintenance system, the template can be directly called and executed, so that the response speed of the operation and maintenance system can be improved. When the target task template is not preconfigured in the operation and maintenance system, the operation and maintenance system can automatically generate the target task template matched with the user intention and execute the target task template, that is, the operation and maintenance system no longer depends on the task template developed by the developer for the operation and maintenance task, so that the operation and maintenance of the network device is more intelligent.

[0011] In another possible design, the target task template is generated according to the task target, including: inputting the task target into a large language model (LLM) to obtain a target task template output by the LLM.

[0012] Through the possible design, the operation and maintenance system learns and reasons the task target through the LLM to obtain the target task template, so that the purpose of automatically generating the target task template matched with the user intention is achieved.

[0013] In yet another possible design, the method further includes: receiving an abnormal alarm reported by the network device; determining a target de-noising rule according to the abnormal alarm and context information of the abnormal alarm; and filtering invalid alarms in the received abnormal alarm according to the target de-noising rule. The context information includes: a historical alarm of the same type as the abnormal alarm and a de-noising rule determined based on the historical alarm.

[0014] Through the possible design, the purpose of filtering invalid alarms from the abnormal alarms received by the operation and maintenance system is achieved, so that not only the storage resources required by the operation and maintenance system for storing the abnormal alarms can be saved, but also the valid alarms can be prevented from being submerged by the invalid alarms so as to be unable to be processed in time, and the human and time costs consumed when all the abnormal alarms are checked and processed one by one can be saved, so that the user experience is improved.

[0015] In yet another possible design, the method further includes: outputting first task information, the first task information including the target task template and a task parameter; and receiving response information of the first task information. The task steps included in the target task template are executed according to the task parameter, including: in a case where the response information indicates that the first task information is correct, the task steps included in the target task template are executed according to the task parameter.

[0016] In a further possible design, the method further includes: in response to the response information carrying the second task information, performing a task step recorded in the second task information according to a task parameter included in the second task information. The second task information is task information obtained by modifying the first task information.

[0017] Through the above two possible designs, the operation and maintenance system can confirm the task parameter and the target task template with the user, and perform the task step included in the task template confirmed by the user according to the task parameter confirmed by the user after the user confirms. That is, the present application can provide the function of interaction between the operation and maintenance system and the user in the process of the operation and maintenance system responding to the task request initiated by the user, so that the user can confirm whether the target task template generated by the operation and maintenance system is reasonable, and / or confirm whether the task parameter is extracted accurately, and / or modify the target task template and / or the task parameter according to the latest demand of the user, and so on, so as to improve the user experience.

[0018] In a further possible design, the task parameter includes an input parameter. Alternatively, the task parameter includes an input parameter and an execution parameter. The input parameter is a parameter input into the target task template when the target task template is executed, and the execution parameter is a parameter triggering execution of the target task template.

[0019] In a further possible design, the execution parameter includes a start time triggering execution of the target task template and / or a cycle duration triggering periodic execution of the target task template. Alternatively, the execution parameter includes a type of an event triggering execution of the target task template and / or an identifier (ID) of the event.

[0020] Through the above two possible designs, the present application obtains the execution parameter included in the user intention, so that the operation and maintenance system can automatically trigger execution of the target task template according to the execution parameter in the subsequent process, thereby realizing truly unattended operation and maintenance.

[0021] In a further possible design, the method includes: receiving the task request input in a natural language by a language user interface (LUI) or a graphical user interface (GUI).

[0022] Through the possible design, the purpose of obtaining the task request described in a natural language by the user can be achieved.

[0023] In a further possible design, the method is applied to a park network.

[0024] In a second aspect, an embodiment of the present application provides an operation and maintenance system, comprising: an acquisition unit configured to acquire a task request, the task request being a user intention described in natural language, the user intention comprising a task parameter and a task target; a generation unit configured to generate a target task template according to the task target, the target task template comprising a task step for realizing the task target; and an execution unit configured to execute the task step comprised in the target task template according to the task parameter, to obtain a task result matched with the task request.

[0025] In a possible design, the operation and maintenance system further comprises a judgment unit configured to judge whether the target task template exists in a task template set according to the task target. The generation unit is specifically configured to generate the target task template according to the task target in a case where the target task template does not exist in the task template set.

[0026] In another possible design, the generation unit is specifically configured to input the task target into an LLM to obtain the target task template output by the LLM.

[0027] In yet another possible design, the operation and maintenance system further comprises: a receiving unit configured to receive an abnormal alarm reported by a network device; a determination unit configured to determine a target de-noising rule according to the abnormal alarm and context information of the abnormal alarm; and a filtering unit configured to filter invalid alarms in the received abnormal alarms according to the target de-noising rule. The context information comprises a historical alarm of the same type as the abnormal alarm and a de-noising rule determined based on the historical alarm.

[0028] In yet another possible design, the operation and maintenance system further comprises: an output unit configured to output first task information, the first task information comprising the target task template and the task parameter; and a receiving unit configured to receive response information of the first task information. The execution unit is specifically configured to execute the task step comprised in the target task template according to the task parameter in a case where the response information indicates that the first task information is correct.

[0029] In yet another possible design, the execution unit is further configured to execute a task step recorded in second task information according to a task parameter contained in the second task information in a case where the response information carries the second task information. The second task information is task information obtained by modifying the first task information.

[0030] In yet another possible design, the task parameter comprises an input parameter. Alternatively, the task parameter comprises an input parameter and an execution parameter. The input parameter is a parameter input into the target task template when the target task template is executed, and the execution parameter is a parameter triggering the execution of the target task template.

[0031] In yet another possible design, the execution parameter includes a start time of triggering the execution of the target task template and / or a period length of triggering the execution of the target task template periodically. Alternatively, the execution parameter includes a type of an event of triggering the execution of the target task template and / or an ID of the event.

[0032] In yet another possible design, the generating unit, the judging unit and the determining unit are implemented by an artificial intelligence agent core (AI agent core) based on LLM.

[0033] In yet another possible design, the generating unit, the judging unit and the determining unit are integrated in one device, and units other than the generating unit, the judging unit and the determining unit are integrated in another device.

[0034] It can be understood that the operation and maintenance system provided by the second aspect and any possible design of the second aspect has the beneficial effects of the corresponding solutions provided by the first aspect and any possible design of the first aspect, which will not be repeated here.

[0035] In a third aspect, the present application provides an operation and maintenance system, comprising a memory, a communication interface and one or more processors, the one or more processors receive or send data through the communication interface, and the one or more processors are configured to read program instructions stored in the memory to execute the method provided by the first aspect and any possible design of the first aspect.

[0036] In a fourth aspect, the present application provides a computer program product comprising instructions, which, when executed by a processor, cause the processor or a device or system comprising the processor to execute the method provided by the first aspect and any possible design of the first aspect.

[0037] In a fifth aspect, the present application provides a computer readable storage medium, which is a non-volatile computer readable storage medium, and the computer readable storage medium comprises computer program instructions, which, when executed by a processor, a device or a system comprising the processor, execute the method provided by the first aspect and any possible design of the first aspect.

[0038] In a sixth aspect, the present application provides a chip, which comprises a processor for running program instructions or codes, and the chip or a device comprising the chip can be used to execute the method provided in the first aspect and any possible implementation manner of the first aspect. For example, the chip further comprises an input interface, an output interface and a memory. The input interface, the output interface, the processor and the memory of the chip are connected through an internal connection path of the chip. The memory in the chip is used to store the program instructions or codes run by the processor. The input interface and the output interface of the chip are used for the connection and communication between the chip and other chips or devices.

[0039] It can be understood that any of the operation and maintenance system, the computer readable storage medium, the computer program product or the chip provided in the above can be applied to the corresponding method provided in the above, and the beneficial effects achieved thereby can refer to the beneficial effects in the corresponding method, which will not be described here.

[0040] In the present application, the names of the operation and maintenance system and the like do not constitute a limitation on the devices or functional modules themselves, and in actual implementation, these devices or functional modules can appear with other names. As long as the functions of the devices or functional modules are similar to those in the present application, they are within the protection scope of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0041] FIG. 1 is a schematic diagram of an operation and maintenance implementation;

[0042] FIG. 2 is a schematic diagram of a network scenario of the method provided in the embodiment of the present application;

[0043] FIG. 3 is a schematic diagram of a network scenario of the method provided in the embodiment of the present application;

[0044] FIG. 4 is another schematic diagram of a network scenario of the method provided in the embodiment of the present application;

[0045] FIG. 5 is still another schematic diagram of a network scenario of the method provided in the embodiment of the present application;

[0046] FIG. 6 is a flowchart of an operation and maintenance method provided in the embodiment of the present application;

[0047] FIG. 7 is a flowchart of an operation and maintenance method provided in the embodiment of the present application;

[0048] FIG. 8 is another flowchart of an operation and maintenance method provided in the embodiment of the present application;

[0049] FIG. 9 is a flowchart of another operation and maintenance method provided in the embodiment of the present application;

[0050] FIG. 10 is a structural schematic diagram of an operation and maintenance system provided in the embodiment of the present application;

[0051] FIG. 11 is another flow diagram of the operation and maintenance method according to an embodiment of the present application;

[0052] FIG. 12 is another flow diagram of the operation and maintenance method according to an embodiment of the present application;

[0053] FIG. 13 is another flow diagram of the operation and maintenance method according to an embodiment of the present application;

[0054] FIG. 14 is another structural diagram of the operation and maintenance system according to an embodiment of the present application;

[0055] FIG. 15 is a structural diagram of a computing device according to an embodiment of the present application. DETAILED DESCRIPTION

[0056] For the purpose, technical solutions and advantages of the present application to be clearer, the embodiments of the present application will be further described in detail below with reference to the drawings.

[0057] For the purpose, technical solutions and advantages of the present application to be clearer, the embodiments of the present application will be further described in detail below with reference to the drawings.

[0058] 1), LLM

[0059] The large language model LLM is also called a large model. The large language model refers to a deep learning model trained using a large amount of text data. The large language model can generate natural language text or understand the meaning of language text. The large language model can be used in natural language processing, machine translation, text generation and other fields. For example, natural language tasks such as text classification tasks, question and answer tasks, dialogue tasks, etc.

[0060] The large language model usually consists of multiple neural network layers, such as feedforward neural network layers, recurrent neural network layers, etc. Among them, the feedforward neural network layer is used for feature extraction of the input text, and the recurrent neural network is used for modeling of the text sequence. In this way, by training the model including multiple neural network layers using a large amount of text data, the large language model can be obtained. The large language model has strong understanding and prediction ability for natural language, and thus can generate accurate and natural text output.

[0061] 2), LUI

[0062] The LUI is an interface for realizing interaction between a device and a user through natural language. The LUI allows the user to communicate with the device using voice or text input, and the device responds to the user through language understanding and language generation technologies. For example, when the device is a terminal, the LUI is implemented as a smart assistant, voice assistant, etc. of the terminal.

[0063] 3), GUI

[0064] A GUI is an interface that enables interaction between a device and a user through graphical elements such as icons, buttons, windows, and the like. A GUI provides an intuitive way to operate applications on a device such as a computer device or a mobile terminal device. Currently, most operating systems and applications use GUIs and user interactions, which are easy to use and user-friendly.

[0065] 4), an artificial intelligence agent core (AI agent core)

[0066] An AI agent core is a reasoning module implemented based on an LLM in an artificial intelligence agent (AI agent). The AI agent core can perceive the environment, make decisions, and perform actions to achieve specific goals, and can optimize its own capabilities through learning to adapt to changes in the environment and solve complex problems. The AI agent core can be applied in multiple fields, such as autonomous driving, medical diagnosis, financial analysis, and the like.

[0067] Exemplarily, the hardware implementation of the AI agent core includes, but is not limited to, a graphics processing unit (GPU), a neural processing unit (NPU), or a central processing unit (CPU), and the like.

[0068] In current large model-based operation and maintenance products, it is usually necessary to develop use cases one by one for operation and maintenance tasks in specific scenarios to obtain task templates capable of executing different operation and maintenance tasks. The task templates include task steps required to achieve the task target. The task steps required to achieve the task target can also be referred to as workflows.

[0069] FIG. 1 shows a schematic diagram of an implementation of an operation and maintenance. As shown in FIG. 1, a developer first develops task templates one by one for operation and maintenance tasks in specific scenarios, and builds a neural network with the ability to select task templates and the ability to extract input parameters. The input parameters refer to parameters that need to be input into the task template when the task template is executed. For example, for a task template for performing connectivity detection between two network devices, the input parameters include the respective internet protocol (IP) addresses of the two network devices, such as IP address 1 and IP address 2. The developer configures the pre-developed task templates into the operation and maintenance system, and integrates the aforementioned neural network into the operation and maintenance system. When the operation and maintenance system receives a user input, such as a user intent expressed in the form of text, the operation and maintenance system uses the neural network to identify the user intent, and performs matching in the pre-configured task templates according to the identification result, so as to determine a task template that matches the user intent. The operation and maintenance system also extracts input parameters from the user intent. The operation and maintenance system executes the matched task template according to the extracted input parameters, obtains a result that matches the user intent, and thus completes an operation and maintenance task. In addition, the developer continues to collect user problem logs, discovers common operation and maintenance tasks used by users, and continues to develop task templates one by one for the discovered common operation and maintenance tasks, and configures the developed task templates into the operation and maintenance system for use. The above scheme requires the developer to obtain task templates according to specific development use cases, which is not only time-consuming and labor-intensive, but also difficult to meet the requirements of diversified operation and maintenance environments.

[0070] To solve the above problems, an embodiment of the present application provides an operation and maintenance method. In the method, after an operation and maintenance system identifies a user intent described in natural language, the operation and maintenance system can automatically disassemble a task target included in the user intent, so as to obtain task steps for implementing the task target, that is, the operation and maintenance system can generate a task template for implementing the task target according to the task target. The operation and maintenance system executes the task template, and thus completes an operation and maintenance task corresponding to the user intent. Through the method provided by the embodiment of the present application, even if there is no task template matching the user intent pre-stored in the operation and maintenance system, the operation and maintenance system can also automatically generate a task template matching the user intent, and execute the task template, so as to complete the operation and maintenance task. It can be seen that the scheme provided by the embodiment of the present application does not require the developer to develop task templates one by one for different operation and maintenance tasks, and thus is more intelligent and has better user experience.

[0071] As an example, the operation and maintenance task includes but is not limited to domain name system (DNS) validity detection, connectivity detection of IP addresses, and the like.

[0072] Referring to FIG. 2, FIG. 2 shows a schematic diagram of a network scenario of the method provided by an embodiment of the present application. As shown in FIG. 2, the network scenario includes an operation and maintenance system and a network device.

[0073] The operation and maintenance system is configured to perform operation and maintenance on the network device, including but not limited to controlling the network device and analyzing possible problems of the network device. For example, the operation and maintenance system is configured to control the on and off of the network device. For another example, the operation and maintenance system receives an abnormal alarm reported by the network device, and analyzes possible problems of the network device based on the abnormal alarm. By executing the operation and maintenance method described below in the embodiments of the present application, the operation and maintenance system no longer relies on the task templates developed by the developers for operation and maintenance tasks, thereby enabling more intelligent operation and maintenance on the network device.

[0074] The network device is, for example, a network device such as a switch, a router, a firewall, an access point (AP), or a server in a user network, without limitation. The user network includes, but is not limited to, a campus network, a school network, or an enterprise network.

[0075] In the embodiments of the present application, the operation and maintenance system includes an operation and maintenance device and an AI agent core. The operation and maintenance device is configured to interact with a user and execute a task template to perform operation and maintenance on the network device. The AI agent core is configured to identify a user intent and generate a task template for achieving a task target included in the user intent, as described in detail below in the method embodiments, which will not be repeated here. Optionally, the operation and maintenance device and the AI agent core can be integrated in one device, or can be deployed in different devices, without limitation in the embodiments of the present application.

[0076] In one example embodiment, referring to FIG. 3 in combination with FIG. 2, the operation and maintenance system is deployed in the user network and configured to perform operation and maintenance on the network device in the user network.

[0077] In another example embodiment, referring to FIG. 4 in combination with FIG. 2, the operation and maintenance device in the operation and maintenance system is deployed in the user network, and the AI agent core in the operation and maintenance system is deployed in a server outside the user network (e.g., in the cloud). The server can be implemented as a single server or as a server cluster, without limitation. Thus, the operation and maintenance device invokes the AI agent core through the network to implement the method described in the embodiments of the present application, thereby achieving intelligent operation and maintenance on the network device in the user network.

[0078] In yet another example embodiment, referring to FIG. 5 in combination with FIG. 2, the operation and maintenance system is deployed in a server outside the user network (e.g., in the cloud). In this case, a client of the operation and maintenance system is deployed in the user network. In this way, the user controls the operation and maintenance system to perform intelligent operation and maintenance on the network device in the user network through the client of the operation and maintenance system.

[0079] It should be understood that the above is an exemplary description of the network scenario of the method provided by the embodiments of the present application, and does not constitute a limitation on the network scenario of the method. Those skilled in the art can know that as the business needs change, the network scenario can be adjusted according to the application needs, and the embodiments of the present application do not enumerate them one by one.

[0080] The embodiments of the present application provide an operation and maintenance system. The operation and maintenance system is used to execute the operation and maintenance method provided by the embodiments of the present application. On the one hand, the operation and maintenance system can generate a task template matched with the user's intention automatically, and execute the task template to complete the operation and maintenance task. Therefore, the operation and maintenance system provided by the embodiments of the present application no longer depends on the task templates developed by the developers one by one for different operation and maintenance tasks, and is more intelligent, and the user experience is better. On the other hand, the operation and maintenance system can extract execution parameters for triggering the execution of the task template from the user's intention, so that the execution of the task template can be triggered automatically according to the indication of the execution parameters, and the operation and maintenance in a true sense of unattended is realized, and therefore the user experience is good. The foregoing technical implementations can be referred to the description in the method embodiments below, and will not be repeated here.

[0081] For example, the operation and maintenance system is the operation and maintenance system described in the network scenario above, which includes an AI agent core and an operation and maintenance device.

[0082] Optionally, the operation and maintenance device can be implemented as a computing device, or as a functional module in a computing device, which is not limited. For example, the computing device can be a general-purpose computer, a notebook computer, a tablet computer, or the like. Of course, the computing device can also be implemented as a server, a controller, or a network management device, which is not limited. The AI agent core can be implemented by a computing device or a computer system including a GPU, an NPU, and / or a CPU, which will not be repeated here.

[0083] The operation and maintenance method provided by the embodiments of the present application will be described below with reference to the accompanying drawings.

[0084] Referring to FIG. 6, FIG. 6 shows a flowchart of an operation and maintenance method according to an embodiment of the present application. Optionally, the method can be executed by the operation and maintenance system of the implementation environment shown in FIGS. 2-5. As shown in FIG. 6, the method includes the following steps 101-103.

[0085] In step 101, the operation and maintenance system obtains a task request. The task request is a user intention described in natural language. The user intention includes a task parameter and a task target.

[0086] In the embodiments of the present application, the user intention represents the actual operation and maintenance requirement of the user, and corresponds to the operation and maintenance task that the user actually wants to implement. Since the task request is a user intention described in natural language, the "user intention includes task parameters and task targets" can also be understood as: the task request obtained by the operation and maintenance system includes task parameters and task targets.

[0087] The task target refers to the operation and maintenance task that the user actually wants to implement. When implementing the task target, the operation and maintenance system generally implements it by executing a task template matched with the task target. The task template matched with the task target is a collection of task steps for implementing the task target. The process of determining the task template matched with the task target is described below in step 1022, which is not repeated here.

[0088] The task parameters refer to the parameters required for implementing the task target, or can be understood as the parameters required for executing the task template matched with the task target. For ease of description, in the embodiments of the present application, the task template matched with the task target described in step 101 is referred to as a target task template, that is, the task parameters are the parameters required for executing the target task template.

[0089] In a possible case, the task parameters include input parameters, which refer to the parameters required for inputting to the target task template when executing the target task template, so that the task steps in the target task template are successfully executed. For example, when the task target in the user intention is to detect the connectivity of two network devices, the input parameters include the IP addresses of the two network devices. For another example, when the task target in the user intention is DNS validity detection, the input parameters include the IP address of the DNS server.

[0090] In another possible implementation, the task parameter includes an execution parameter and an input parameter. The content of the input parameter is described above. The execution parameter refers to a parameter for triggering execution of the target task template. In one possible implementation, the execution parameter includes a start time for triggering execution of the target task template and / or a cycle duration for periodically triggering execution of the target task template. For example, the execution parameter can be: starting execution of the target task template at time t2, and / or periodically executing the target task template at an interval of duration T1, where T is a positive integer. Here, t2 and t1 can be the same time, which means that the operation and maintenance system starts to execute the target task template immediately after receiving the task request and obtaining the target task template. Alternatively, t2 is a time after t1, which means that the operation and maintenance system starts to execute the target task template after waiting for a time t2 after receiving the task request and obtaining the target task template. In another possible implementation, the execution parameter includes information of an event for triggering execution of the target task template. The information of the event includes, but is not limited to, a type of the event, an identifier (ID) of the event (such as an event name), and the like. Here, the event includes, but is not limited to, an online / offline event of a network device, an abnormal alarm reported by a network device, and the like.

[0091] In one example embodiment, the operation and maintenance system obtains the task request, including: the operation and maintenance system receiving a task request input by a user through a front end. The front end of the operation and maintenance system is configured to implement interaction between the operation and maintenance system and the user. For example, the front end of the operation and maintenance system can be an interface (such as a website interface or a client interface) presented to the user by the operation and maintenance system when providing operation and maintenance services to the user. The interface can be an LUI or a GUI, which is not limited in this regard.

[0092] For example, the front end of the operation and maintenance system is implemented as an LUI. In this case, the user can input a user intent described in natural language to the operation and maintenance system through an input interface (such as a microphone) of the operation and maintenance system capable of inputting voice, so as to achieve the purpose of inputting a task request to the operation and maintenance system. In response, the operation and maintenance system receives the task request input by the user.

[0093] For another example, the front end of the operation and maintenance system is implemented as a GUI. In this case, the user can input a user intent described in natural language to the operation and maintenance system through an input interface (such as a keyboard or a touch screen) of the operation and maintenance system capable of inputting text, so as to achieve the purpose of inputting a task request to the operation and maintenance system. In response, the operation and maintenance system receives the task request input by the user.

[0094] In step 102, the operation and maintenance system generates a target task template according to a task target.

[0095] As described above, the target task template is used to implement the task target in the task request obtained by the operation and maintenance system in step 101, and the target task template includes a task step for implementing the task target.

[0096] Specifically, the process of generating the target task template according to the task target by the operation and maintenance system can be implemented through steps 1021-1023 shown in FIG. 7.

[0097] In step 1021, the operation and maintenance system obtains the task target from the obtained task request.

[0098] Since the task request obtained by the operation and maintenance system is a user intention described in natural language, after obtaining the task request, the operation and maintenance system can identify the user intention indicated by the task request through the AI agent core, so as to extract the task target from the user intention.

[0099] As an example, after receiving the task request input by the user through the front end, the operation and maintenance system identifies the user intention indicated by the task request through the AI agent core, so as to extract the task target from the user intention.

[0100] In step 1022, the operation and maintenance system determines whether there is a target task template in the task template set according to the task target.

[0101] Optionally, in the embodiment of the present application, the operation and maintenance system is pre-stored with a task template set containing a plurality of task templates. The task templates in the task template set can be task templates developed by the developer for specific operation and maintenance tasks, or task templates generated by the operation and maintenance system according to the task target in the obtained task request in the history, which is not limited. In an example, the task template set can be stored in the operation and maintenance system in the form of a database or stored in a device accessible by the operation and maintenance system.

[0102] In this case, after obtaining the task target, the operation and maintenance system first queries the task template set according to the task target to determine whether there is a target task template in the task template set.

[0103] In an example embodiment, after the operation and maintenance system extracts the task target from the user intent through the AI agent core, the AI agent core further calls the LLM to determine whether there is a target task template capable of achieving the task target in the task template set through natural language matching. For example, the AI agent core calls the LLM to match and compare the extracted task target with the description information of each task template in the task template set, so as to determine whether there is a target task template capable of achieving the task target in the task template set. The description information of the task template may, for example, be the name, use and / or function of the task template.

[0104] When the AI agent core determines that there is a target task template in the task template set, the operation and maintenance system retrieves the data of the target task template from the task template set according to the identifier of the target task template. The identifier of the target task template includes, but is not limited to, the name, ID, etc. of the target task template.

[0105] When the AI agent core determines that there is no target task template in the task template set, the operation and maintenance system performs step 1023.

[0106] Step 1023, the operation and maintenance system generates a target task template according to the task target.

[0107] In the embodiments of the present application, the operation and maintenance system generates a target task template through the AI agent core. Specifically, the AI agent core can input the task target into the LLM, so as to output a task step set for achieving the task target from the LLM, that is, obtain the target task template. The LLM can arrange a task step set for achieving the task target through learning and reasoning of the task target and the application programming interface (API) calling capability supported by the operation and maintenance system. For example, the task steps include, but are not limited to: API interface calling, branch of logical judgment of API interface return value, different API interface calling according to different branches, branch of logical judgment of API interface return value of each branch, etc.

[0108] In some embodiments, when generating the target task template according to the task target, the AI agent core of the operation and maintenance system sets a post task for the task target in addition to arranging the task steps for achieving the task target. For example, when the task target is to detect the connectivity of the network device, the target task template generated by the operation and maintenance system according to the task target includes, in addition to the task steps for achieving the detection of the connectivity of the network device, the task steps for achieving the positioning of the link fault when the connectivity of the network device is abnormal. The task achieved by the task steps for achieving the positioning of the link fault is the post task of the operation task "detecting the connectivity of the network device".

[0109] Optionally, the task steps for achieving the post task can correspond to one of the task templates in the task template set, denoted as a post task template. In this case, in addition to including the task steps for achieving the task target in the task request, the target task template also includes the task steps of the post task template when the task result matching the task request meets the preset condition. Of course, the task steps for achieving the post task can also be the task steps newly arranged by the operation and maintenance system through the AI agent core for the task target of the post task, which is not limited.

[0110] It can be understood that in some examples, the operation and maintenance system can also not perform step 1022. In this case, after obtaining the task target through step 1021, the operation and maintenance system directly performs step 1023, that is, directly generates the target task template according to the task target by the AI agent core of the operation and maintenance system.

[0111] Optionally, the operation and maintenance system can also store the target task template. In one example, when the task template set pre-stored in the operation and maintenance system is implemented by a database, the operation and maintenance system adds the newly generated target task template to the database to achieve the persistent storage of the target task template.

[0112] Optionally, the operation and maintenance system can store the target task template according to the matching degree of the task result and the task request after executing the target task template to obtain the task result.

[0113] For example, when the task result can correctly respond to the task request initiated by the user, the operation and maintenance system stores the target task template. In one implementation, the operation and maintenance system can display the task result to the user through the front end after executing the target task template to obtain the task result, and store the target task template when receiving the response information input by the user indicating that the task result matches the task request. The detailed description of the operation and maintenance system executing the target task template to obtain the task result can be referred to the description of step 103 below, which is not repeated here.

[0114] For example, when the task result cannot correctly respond to the task request initiated by the user, the operation and maintenance system does not need to store the target task template. In an implementation manner, the operation and maintenance system displays the task result obtained by executing the target task template to the user through the front end, and deletes the target task template or invalidates the target storage template when receiving the response information input by the user indicating that the task result does not match the task request. In this case, the operation and maintenance system can execute the method provided in the embodiments of the present application again according to the task request initiated by the user, to generate a new target task template for the task target in the task request, execute the new target task template according to the task parameters, and determine whether to store the new target task template, which will not be repeated here.

[0115] For the target task template that can obtain a task result matching the task request, by persistently storing the target task template, when the user initiates the task request described in step 101 again subsequently, the operation and maintenance system can directly match the target task template that has been persistently stored. In this way, the operation and maintenance system does not need to execute the process of generating the target task template again, so that the resource consumption of the operation and maintenance system can be saved, the speed of responding to the task request initiated by the user can be improved, and thus the operation and maintenance efficiency of the operation and maintenance system can be improved.

[0116] Step 103: The operation and maintenance system executes the task steps included in the target task template according to the task parameters, to obtain a task result matching the task request.

[0117] It should be understood that after the operation and maintenance system obtains the task request in step 101, in addition to extracting the task target from the user intent corresponding to the task request through the process described in step 1021, the operation and maintenance system also needs to extract the task parameters from the user intent corresponding to the task request. The process of extracting the task parameters from the user intent corresponding to the task request by the operation and maintenance system can refer to the process of extracting the task target from the user intent corresponding to the task request by the operation and maintenance system described in step 1021, which will not be repeated here.

[0118] In the embodiments of the present application, in addition to extracting the input parameters in the task parameters from the user intent, the execution parameters for triggering the execution of the target task template are also extracted, so that the operation and maintenance system can automatically trigger the execution of the target task template according to the execution parameters subsequently, thereby realizing truly unattended operation and maintenance.

[0119] Optionally, after extracting the task parameter from the user intent corresponding to the task request, the operation and maintenance system can perform parameter verification on the task parameter. For example, the operation and maintenance system verifies whether the operation and maintenance personnel who inputs the current task request has the operation and maintenance permission of the network device involved in the task parameter, and the like. It should be understood that the operation and maintenance system can execute the target task template according to the task parameter only in the case that the task parameter verification is successful, so as to complete the operation and maintenance task. Here, the detailed description of the operation and maintenance system verifying the task parameter can refer to related technologies, and will not be described here.

[0120] In the case that the task parameter verification is successful, in the first possible implementation manner, the operation and maintenance system directly executes the task steps included in the target task template obtained in step 102 according to the task parameter extracted above, so as to obtain the task result matched with the task request obtained in step 101.

[0121] In the second possible implementation manner, the operation and maintenance system can submit the task parameter extracted above and the target task template obtained in step 102 to the user for confirmation, and execute the task steps included in the target task template confirmed by the user according to the task parameter confirmed by the user. That is, the embodiment of the present application can provide the function of interaction between the operation and maintenance system and the user in the process that the operation and maintenance system responds to the task request initiated by the user, so that the user can confirm whether the target task template generated by the operation and maintenance system is reasonable, and / or confirm whether the task parameter is extracted accurately, and / or modify the target task template and / or the task parameter according to the latest demand of the user, and the like, so as to improve the user experience.

[0122] Referring to FIG. 8, the implementation process of the second possible implementation manner can be implemented through steps 1031-1034 shown in FIG. 8.

[0123] In step 1031, the operation and maintenance system outputs first task information, and the first task information includes the target task template and the task parameter extracted from the user intent.

[0124] After obtaining the target task template and extracting the task parameter from the user intent, the operation and maintenance system records the target task template and the task parameter as the first task information, and outputs the first task information to the user through the front end of the operation and maintenance system. In this way, in the front end interface of the operation and maintenance system, the user can intuitively understand the task steps required to be executed by the operation and maintenance system in response to the task request input by the user and the corresponding task parameter (i.e., the first task information). In this way, the user can confirm whether the task steps and the task parameter in the first task information meet the demand of the user according to the first task information displayed on the front end interface of the operation and maintenance system.

[0125] When the user determines that the first task information displayed on the front-end interface of the operation and maintenance system meets the user's own needs, the user can perform a confirmation operation on the front-end interface of the operation and maintenance system to respond to the first task information output by the operation and maintenance system. In an example, the confirmation operation includes but is not limited to an operation (such as a click operation performed by a finger or a mouse, etc.) on a "confirmation" control on the interface of the operation and maintenance system front-end interface where the first task information is displayed.

[0126] When the user determines that the first task information displayed on the front-end interface of the operation and maintenance system does not meet the user's own needs, the user can perform a modification operation on the front-end interface of the operation and maintenance system to modify the task steps and / or task parameters in the first task information displayed on the front-end interface of the operation and maintenance system. For example, the user can operate a "modification" control (such as a click operation performed by a finger or a mouse, etc.) on the front-end interface of the operation and maintenance system to achieve the purpose of modifying the task steps and / or task parameters in the first task information displayed on the front-end interface of the operation and maintenance system. In an example, for a task request whose task goal is to detect network device connectivity and whose task parameters include the IP addresses (IP address 1 and IP address 2, respectively) of two network devices, the first task information displayed on the front-end interface of the operation and maintenance system includes the task step of detecting network device connectivity and the task parameters [IP address 1, IP address 2]. If the user's current latest needs are to detect the connectivity of the network devices to which IP address 1 and IP address 3 belong, the user can perform a modification operation on the front-end interface of the operation and maintenance system to modify the task parameters displayed on the front-end interface of the operation and maintenance system to [IP address 1, IP address 3]. In another example, the user can also choose not to execute a part of the task steps displayed on the front-end interface of the operation and maintenance system, and so on.

[0127] For the second task information obtained by the user performing a modification operation on the front-end interface of the operation and maintenance system, the user performs a confirmation modification operation on the front-end interface of the operation and maintenance system to respond to the first task information output by the operation and maintenance system. The second task information is the task information obtained by the user modifying the first task information. In an example, the confirmation modification operation includes but is not limited to an operation (such as a click operation performed by a finger or a mouse, etc.) on a "confirmation" or "confirmation modification" control on the interface of the operation and maintenance system front-end where the second task information is displayed.

[0128] Step 1032, the operation and maintenance system receives response information of the first task information.

[0129] In response to the confirmation operation or the confirmation modification operation performed by the user on the front-end interface of the operation and maintenance system in step 1301, the operation and maintenance system receives the response information of the first task information input by the user to the operation and maintenance system.

[0130] Step 1033, when the response information of the first task indicates that the user confirms that the first task information is correct, the operation and maintenance system executes the task steps included in the target task template according to the task parameters extracted from the user intent.

[0131] Among them, the task parameters extracted from the user intent are the task parameters in the first task information.

[0132] It should be understood that when the response information of the first task information is the response information input by the user through the execution confirmation operation, indicating that the user confirms that the first task information is correct. At this time, the operation and maintenance system executes the target task template in the first task information according to the task parameters in the first task information.

[0133] In one example, when the execution parameter in the task parameter indicates executing the target task template immediately, the operation and maintenance system starts executing the task steps included in the target task template based on the input parameters in the task parameter immediately after receiving the response information of the first task information.

[0134] In another example, when the execution parameter in the task parameter indicates executing the target task template at time t2, the operation and maintenance system starts executing the task steps included in the target task template based on the input parameters in the task parameter after waiting for time t2 to arrive after receiving the response information of the first task information.

[0135] In yet another example, when the execution parameter in the task parameter indicates executing the target task template periodically with an interval of time length T1 from the moment or time t2, the operation and maintenance system starts executing the task steps included in the target task template based on the input parameters in the task parameter periodically with an interval of time length T1 after receiving the response information of the first task information or waiting for time t2 to arrive.

[0136] The following is an example of a task request in which the task target is to detect network device connectivity, the input parameters in the task parameters include the IP addresses (IP address 1 and IP address 2, respectively) of two network devices, and the target task template generated based on the task target is template 1. In this case, it is assumed that the task steps in template 1 include: obtaining the detection result of the connectivity between the operation and maintenance system and the network device; analyzing the detection result; and when the detection result indicates that the connectivity between the network devices is abnormal, executing the post-task of link fault localization. In this case, the operation and maintenance system sends connectivity detection instructions (such as ping instructions) to network device 1 to which IP address 1 belongs and network device 2 to which IP address 2 belongs, respectively, so as to obtain detection result 1 returned by network device 1, which indicates whether the operation and maintenance system and network device 1 are connected, and obtain detection result 2 returned by network device 2, which indicates whether the operation and maintenance system and network device 2 are connected. The operation and maintenance system analyzes and determines whether network device 1 and network device 2 are connected according to detection result 1 and detection result 2. When it is determined that network device 1 and network device 2 are not connected, the task steps required to implement the post-task "link fault localization" are executed to implement the link fault localization between network device 1 and network device 2, which will not be described again.

[0137] In step 1034, when the response information of the first task carries the second task information, the operation and maintenance system executes the task steps recorded in the second task information according to the task parameters contained in the second task information.

[0138] It can be understood that when the response information of the first task information is the response information input by the user to the operation and maintenance system by executing the confirmation modification operation after modifying the first task information to obtain the second task information, at this time, the response information of the first task carries the second task information. In this case, the operation and maintenance system executes the task steps recorded in the second task information according to the task parameters contained in the second task information. The detailed description of the operation and maintenance system executing the task steps recorded in the second task information according to the task parameters contained in the second task information can be referred to the description of the operation and maintenance system executing the target task template in the first task information according to the task parameters in the first task information in step 1033, which will not be described again.

[0139] Optionally, in the embodiments of the present application, logs are recorded in the process from obtaining the task request to executing the target task template to obtain the task result, so as to record the process of the operation and maintenance system responding to the task request input by the user.

[0140] Optionally, after the operation and maintenance system executes the task result of the target task template, the operation and maintenance system can show the task result to the user through the front end, so that the user determines whether the task result matches the task request initiated by the user. Optionally, when the operation and maintenance system determines that the task result obtained by executing the target task matches the task request initiated by the user, the task request and the task result matching the task request can be recorded for use as reference information for subsequent operation and maintenance tasks.

[0141] In summary, by the method described in steps 101 to 103, the operation and maintenance system can automatically orchestrate a target task template for achieving the task target in the user intent described in natural language by the user, so that the developer does not need to develop task templates one by one for different operation and maintenance tasks, which can save a lot of human cost. Moreover, by persistently storing the target task template obtained by the automatic orchestration task step, when the user inputs the same user intent again, the operation and maintenance system can only need to call and execute the stored task template to achieve the task target in the user intent, which can improve the response time of the operation and maintenance system and improve the user experience.

[0142] In addition, the scheme provided by the embodiments of the present application can directly extract the execution parameters of the task template from the user intent, so that the time, period duration or event information for triggering the execution of the task template can be quickly configured, which can improve the automation degree of the operation and maintenance system and realize truly unattended operation and maintenance, thereby improving the user experience.

[0143] In some other embodiments, during the operation of the operation and maintenance system, the operation and maintenance system can receive a large number of abnormal alarms reported by network devices, including but not limited to device fault alarms, link abnormal alarms, network system clock abnormal alarms and the like of the network devices. In practice, there are many meaningless alarms (referred to as invalid alarms) in the abnormal alarms received by the operation and maintenance system. For example, when the network cable connecting two network devices is disconnected, the two devices will continuously report link abnormal alarms to the operation and maintenance system, and all the alarms except the first abnormal alarm are repeated invalid alarms. Based on this, the embodiments of the present application further provide an operation and maintenance method, which can filter the abnormal alarms received by the operation and maintenance system to filter out invalid alarms. By this method, not only the storage resource consumption of the operation and maintenance system can be saved, but also the effective alarms can be prevented from being submerged by invalid alarms so as to be unable to be processed in time, and the human cost and time cost consumed when checking and processing the abnormal alarms one by one can be saved.

[0144] Referring to FIG. 9, FIG. 9 shows a flowchart of another operation and maintenance method provided by the embodiments of the present application. Optionally, the method can be executed by the operation and maintenance system of the implementation environment shown in FIGS. 2 to 5. As shown in FIG. 9, the method includes the following steps 201 to 203.

[0145] Step 201, the operation and maintenance system receives an abnormal alarm reported by a network device.

[0146] Optionally, the operation and maintenance system can receive the abnormal alarm sent by the network device through its own communication interface. It can be understood that the type and content of the abnormal alarm reported by the network device are not limited in the embodiments of the present application.

[0147] Optionally, the operation and maintenance system performs a deduplication operation on the current received abnormal alarm (denoted as the current abnormal alarm).

[0148] In an exemplary implementation, the operation and maintenance system can query the historical received abnormal alarms according to the current abnormal alarm to determine whether the current abnormal alarm is a repeated invalid alarm. Wherein, the "historical received abnormal alarm" can be the abnormal alarm received by the operation and maintenance system in a period with the end time being the current time and the time length being T2, or the "historical received abnormal alarm" is all the abnormal alarms received by the operation and maintenance system before the current time, which is not limited. In addition, the specific value of the time length T2 is not limited in the embodiments of the present application. Further, when the operation and maintenance system queries that the current abnormal alarm exists in the "historical received abnormal alarm", it is determined that the current abnormal alarm is a repeated invalid alarm. At this time, the operation and maintenance system can set the current abnormal alarm as invalid. When the operation and maintenance system queries that the current abnormal alarm does not exist in the "historical received abnormal alarm", it is determined that the current abnormal alarm is not a repeated abnormal alarm, in which case, the operation and maintenance system performs step 202 on the current abnormal alarm.

[0149] Step 202, the operation and maintenance system determines a target de-noising rule according to the abnormal alarm and context information of the abnormal alarm.

[0150] The context information of the abnormal alarm includes a historical alarm of the same type as the abnormal alarm and a de-noising rule determined based on the historical alarm.

[0151] Wherein, the historical alarm is an abnormal alarm received by the operation and maintenance system before receiving the abnormal alarm (i.e. the current abnormal alarm) in step 201.

[0152] The de-noising rule refers to a strategy / rule for filtering invalid alarms in the abnormal alarms received by the operation and maintenance system. For example, the de-noising rule is to filter the abnormal alarms reported by the network device 3 in the non-working time. For another example, the de-noising rule is to filter the abnormal alarms reported by the network device 4 to indicate network link failure.

[0153] The de-noising rule determined based on the historical alarm is the de-noising rule determined after step 202 is performed on the historical alarm.

[0154] In a first possible implementation, the historical alarms and the noise elimination rules determined based on the historical alarms are recorded in the operation and maintenance system. Therefore, in step 202, the operation and maintenance system determines the target noise elimination rule according to the abnormal alarm and the context information of the abnormal alarm, including: the operation and maintenance system performs inference learning on the current abnormal alarm, at least one historical alarm of the same type as the abnormal alarm, and the noise elimination rule determined based on the at least one historical alarm through the AI agent core, to obtain the target noise elimination rule.

[0155] It can be understood that in this embodiment, when the AI agent core performs inference learning on the current abnormal alarm, at least one historical alarm of the same type as the abnormal alarm, and the noise elimination rule determined based on the at least one historical alarm, there can also be a case where the target noise elimination rule cannot be inferred and learned. At this time, optionally, when the AI agent core performs inference learning on the current abnormal alarm, at least one historical alarm of the same type as the abnormal alarm, and the noise elimination rule determined based on the at least one historical alarm, and the target noise elimination rule cannot be obtained, the AI agent core stops the process of determining the target noise elimination rule based on the current abnormal alarm. Optionally, after the AI agent core performs inference learning on the current abnormal alarm, at least one historical alarm of the same type as the abnormal alarm, and the noise elimination rule determined based on the at least one historical alarm, the AI agent core outputs indication information that the target noise elimination rule cannot be obtained at present.

[0156] In a second possible implementation, in addition to the historical alarms and the noise elimination rules determined based on the historical alarms, the operation and maintenance system is also pre-provisioned with a plurality of preset noise elimination rules. The preset noise elimination rules can be the target noise elimination rules stored after steps 201-203 are performed, or can be noise elimination rules developed by the developer based on expert experience, which are not limited. For example, when the developer finds that the network device 5 will stop running the service 1 during non-working hours, the developer can develop a noise elimination rule: filtering the abnormal alarms related to the service 1 reported by the network device 5 during non-working hours.

[0157] In this case, in step 202, the operation and maintenance system determines a target noise elimination rule according to the abnormal alarm and context information of the abnormal alarm, including: the operation and maintenance system performs inference learning on the current abnormal alarm, at least one historical alarm of the same type as the current abnormal alarm, a noise elimination rule determined based on the at least one historical alarm, and at least one preset noise elimination rule through an AI agent core, to obtain the target noise elimination rule. The at least one preset noise elimination rule is a preset noise elimination rule associated with the current abnormal alarm. For example, the type of invalid alarm that can be filtered by the at least one preset noise elimination rule is the same as the type of the current abnormal alarm. For another example, the occurrence time of the invalid alarm that can be filtered by the at least one preset noise elimination rule and the time when the current abnormal alarm occurs belong to the same time period, such as a non-working time period. For another example, the types of the first network device and the second network device are the same, and the invalid alarm filtered by the at least one preset noise elimination rule is reported by the first network device, and the current abnormal alarm is reported by the second network device. And the like, not limited thereto.

[0158] Similarly to the first possible implementation, in the second possible implementation, there is also a case where the operation and maintenance system cannot infer and learn the target noise elimination rule, which will not be described in detail.

[0159] In step 203, the operation and maintenance system filters invalid alarms in the received abnormal alarms according to the target noise elimination rule.

[0160] After the operation and maintenance system determines the target noise elimination rule, the operation and maintenance system filters the abnormal alarms received by the operation and maintenance system before the current abnormal alarm based on the indication of the target noise elimination rule, to filter out invalid alarms therein.

[0161] Optionally, after the operation and maintenance system filters invalid alarms in the received abnormal alarms according to the target noise elimination rule, the operation and maintenance system can delete the target noise elimination rule.

[0162] Optionally, after the operation and maintenance system filters invalid alarms in the received abnormal alarms according to the target noise elimination rule, the operation and maintenance system can determine whether to store the target noise elimination rule according to the number of filtered invalid alarms, so as to store the target noise elimination rule as the preset noise elimination rule. In one example, when the operation and maintenance system determines that the number of filtered invalid alarms exceeds a threshold, the operation and maintenance system stores the target noise elimination rule and takes the target noise elimination rule as the preset noise elimination rule. The present application embodiment does not make specific limitation on the value of the threshold.

[0163] Optionally, the operation and maintenance system can also display the target noise elimination rule to the user through the front end.

[0164] In this way, by the method described in steps 201-203, the target de-noising rule for filtering invalid alarms in the received abnormal alarms is determined based on the currently received abnormal alarm, and the purpose of filtering invalid alarms according to the target de-noising rule is achieved. By this method, not only the storage resources required by the operation and maintenance system for storing abnormal alarms can be saved, but also the effective alarms can be prevented from being overwhelmed by invalid alarms so as to be unable to handle the effective alarms in time, and the human cost and time cost consumed when checking and processing all abnormal alarms one by one can be saved, thereby improving the user experience.

[0165] To further understand the scheme described in the embodiments of the present application, the method described in the embodiments of the present application is further described below in conjunction with examples.

[0166] Referring to FIG. 10, FIG. 10 exemplarily shows a structural schematic diagram of an operation and maintenance system provided by the embodiments of the present application. As shown in FIG. 10, the operation and maintenance system 100 includes an operation and maintenance device and an AI agent core. The operation and maintenance device includes a task management module, a task scheduling module, a front end, and a management and control module. The AI agent core includes an intention recognition module, a task arrangement module, and an alarm analysis module. The functions of each module can be referred to the description of the method in FIGS. 11-13.

[0167] In conjunction with FIG. 10, referring to FIG. 11, FIG. 11 shows another flowchart of the operation and maintenance method provided by the embodiments of the present application. As shown in FIG. 11, the method 110 can include S1-S11.

[0168] S1, the task management module receives a task request input by a user in natural language through the front end.

[0169] The detailed description of S1 can be referred to the related description in step 101, and will not be repeated here.

[0170] S2, the task management module transmits the task request to the intention recognition module of the AI agent core.

[0171] S3, the intention recognition module identifies the task request described in natural language, and extracts the task target and the task parameter therefrom.

[0172] The detailed description of the intention recognition module extracting the task target and the task parameter can be referred to the related description in steps 1021 and 103, and will not be repeated here.

[0173] S4, the intention recognition module queries whether there is a target task template in the task template set according to the task target.

[0174] When the intent recognition module determines that the target task template exists in the task template set, the identifier of the target task template and the task parameter are sent to the task management module, and step S5 is performed by the task management module. The identifier of the target task template includes but is not limited to the name, ID, etc. of the target task template.

[0175] When the intent recognition module determines that the target task template does not exist in the task template set, the task parameter and the task target are sent to the task orchestration module of the AI agent core, and step S6 is performed by the task orchestration module.

[0176] S5, the task management module retrieves the data of the target task template from the task template set according to the identifier of the target task template.

[0177] Subsequently, the task management module performs S7.

[0178] S6, the task orchestration module generates a target task template according to the task target, and sends the target task template and the task parameter to the task management module.

[0179] The detailed description of how the task orchestration module generates the target task template according to the task target can be referred to the description of step 1023, and will not be repeated.

[0180] Optionally, the task parameter can also be sent to the task management module by the intent recognition module in S4, which is not limited.

[0181] Subsequently, the task management module records the first task information based on the received task parameter and the target task template, and performs S7.

[0182] S7, the task management module records the first task information including the task parameter and the target task template.

[0183] S8, the task management module outputs the first task information to the user through the front end.

[0184] The detailed description of S8 can be referred to the description of step 1031, and will not be repeated.

[0185] S9, the task management module receives the response information of the first task information input by the user through the front end.

[0186] The detailed description of S9 can be referred to the description of step 1032, and will not be repeated.

[0187] S10, when the response information of the first task indicates that the user confirms that the first task information is correct, the task management module determines to execute the target task template according to the task parameter extracted from the user intent.

[0188] S11, when the response information of the first task carries the second task information, the task management module determines to execute the task steps recorded in the second task information according to the task parameters contained in the second task information.

[0189] The detailed description of S10-S11 can refer to the description of steps 1033-1034, and will not be repeated here.

[0190] In some example embodiments, the process of executing the target task template according to the task parameters extracted from the user intention by the operation and maintenance device can be implemented by the method 120 shown in FIG. 12. That is, the process of executing the target task template in the first task information according to the task parameters in the first task information by the operation and maintenance device can be implemented by the method 120 shown in FIG. 12. As shown in FIG. 12, the method 120 includes S12-S14.

[0191] S12, the task scheduling module obtains the target task template and the corresponding input parameters from the first task information maintained by the task management module.

[0192] Specifically, the task scheduling module can periodically access the task information maintained by the task management module. When the execution parameters in a certain task information, such as the first task information, meet the requirements, the task scheduling module obtains the target task template and the corresponding input parameters from the first task information and performs S12, so as to realize the purpose of starting the operation and maintenance task corresponding to the target task template.

[0193] S13, the task scheduling module sends the target task template and the corresponding input parameters to the control analysis module.

[0194] S14, the control analysis module executes the target task template according to the input parameters.

[0195] It should be understood that in S14, the process of executing the target task template according to the input parameters by the control analysis module can interact with the network device to obtain the information required by the task steps in the target task template, which is not specifically limited.

[0196] Optionally, after the control analysis module executes the target task template and obtains the task result, the control analysis module can also show the task result to the user through the front end.

[0197] Optionally, the control analysis module can store the task result and the task request for obtaining the task result.

[0198] In still some example embodiments, in combination with FIG. 10, referring to FIG. 13, FIG. 13 shows a flow diagram of another operation and maintenance method provided by the embodiments of the present application. As shown in FIG. 13, the method 130 includes S15-S19.

[0199] S15, the control analysis module receives the abnormal alarm reported by the network device, and performs a deduplication operation on the currently received abnormal alarm.

[0200] The detailed description of S15 can refer to the description of step 201, and will not be repeated.

[0201] S16, the control analysis module determines that the currently received abnormal alarm is not a repeated invalid alarm, and transmits the abnormal alarm to the alarm analysis module of the AI agent core.

[0202] S17, the alarm analysis module determines the target noise elimination rule according to the received abnormal alarm and the context information of the abnormal alarm.

[0203] The detailed description of S17 can refer to the description of step 202, and will not be repeated.

[0204] S18, the alarm analysis module sends the determined target noise elimination rule to the control analysis module.

[0205] S19, the control analysis module filters the invalid alarms in the received abnormal alarms according to the received target noise elimination rule.

[0206] The detailed description of S19 can refer to the description of step 203, and will not be repeated.

[0207] Optionally, the control analysis module can also show the target noise elimination rule to the user through the front end.

[0208] It should be noted that the various modules shown in FIG. 10 and the method steps executed by the various modules in FIGS. 11-13 are only exemplary and do not constitute a limitation on the scope of protection of the scheme described in the embodiments of the present application.

[0209] In order to realize the functions described in the above method embodiment, as shown in FIG. 14, FIG. 14 shows another structure diagram of the operation and maintenance system provided by the embodiments of the present application. The operation and maintenance system 1400 is used to execute the operation and maintenance method described above, for example, to execute the method shown in FIGS. 6-9 or FIGS. 11-13. The operation and maintenance system 1400 can include an acquisition unit 1401, a generation unit 1402, and an execution unit 1403.

[0210] The acquisition unit 1401 is configured to acquire a task request, the task request being a user intent described in natural language, the user intent including a task parameter and a task target. The generation unit 1402 is configured to generate a target task template according to the task target, the target task template including a task step for realizing the task target. The execution unit 1403 is configured to execute the task step included in the target task template according to the task parameter, to obtain a task result matched with the task request.

[0211] As an example, in combination with FIG. 6, the acquisition unit 1401 can be configured to perform step 101, the generation unit 1402 can be configured to perform step 102, and the execution unit 1403 can be configured to perform step 103.

[0212] Optionally, the operation and maintenance system 1400 further includes a judgment unit 1404 configured to judge whether there is a target task template in the task template set according to the task target. The generation unit 1402 is specifically configured to generate the target task template according to the task target in a case where there is no target task template in the task template set.

[0213] As an example, in combination with FIG. 7, the judgment unit 1404 can be configured to perform step 1022, and the generation unit 1402 can be configured to perform step 1023.

[0214] Optionally, the generation unit 1402 is specifically configured to input the task target into the LLM to obtain the target task template output by the LLM.

[0215] Optionally, the operation and maintenance system 1400 further includes a receiving unit 1405 configured to receive an abnormal alarm reported by a network device, a determination unit 1406 configured to determine a target de-noising rule according to the abnormal alarm and context information of the abnormal alarm, and a filtering unit 1407 configured to filter invalid alarms in the received abnormal alarm according to the target de-noising rule. The context information includes historical alarms of the same type as the abnormal alarm and a de-noising rule determined based on the historical alarms.

[0216] As an example, in combination with FIG. 9, the receiving unit 1405 can be configured to perform step 201, the determination unit 1406 can be configured to perform step 202, and the filtering unit 1407 can be configured to perform step 203.

[0217] Optionally, the operation and maintenance system 1400 further includes an output unit 1408 configured to output first task information, the first task information including the target task template and the task parameter, a receiving unit 1405 configured to receive response information of the first task information, and an execution unit 1403 configured to execute the task steps included in the target task template according to the task parameter in a case where the response information indicates that the first task information is correct.

[0218] As an example, in combination with FIG. 8, the output unit 1408 can be configured to perform step 1031, the receiving unit 1405 can be configured to perform step 1032, and the execution unit 1403 can be configured to perform step 1033.

[0219] Optionally, the execution unit 1403 is further configured to execute the task step recorded in the second task information according to a task parameter contained in the second task information, in response to the response information carrying the second task information. The second task information is task information obtained by modifying the first task information.

[0220] As an example, in combination with FIG. 8, the execution unit 1403 can be configured to execute step 1034.

[0221] Optionally, the task parameter includes an input parameter. Alternatively, the task parameter includes an input parameter and an execution parameter. The input parameter is a parameter input into the target task template when the target task template is executed, and the execution parameter is a parameter triggering execution of the target task template.

[0222] Optionally, the execution parameter includes a start time triggering execution of the target task template and / or a cycle duration of periodically triggering execution of the target task template. Alternatively, the execution parameter includes a type of event triggering execution of the target task template and / or an ID of the event.

[0223] Optionally, the generation unit 1402, the judgment unit 1404, and the determination unit 1406 are implemented by an LLM-based AI agent core.

[0224] Optionally, the generation unit 1402, the judgment unit 1404, and the determination unit 1406 are integrated in one device (such as a device configured with the AI agent core shown in FIG. 10). Units other than the generation unit 1402, the judgment unit 1404, and the determination unit 1406 are integrated in another device (such as the operation and maintenance apparatus shown in FIG. 10).

[0225] For specific descriptions of the above optional manners, refer to the foregoing method embodiments, which will not be described here again. In addition, the explanations and beneficial effect descriptions of any one of the operation and maintenance systems 1400 provided above can refer to the corresponding method embodiments described above, which will not be described here again.

[0226] As an example, in combination with FIG. 10, the functions implemented by the acquisition unit 1401, the receiving unit 1405, and the output unit 1408 in the operation and maintenance system 1400 can be implemented by the front end in the operation and maintenance device shown in FIG. 10. The function implemented by the generation unit 1402 in the operation and maintenance system 1400 can be implemented by the task scheduling module in the AI agent core shown in FIG. 10. The function implemented by the execution unit 1403 in the operation and maintenance system 1400 can be implemented by the task scheduling module and the control analysis module in the operation and maintenance device shown in FIG. 10. The function implemented by the judgment unit 1404 in the operation and maintenance system 1400 can be implemented by the intention recognition module in the AI agent core shown in FIG. 10. The function implemented by the determination unit 1406 in the operation and maintenance system 1400 can be implemented by the alarm analysis module in the AI agent core shown in FIG. 10. The function implemented by the filtering unit 1407 in the operation and maintenance system 1400 can be implemented by the control analysis module in the operation and maintenance device shown in FIG. 10.

[0227] Those skilled in the art should easily understand that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is implemented in hardware or computer software driven hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0228] It should be noted that the division of modules / units in FIG. 10 or FIG. 14 is illustrative, and is only a logical functional division. When actually implemented, there can be another division manner. For example, two or more functions can be integrated in one processing module. The functions implemented by the integrated module can be implemented in the form of hardware or in the form of a software function module.

[0229] The embodiments of the present application provide a computing device for implementing part or all of the functions of the operation and maintenance device or the AI agent core in the operation and maintenance system provided by the embodiments of the present application.

[0230] FIG. 15 is a structural schematic diagram of a computing device provided in an embodiment of the present application. As shown in FIG. 15, the computing device 1500 includes a processor 1501, a memory 1502, a communication interface 1503, and a bus 1504. The processor 1501, the memory 1502, and the communication interface 1503 are communicatively connected to each other through the bus 1504. The computing device 1500 further includes an input / output interface 1505, which is communicatively connected to the processor 1501, the memory 1502, and the communication interface 1503 through the bus 1504.

[0231] The processor 1501 can include a general-purpose processor and / or a special-purpose hardware chip. The general-purpose processor can include a CPU, a microprocessor, or a GPU. The CPU is, for example, a single-CPU, or is, for example, a multi-CPU. The special-purpose hardware chip is a hardware module with high processing performance. The special-purpose hardware chip includes at least one of a digital signal processor (DSP), a data processing unit (DPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, an NPU, a tensor processing unit (TPU), an AI chip, or a network processer (NP). The processor 1501 can also be an integrated circuit chip with signal processing capability. In the implementation process, part or all of the functions of the method provided in the embodiments of the present application can be completed by the integrated logic circuit of the hardware in the processor 1501 or the instruction in the form of software.

[0232] The memory 1502 is configured to store computer programs, including an operating system 1502a and executable code (i.e., program instructions) 1502b. The memory 1502 is, for example, a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), a flash memory, or another type of static storage device that can store static information and instructions, or is, for example, a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous DRAM (SDRAM), a double data rate SDRAM (DDR SDRAM), an enhanced SDRAM (ESDRAM), a synchlink DRAM (SLDRAM), or another type of dynamic storage device that can store information and instructions, or is, for example, a read-only light disc or another optical disc storage, an optical disc storage (including a compact disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk storage medium or another magnetic storage device, or is any other medium that can be used to carry or store desired executable code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this. For example, the memory 1502 is configured to store a task template set. The memory 1502 is, for example, independent and connected to the processor 1501 through the bus 1504. Alternatively, the memory 1502 and the processor 1501 are integrated together. The memory 1502 can store executable code, and when the executable code stored in the memory 1502 is executed by the processor 1501, the processor 1501 is configured to perform part or all of the functions of the operation and maintenance method provided in the embodiments of the present application. For implementation of the processor 1501 to execute the process, please refer to the related description in the foregoing embodiments. The memory 1502 can further include software modules and data required by other running processes such as an operating system.

[0233] The communication interface 1503 uses a transceiving module such as, but not limited to, a transceiver, to implement communication with other devices or communication networks. For example, the communication interface 1503 can be any one or any combination of the following devices: a communication interface (such as an Ethernet interface), a wireless network card, and the like, which has a network access function. Among them, the communication interface 1503 includes a receiving unit for receiving data / packets, and a sending unit for sending data / packets.

[0234] The bus 1504 is any type of communication bus for interconnecting internal devices (e.g., the memory 1502, the processor 1501, the communication interface 1503) of the computing device 1500. For example, a system bus. The embodiments of the present application take the above-mentioned devices inside the computing device 1500 as an example to illustrate that the above-mentioned devices inside the computing device 1500 are interconnected by the bus 1504. Alternatively, the above-mentioned devices inside the computing device 1500 can also be connected to each other in communication by means other than the bus 1504, for example, the above-mentioned devices inside the computing device 1500 are interconnected by internal logic interfaces.

[0235] The input / output interface 1505 is used to realize the human-computer interaction between the user and the computing device 1500. For example, to realize the text interaction or voice interaction between the user and the computing device 1500, etc. Among them, the input / output interface 1505 includes an input interface for realizing the user inputting information to the computing device 1500, and includes an output interface for realizing the computing device 1500 outputting information to the user.

[0236] As an example, the input interface includes but is not limited to a touch screen, a keyboard, a mouse, or a microphone, etc., and the output interface includes but is not limited to a display screen, a loudspeaker, etc. Among them, the touch screen, the keyboard, or the mouse are used to input text / image information, the microphone is used to input voice information, the display screen is used to output text / image information, and the loudspeaker is used to output voice information.

[0237] It should be noted that the above-mentioned devices can be respectively arranged on independent chips, or at least part or all of them can be arranged on the same chip. Whether to arrange each device independently on different chips or to integrate them on one or more chips often depends on the needs of product design. The embodiments of the present application do not limit the specific implementation form of the above-mentioned devices. And the description of the corresponding flow of each of the above-mentioned figures has its own emphasis, and the part not described in detail in a certain flow can be referred to the related description of other flows.

[0238] In the above embodiments, the implementation can be wholly or partially by software, hardware, firmware or any combination thereof. When implemented by software, the implementation can be wholly or partially in the form of a computer program product. The computer program product providing a program development platform includes one or more computer instructions, which when loaded and executed on the computing device 1500, wholly or partially implement the functions of the operation and maintenance method provided by the embodiments.

[0239] Furthermore, the computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium stores computer program instructions providing a program development platform.

[0240] As an example, in combination with FIG. 14, the functions implemented by the acquisition unit 1401, the receiving unit 1405 and the output unit 1408 in the operation and maintenance system 1400 can be implemented by the input / output interface 1505 shown in FIG. 15. The functions implemented by the generation unit 1402, the execution unit 1403, the determination unit 1404, the determination unit 1406 and the filtering unit 1407 in the operation and maintenance system 1400 can be implemented by the processor 1501 executing the program code in the memory 1502.

[0241] The embodiments of the present application also provide a computer readable storage medium, which is a non-volatile computer readable storage medium, and includes computer program instructions. When the computer program instructions are executed by a processor, a device or a system including the processor, the processor, the device or the system executes the operation and maintenance method provided by the embodiments of the present application.

[0242] The embodiments of the present application also provide a computer program product including instructions, which when executed by a processor, a device or a system including the processor, cause the processor, the device or the system to implement the operation and maintenance method provided by the embodiments of the present application.

[0243] The computer system is a system with computing processing capability. The computer system generally includes a processor and a memory, and the processor is used to call and run the instructions stored in the memory, so that the computer system implements the operation and maintenance method described above. Optionally, the computer system can also include at least one of an input interface or an output interface. The processor, the memory, the input interface and the output interface of the computer system are connected through an internal connection path.

[0244] A person of ordinary skill in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by hardware, or can be instructed by a program to complete the related hardware, and the program can be stored in a computer readable storage medium, such as a read-only memory, a magnetic disk or an optical disk.

[0245] It should be noted that the information (including but not limited to user equipment information, user personal information, etc.), data (including but not limited to data for analysis, stored data, displayed data, etc.) and signals involved in the present application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0246] The chip provided in the embodiments of the present application also includes a processor for running program instructions or codes, and the chip or the device containing the chip can be used to execute the operation and maintenance method provided in the embodiments of the present application. For example, the chip further includes an input interface, an output interface and a memory. The input interface, the output interface, the processor and the memory of the chip are connected through the internal connection path of the chip, and the memory in the chip is used to store the program instructions or codes run by the processor, and the input interface and the output interface of the chip are used for the connection and communication of the chip with other chips or devices.

[0247] In the embodiments of the present application, the terms "first", "second" and "third" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance. The term "at least one" means one or more, and the term "multiple" means at least two, unless otherwise explicitly limited.

[0248] In the present application, the term "and / or" is only used to describe the association relationship of the associated objects, which means that there can be three kinds of relationships, for example, A and / or B, which can represent the existence of A alone, the existence of A and B together, and the existence of B alone. In addition, the character " / " in this paper generally represents that the front and rear associated objects are a "or" relationship.

[0249] It should be understood that the terms used in the description of various described examples herein are only for the purpose of describing specific examples, and are not intended to be limiting. As used in the description of various described examples and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0250] It should be understood that determining B according to A does not mean that B is determined only according to A, but B can also be determined according to A and / or other information.

[0251] It should be understood that the terms "comprise", "comprises", "comprising", "includes", "including", "comprises", "comprising", "comprises" and / or "comprising" when used in this specification specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0252] It should also be understood that in various embodiments of the present application, the size of the sequence number of various processes does not mean the order of execution, the execution order of various processes should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0253] The above is only an optional embodiment of the present application, and does not limit the present application, any modification, equivalent replacement, improvement, etc. within the concept and principle of the present application should be included in the protection scope of the present application.

[0254] Finally, it should be pointed out that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the protection scope of the technical solutions of the embodiments of the present application.

Claims

1. An operation and maintenance method, characterized by, The method comprises: obtaining a task request, the task request being a user intention described in natural language, the user intention comprising a task parameter and a task target; generating a target task template according to the task target, the target task template comprising a task step for achieving the task target; executing the task step comprised in the target task template according to the task parameter to obtain a task result matched with the task request.

2. The method of claim 1, wherein, The method further comprises: determining whether the target task template exists in a task template set according to the task target; generating the target task template according to the task target in a case where the target task template does not exist in the task template set.

3. The method of claim 1 or 2, wherein, The generating of the target task template according to the task target comprises: inputting the task target into a large language model (LLM) to obtain the target task template output by the LLM.

4. The method of any one of claims 1 to 3, wherein, The method further comprises: receiving an abnormal alarm reported by a network device; determining a target de-noising rule according to the abnormal alarm and context information of the abnormal alarm, wherein the context information comprises a historical alarm of the same type as the abnormal alarm and a de-noising rule determined based on the historical alarm; filtering invalid alarms in the received abnormal alarms according to the target de-noising rule.

5. The method of any one of claims 1 to 4, wherein, The method further comprises: outputting first task information, the first task information comprising the target task template and the task parameter; receiving response information of the first task information; The executing of the task step comprised in the target task template according to the task parameter comprises: in a case where the response information indicates that the first task information is correct, executing the task step comprised in the target task template according to the task parameter.

6. The method of claim 5, wherein, The method further comprises: in a case where the response information carries second task information, executing a task step recorded in the second task information according to a task parameter contained in the second task information, the second task information being task information obtained by modifying the first task information.

7. The method of any one of claims 1 to 6, wherein, The task parameter comprises an input parameter; or the task parameter comprises the input parameter and an execution parameter; wherein the input parameter is a parameter input into the target task template when the target task template is executed, and the execution parameter is a parameter triggering execution of the target task template.

8. The method of claim 7, wherein: the execution parameter comprises a start time triggering execution of the target task template and / or a cycle duration triggering periodic execution of the target task template; or the execution parameter comprises a type of an event triggering execution of the target task template and / or an identification (ID) of the event.

9. An operation and maintenance system, characterized by The method comprises: an obtaining unit, configured to obtain a task request, the task request being a user intention described in natural language, the user intention comprising a task parameter and a task target; a generating unit, configured to generate a target task template according to the task target, the target task template comprising a task step for achieving the task target; An execution unit is configured to execute the task steps included in the target task template according to the task parameters, to obtain a task result matched with the task request.

10. The system of claim 9, wherein, The system further includes: A judgment unit is configured to determine whether the target task template exists in the task template set according to the task target. The generation unit is specifically configured to generate the target task template according to the task target in a case where the target task template does not exist in the task template set.

11. The system of claim 9 or 10, wherein The generation unit is specifically configured to input the task target into a large language model (LLM) to obtain the target task template output by the LLM.

12. The system of any one of claims 9 to 11, wherein, The system further includes: A receiving unit is configured to receive an abnormal alarm reported by a network device. A determination unit is configured to determine a target de-noising rule according to the abnormal alarm and context information of the abnormal alarm, wherein the context information includes a historical alarm of the same type as the abnormal alarm and a de-noising rule determined based on the historical alarm. A filtering unit is configured to filter invalid alarms in the received abnormal alarms according to the target de-noising rule.

13. The system of any one of claims 9 to 12, wherein, The system further includes: An output unit is configured to output first task information, wherein the first task information includes the target task template and the task parameters. A receiving unit is configured to receive response information of the first task information. The execution unit is configured to execute the task steps included in the target task template according to the task parameters in a case where the response information indicates that the first task information is correct.

14. The system of claim 13, wherein The execution unit is further configured to execute task steps recorded in second task information according to task parameters included in the second task information in a case where the response information carries the second task information, wherein the second task information is task information obtained by modifying the first task information.

15. The system of any one of claims 9 to 14, wherein, The task parameters include input parameters; or the task parameters include the input parameters and execution parameters. The input parameters are parameters input into the target task template when the target task template is executed, and the execution parameters are parameters triggering execution of the target task template.

16. The system of claim 15, wherein The execution parameters include a start time triggering execution of the target task template and / or a cycle duration triggering periodic execution of the target task template; or The execution parameters include a type of an event triggering execution of the target task template and / or an identification (ID) of the event.

17. The system of any one of claims 12 to 16, wherein, The generation unit, the judgment unit, and the determination unit are implemented by an artificial intelligence agent core (AI agent core) based on a large language model (LLM).

Images