Network service processing method, electronic equipment and device

Abstract

The embodiment of the invention provides a network service processing method, electronic equipment and a network service processing device, relates to the technical field of artificial intelligence, is applied to a first agent, and comprises the following steps: obtaining a to-be-processed service demand in a network environment; forwarding the service demand to a plurality of second intelligent agents, so that each second intelligent agent judges whether the second intelligent agent can execute the task to meet the service demand based on the memory information; determining a target agent for executing the business demand according to the judgment result of each second agent; and controlling the target agent to execute the task so as to meet business requirements. By applying the scheme provided by the embodiment of the invention, the processing of the network service can be conveniently realized.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a network service processing method, electronic equipment, and device. Background Art

[0002] Network service scenarios present extremely complex challenges, encompassing five broad scenarios: network planning and design, network construction, network maintenance, network optimization, and network operations. These scenarios also encompass hundreds of sub-scenarios, including solution planning, wiring layout, and fault location. These scenarios involve a wide range of devices, including wireless equipment, access switches, aggregation switches, core switches, and routers. For different scenarios and devices, personnel must control network equipment in different ways and perform different processing to implement diverse services and meet diverse network requirements. This results in highly complex processes and low efficiency for implementing network services. Summary of the Invention

[0003] The purpose of the embodiments of the present application is to provide a network service processing method, electronic device, and apparatus to conveniently process network services. The specific technical solutions are as follows:

[0004] In a first aspect, an embodiment of the present application provides a network service processing method, which is applied to a first agent and includes:

[0005] Obtain pending business requirements in the network environment;

[0006] Forwarding the business demand to multiple second agents, so that each second agent determines whether it can perform tasks to meet the business demand based on memory information, wherein each second agent is used to perform different tasks in the network environment, and the memory information includes environmental memory information and / or exclusive memory information representing business demands that have been processed in the network environment. Each set of exclusive memory information corresponds to a first agent or a second agent, indicating the business processing capability of the corresponding first agent or second agent and / or its historical task execution process;

[0007] Determining a target agent for executing the business requirement based on the judgment results of each second agent;

[0008] Control the target agent to perform tasks to meet the business needs.

[0009] In one embodiment of the present application, the environmental memory information includes global memory information and / or short-term memory information;

[0010] The global memory information includes at least one of the following information: historical process information of historical workflows for processing historical business needs, capability information of each agent, environmental information of the network environment, knowledge information required for each agent to perform tasks, and a report template for a result report representing the result of processing business needs;

[0011] The short-term memory information includes at least one of the following information: a first business requirement processed within a recent preset time period, a time when the first business requirement was generated, first process information of a first workflow for processing the first business requirement, a processing result of the first business requirement, and event information of an event occurring in the network environment within a recent preset time period;

[0012] Each group of exclusive memory information includes at least one of the following information: capability information of the corresponding intelligent agent, tool information of the tools that the corresponding intelligent agent can dispatch, task processing flow information of the corresponding intelligent agent performing tasks, scheduling information indicating other intelligent agents dispatched by the corresponding intelligent agent and / or dispatched by other intelligent agents, and knowledge information required for the corresponding intelligent agent to perform tasks.

[0013] In one embodiment of the present application, when the memory information includes the first service requirement and the first process information in the short-term memory information, after obtaining the service requirement to be processed in the network environment, the method further includes:

[0014] If the business requirement is the same as the first business requirement in the short-term memory information, the second agent included in the first workflow represented by the first process information corresponding to the first business requirement is controlled to perform the task according to the first workflow.

[0015] In one embodiment of the present application, when the environment memory information includes the historical process information and the first process information, the method further includes:

[0016] If the first workflow indicated by the first process information is capable of achieving the corresponding first business requirement, determining, based on the first process information and historical process information of a preset business process, whether the execution time of the first workflow is less than the execution time of the preset workflow, wherein the preset workflow is the historical workflow with the shortest execution time for achieving the first business requirement;

[0017] If the execution time of the first workflow is shorter than the execution time of the preset workflow, the first process information of the first workflow is recorded as historical process information in the global memory information, and the first workflow is used as a new preset workflow.

[0018] In one embodiment of the present application, the method further includes:

[0019] For a preset simulated business requirement, generating a simulated workflow based on the memory information, wherein the simulated business requirement is a business requirement expected to be generated;

[0020] In the case that the simulated workflow cannot realize the simulated business requirements, if the business requirements actually obtained are the simulated business requirements, other workflows outside the simulated workflow are generated, and the intelligent agents contained in the generated workflow are controlled to perform tasks according to the generated workflow.

[0021] In one embodiment of the present application, controlling the target agent to perform tasks to meet the business needs includes:

[0022] Based on the environmental memory information and / or the exclusive memory information corresponding to the target agent, a workflow is generated to implement the business requirement, wherein the workflow represents the tasks to be performed by the target agent and the order in which the tasks are performed;

[0023] Control the target agent to perform the corresponding tasks in the order indicated by the workflow.

[0024] In one embodiment of the present application, after controlling the target agent to perform tasks to meet the business requirements, the method further includes:

[0025] Generate a result report indicating the processing result of the business requirement according to a preset report template and the processing result of the business requirement.

[0026] In a second aspect, an embodiment of the present application provides an electronic device, the electronic device comprising:

[0027] processor;

[0028] transceiver;

[0029] A machine-readable storage medium storing machine-executable instructions that can be executed by the processor, wherein the machine-executable instructions cause the processor to control the first agent to perform the following steps:

[0030] Obtain pending business requirements in the network environment;

[0031] Forwarding the business demand to multiple second agents, so that each second agent determines whether it can perform tasks to meet the business demand based on memory information, wherein each second agent is used to perform different tasks in the network environment, and the memory information includes environmental memory information and / or exclusive memory information representing business demands that have been processed in the network environment. Each set of exclusive memory information corresponds to a first agent or a second agent, indicating the business processing capability of the corresponding first agent or second agent and / or its historical task execution process;

[0032] Determining a target agent for executing the business requirement based on the judgment results of each second agent;

[0033] Control the target agent to perform tasks to meet the business needs.

[0034] In one embodiment of the present application, the environmental memory information includes global memory information and / or short-term memory information;

[0035] The global memory information includes at least one of the following information: historical process information of historical workflows for processing historical business needs, capability information of each agent, environmental information of the network environment, knowledge information required for each agent to perform tasks, and a report template for a result report representing the result of processing business needs;

[0036] The short-term memory information includes at least one of the following information: a first business requirement processed within a recent preset time period, a time when the first business requirement was generated, first process information of a first workflow for processing the first business requirement, a processing result of the first business requirement, and event information of an event occurring in the network environment within a recent preset time period;

[0037] Each group of exclusive memory information includes at least one of the following information: capability information of the corresponding intelligent agent, tool information of the tools that the corresponding intelligent agent can dispatch, task processing flow information of the corresponding intelligent agent performing tasks, scheduling information indicating other intelligent agents dispatched by the corresponding intelligent agent and / or dispatched by other intelligent agents, and knowledge information required for the corresponding intelligent agent to perform tasks.

[0038] In one embodiment of the present application, when the memory information includes the first service requirement and the first process information in the short-term memory information, after obtaining the service requirement to be processed in the network environment, the method further includes:

[0039] If the business requirement is the same as the first business requirement in the short-term memory information, the second agent included in the first workflow represented by the first process information corresponding to the first business requirement is controlled to perform the task according to the first workflow.

[0040] In one embodiment of the present application, when the environment memory information includes the historical process information and the first process information, the machine executable instructions further cause the processor to execute the following steps:

[0041] If the first workflow indicated by the first process information is capable of achieving the corresponding first business requirement, determining, based on the first process information and historical process information of a preset business process, whether the execution time of the first workflow is less than the execution time of the preset workflow, wherein the preset workflow is the historical workflow with the shortest execution time for achieving the first business requirement;

[0042] If the execution time of the first workflow is shorter than the execution time of the preset workflow, the first process information of the first workflow is recorded as historical process information in the global memory information, and the first workflow is used as a new preset workflow.

[0043] In one embodiment of the present application, the machine executable instructions further cause the processor to perform the following steps:

[0044] For a preset simulated business requirement, generating a simulated workflow based on the memory information, wherein the simulated business requirement is a business requirement expected to be generated;

[0045] When it is determined that the simulated workflow cannot achieve the simulated business requirements, if the business requirements actually obtained are the simulated business requirements, other workflows outside the simulated workflow are generated, and the intelligent agents contained in the generated workflow are controlled to perform tasks according to the generated workflow.

[0046] In one embodiment of the present application, controlling the target agent to perform tasks to meet the business needs includes:

[0047] Based on the environmental memory information and / or the exclusive memory information corresponding to the target agent, a workflow is generated to implement the business requirement, wherein the workflow represents the tasks to be performed by the target agent and the order in which the tasks are performed;

[0048] Control the target agent to perform the corresponding tasks in the order indicated by the workflow.

[0049] In one embodiment of the present application, after controlling the target agent to perform tasks to meet the business requirements, the method further includes:

[0050] Generate a result report indicating the processing result of the business requirement according to a preset report template and the processing result of the business requirement.

[0051] In a third aspect, an embodiment of the present application provides a network service processing device, which is applied to a first agent and includes:

[0052] Demand acquisition module, used to obtain business requirements to be processed in the network environment;

[0053] a demand forwarding module, configured to forward the business demand to a plurality of second agents, so that each second agent determines whether it is capable of performing tasks to meet the business demand based on memory information, wherein each second agent is respectively configured to perform different tasks in the network environment, the memory information including environmental memory information and / or exclusive memory information representing business demands that have been processed in the network environment, each set of exclusive memory information corresponding to a first agent or a second agent, indicating the business processing capability of the corresponding first agent or second agent and / or its historical task execution process;

[0054] A target agent determination module, configured to determine a target agent for executing the business requirement based on the judgment results of each second agent;

[0055] The first task execution module is used to control the target agent to execute tasks to meet the business needs.

[0056] In one embodiment of the present application, the environmental memory information includes global memory information and / or short-term memory information;

[0057] The global memory information includes at least one of the following information: historical process information of historical workflows for processing historical business needs, capability information of each agent, environmental information of the network environment, knowledge information required for each agent to perform tasks, and a report template for a result report representing the result of processing business needs;

[0058] The short-term memory information includes at least one of the following information: a first business requirement processed within a recent preset time period, a time when the first business requirement was generated, first process information of a first workflow for processing the first business requirement, a processing result of the first business requirement, and event information of an event occurring in the network environment within a recent preset time period;

[0059] Each group of exclusive memory information includes at least one of the following information: capability information of the corresponding intelligent agent, tool information of the tools that the corresponding intelligent agent can dispatch, task processing flow information of the corresponding intelligent agent performing tasks, scheduling information indicating other intelligent agents dispatched by the corresponding intelligent agent and / or dispatched by other intelligent agents, and knowledge information required for the corresponding intelligent agent to perform tasks.

[0060] In one embodiment of the present application, when the memory information includes the first business requirement and the first process information in the short-term memory information, the apparatus further includes:

[0061] The second task execution module is used to control the second agent included in the first workflow represented by the first process information corresponding to the first business requirement to execute the task according to the first workflow if the above-mentioned business requirement is the same as the first business requirement in the short-term memory information.

[0062] In one embodiment of the present application, when the environment memory information includes the historical process information and the first process information, the apparatus further includes:

[0063] a duration determination module configured to determine, if the first workflow indicated by the first process information is capable of achieving the corresponding first business requirement, whether the execution duration of the first workflow is less than the execution duration of the preset workflow based on the first process information and historical process information of the preset business process, wherein the preset workflow is the historical workflow with the shortest execution duration for achieving the first business requirement;

[0064] The workflow determination module is used to record the first process information of the first workflow as historical process information in the global memory information if the execution time of the first workflow is shorter than the execution time of the preset workflow, and to use the first workflow as a new preset workflow.

[0065] In one embodiment of the present application, the device further comprises:

[0066] A simulation process generation module, configured to generate a simulation workflow based on the memory information for a preset simulation business requirement, wherein the simulation business requirement is a business requirement expected to be generated;

[0067] The third task execution module is used to generate other workflows outside the simulated workflow if the simulated workflow cannot achieve the simulated business requirements, and control the intelligent agents contained in the generated workflow to perform tasks according to the generated workflow if the business requirements actually obtained are the simulated business requirements.

[0068] In one embodiment of the present application, the first task execution module is specifically configured to:

[0069] Based on the environmental memory information and / or the exclusive memory information corresponding to the target agent, a workflow is generated to implement the business requirement, wherein the workflow represents the tasks to be performed by the target agent and the order in which the tasks are performed;

[0070] Control the target agent to perform the corresponding tasks in the order indicated by the workflow.

[0071] In one embodiment of the present application, the device further comprises:

[0072] The result report generating module is used to generate a result report indicating the processing result of the business requirement according to a preset report template and the processing result of the business requirement.

[0073] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any of the method steps described in the first aspect is implemented.

[0074] In a fifth aspect, an embodiment of the present application further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute any of the methods described in the first aspect above.

[0075] Beneficial effects of the embodiments of the present application:

[0076] In the network service processing method provided in the embodiment of the present application, the first agent and the second agent can interact collaboratively, and the second agent can determine whether it needs to participate in the processing of business needs as a target agent based on the memory information. In this way, the first agent and the target agent can collaboratively complete the processing of business needs. This process is automatically implemented by the agent, which can greatly reduce the complexity of the network service processing flow, thereby conveniently realizing the processing of network services. In addition, whether each agent needs to process business needs is not pre-set, so the staff does not need to pre-configure the agents that participate in the business demand processing for various possible business needs, which can reduce labor costs. In addition, compared with the fixed agents that participate in the business demand processing pre-configured for business needs, the target agent determined based on the actual network environment represented by the memory information and / or the situation of the agent itself can make the selected target agent more compatible with the network environment and / or the situation of the agent itself, thereby improving the effect of the target agent in completing business needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0078] Figure 1 A schematic diagram of a business scenario provided in an embodiment of the present application;

[0079] Figure 2 A flowchart of a first network service processing method provided in an embodiment of the present application;

[0080] Figure 3 A schematic diagram of memory information provided in an embodiment of the present application;

[0081] Figure 4 A flowchart of a second network service processing method provided in an embodiment of the present application;

[0082] Figure 5 A flowchart of a third network service processing method provided in an embodiment of the present application;

[0083] Figure 6 A flowchart of a fourth network service processing method provided in an embodiment of the present application;

[0084] Figure 7 A flowchart of a fifth network service processing method provided in an embodiment of the present application;

[0085] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0086] Figure 9 A schematic diagram of the structure of a network service processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0087] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field based on this application are within the scope of protection of this application.

[0088] In order to better describe the embodiments of the present application, first, a business scenario in a network environment is introduced.

[0089] See also Figure 1 , which is a schematic diagram of a business scenario provided in an embodiment of the present application.

[0090] The diagram includes planning, construction, maintenance, optimization, and operation scenarios. The planning scenario includes network planning sub-scenarios, the construction scenario includes design and deployment sub-scenarios, the maintenance scenario includes monitoring and troubleshooting sub-scenarios, network inspection sub-scenarios, and network testing sub-scenarios, the optimization scenario includes quality optimization sub-scenarios and energy efficiency optimization sub-scenarios, and the operation scenario includes service provisioning sub-scenarios and network complaint sub-scenarios.

[0091] The network planning sub-scenario includes 1. demand forecasting business and 2. planning solution business.

[0092] The design and deployment sub-scenarios include 3. Solution Integration, 4. Project Acceptance and Quality Inspection, and 5. Deployment Configuration.

[0093] The monitoring and troubleshooting sub-scenario includes 6. Fault identification service, 7. Hidden danger identification service, 8. Delimitation and positioning service, 9. Treatment solution service, 10. Fault handling service, and 11. Scenario monitoring service.

[0094] The network inspection sub-scenario includes 12. command inspection service and 13. on-site inspection service.

[0095] The network test sub-scenario includes 14. Network test service and 15. Service test service.

[0096] The quality optimization sub-scenarios include 16. quality identification business, 17. quality positioning business, 18. quality optimization solution business, and 19. quality optimization verification business.

[0097] The energy efficiency optimization sub-scenario includes 20. Energy efficiency problem identification business, 21. Energy efficiency problem location business, 22. Energy efficiency optimization solution business, and 23. Energy efficiency optimization verification business.

[0098] The service activation sub-scenarios include 24. Activating the scheduling process service, 25. Resource exploration service, and 26. Service installation, maintenance, and quality inspection service.

[0099] The online complaint sub-scenarios include 27. Batch complaint warning service, 28. User complaint prediction service, and 29. Complaint demarcation service.

[0100] Each business shown in the figure can be completed by one or more intelligent agents in collaboration, and each intelligent agent implements one or more tasks in the workflow of the business.

[0101] See also Figure 2 , which is a flow chart of the first network business processing method provided in an embodiment of the present application, applied to the first intelligent agent, including the following steps S201-S204.

[0102] S201: Obtaining business requirements to be processed in a network environment.

[0103] S202: Multiple second agents forward the above business requirements, so that each second agent determines whether it can perform the task to meet the above business requirements based on the memory information.

[0104] Among them, each second intelligent agent is used to perform different tasks in the network environment. The above-mentioned memory information includes environmental memory information and / or exclusive memory information representing the business needs that have been processed in the network environment. Each set of exclusive memory information corresponds to a first intelligent agent or a second intelligent agent, indicating the corresponding first intelligent agent or second intelligent agent's ability to handle business and / or its process of executing tasks in history.

[0105] Each agent, including a first agent and a second agent, is used to perform different tasks in a network environment.

[0106] Individual agents form a multi-agent system. These agents can communicate with each other through APIs, interacting and collaborating to solve complex business needs or execute tasks. APIs are tools that enable interoperability between agents, acting as an intermediary layer for data transmission between agent systems.

[0107] Each agent is a software module running on an electronic device. It can perform tasks based on a large artificial intelligence language model to meet business needs.

[0108] For example, in one embodiment of the present application, there may be a user interaction agent (used to interact directly with the user and understand the user's intentions), an access management agent (used to manage whether the user is allowed access), a network monitoring agent (used to analyze the monitoring logs of network equipment and discover faults), an alarm processing agent (used to generate fault solutions for network faults), and a configuration management agent (used to be responsible for the generation, review and issuance of configuration commands, scripts, etc.).

[0109] In one embodiment of the present application, the above-mentioned business requirements can be manually input by the user. For example, the business requirements manually input by the user are "solving the problem of the conference room being unable to access the Internet", "how to solve the problem of router disconnection", "how to configure the switch", etc., then the above-mentioned first intelligent agent is a user interaction intelligent agent.

[0110] In another embodiment of the present application, the aforementioned service requirement may also be generated by the first agent itself. For example, if the first agent is a network monitoring agent, the network monitoring agent may generate a service requirement for resolving a network failure upon detecting the failure. Alternatively, if the first agent is a configuration management agent, the configuration management agent may generate a service requirement for updating the network configuration upon determining that the configuration of a network device needs to be updated.

[0111] Specifically, each second agent inputs the business requirements into the large language model configured by itself, and the large language model uses memory information to determine whether it is the target agent.

[0112] The above-mentioned environmental memory information includes global memory information and / or short-term memory information.

[0113] Among them, the above-mentioned global memory information is the continuously stored information shared by each intelligent agent, the short-term memory information is the information shared by each intelligent agent and will be deleted after the preset storage period, each set of exclusive memory information corresponds to an intelligent agent, and each set of exclusive memory information is only used by the corresponding intelligent agent.

[0114] The above-mentioned global memory information includes at least one of the following information: historical process information of historical workflows for processing historical business needs, capability information of each intelligent agent, environmental information of the above-mentioned network environment, knowledge information required for each intelligent agent to perform tasks, and a report template for a result report representing the results of business demand processing.

[0115] Among them, historical process information includes historical business needs, the intelligent agents involved in processing historical business needs, the tasks performed by each intelligent agent, the tools used by the intelligent agent to perform tasks, the time it takes for the intelligent agent to perform tasks, the order in which the intelligent agent performs tasks, the processing results of historical business needs, the processing effect of historical business needs, etc. The processing effect can be measured by whether the historical business needs are resolved, the number of intelligent agents that need to be called to resolve the historical business needs, and the time required to resolve the historical business needs.

[0116] The capability information of each agent includes: tasks that the agent can handle, tools that the agent can call, interfaces that can be called, etc. The above tools include interfaces, models, algorithm blocks, etc.

[0117] The environmental information of the network environment includes: network topology, device models of network devices, hardware configuration of network devices and other information.

[0118] Global memory information contains knowledge shared by all agents, such as network equipment configuration guides, functional specifications, configuration upgrade manuals, operation manuals, communication protocol information, industry white papers, etc. This knowledge information can come from industry experts or knowledge bases in the network.

[0119] The report template is used to define the content that needs to be included in the result report, the data type, length, format, etc. of each item.

[0120] The above-mentioned short-term memory information includes at least one of the following information: the first business demand processed within the most recent preset time period, the time when the above-mentioned first business demand was generated, the first process information of the first workflow for processing the above-mentioned first business demand, the processing result of the above-mentioned first business demand, and the event information of the event occurring in the above-mentioned network environment within the most recent preset time period.

[0121] The first process information includes the agents involved in the first workflow, the processing order of the agents involved, and the execution time of the first workflow.

[0122] Events occurring in the network environment include: network failures, network congestion, network equipment system upgrades, etc. Event information includes: the location of the failure, the time of failure, the number of failures, the result of the failure handling, the location of the congestion, the time of congestion, the duration of the congestion, the network equipment system upgraded, and the system version numbers of the network equipment before and after the upgrade.

[0123] Each group of exclusive memory information includes at least one of the following information: capability information of the corresponding intelligent agent, tool information of the tools that the corresponding intelligent agent can dispatch, task processing flow information of the corresponding intelligent agent performing tasks, scheduling information indicating other intelligent agents dispatched by the corresponding intelligent agent and / or dispatched by other intelligent agents, and knowledge information required for the corresponding intelligent agent to perform tasks.

[0124] For example, the knowledge information in the exclusive memory information includes detailed information of the API that can be called by the agent corresponding to the exclusive memory information and information related to the task that the agent needs to perform.

[0125] See also Figure 3 , is a schematic diagram of memory information provided in an embodiment of the present application.

[0126] The diagram contains global memory information and exclusive memory information. As can be seen from the diagram, global memory information includes networking information, configuration details, business processes, agent information, general knowledge information, and report templates.

[0127] Exclusive memory information includes knowledge base, information query, configuration distribution, troubleshooting, and report generation.

[0128] The knowledge base includes configuration guides, functional specifications, white papers, deployment and upgrade manuals, and operation manuals.

[0129] Information query includes the API library and user information. The API library includes the API name, API parameters, and the meaning of the API return content used for information query. User information includes user information and service and device association information.

[0130] Configuration delivery includes an API library, code, and scripts. The API library contains the API name, API parameters, and the meaning of the API response data used for configuration delivery. The code and scripts include templates, problem-solving tools, and customizable variables and parameters.

[0131] Troubleshooting includes a troubleshooting process knowledge base and a root cause location expert experience base. The root cause location expert experience base includes business phenomena, network phenomena, and root causes.

[0132] Report generation includes APIs and templates for generating reports. Templates include report templates and report templates.

[0133] From the above, it can be seen that the memory information in the embodiments of the present application may include global memory information, short-term memory information, and exclusive memory information. The usage scope and storage time of different information are different, that is, the memory information may be composed of rich information of multiple different dimensions. Based on the rich information, the intelligent agent can accurately understand complex network problems and provide more accurate solutions.

[0134] S203: Determine a target agent for executing the above business requirement based on the judgment results of each second agent.

[0135] S204: Control the target agent to execute tasks to meet the business requirements.

[0136] In one embodiment of the present application, for question-and-answer business needs, the first intelligent agent can control itself and the target intelligent agent to call the artificial intelligence model to generate answers to questions, and feed back the generated answers to the first intelligent agent. The first intelligent agent summarizes the answers to obtain processing results.

[0137] For example, the question-and-answer business requirement mentioned above is a business requirement that requires outputting corresponding answers to user questions, such as "How to configure network equipment" and "What faults have occurred in routers in the past month?"

[0138] In another embodiment of the present application, for business needs requiring automatic processing, step S204 can be implemented by following steps A and B. Business needs requiring automatic processing are business needs where the agent needs to analyze and then operate network devices to control their operation, for example, troubleshooting network device failures, controlling network devices for system upgrades, etc. The first agent can invoke a large language model or a small classification model to determine whether the business need is a question-and-answer business need or a business need requiring automatic processing.

[0139] Step A: Based on the above-mentioned environmental memory information and / or the exclusive memory information corresponding to the above-mentioned target intelligent agent, generate a workflow to achieve the above-mentioned business needs.

[0140] The above workflow represents the tasks to be performed by the above target agent and the order in which each task is performed.

[0141] The first agent can use the large language model to determine business needs and, based on environmental memory information and / or the target agent's unique memory information, determine the current network conditions, the capabilities of the target agents, and the tools they can use. It then determines the role each target agent can play in handling the business needs, thereby assigning tasks to the target agents.

[0142] Step B: Control the target agent to perform the corresponding tasks in the order indicated by the workflow.

[0143] For example, a business requirement is to address the issue of mobile phones being unable to access the internet in a conference room. The user interaction agent, acting as the first agent, determines the following workflow: It dispatches the access management agent to check whether the mobile phone has recently requested access to the conference room's equipment; it dispatches the network monitoring agent to check whether there are any faults in the network equipment and links from the conference room's access point to the core; and it dispatches the alarm processing agent to determine which alarm actions related to the conference room's access point have been executed within a limited period of time (e.g., half an hour, an hour, etc.). The user interaction agent collects feedback from the target agents (i.e., the access management agent, the network monitoring agent, and the alarm processing agent). For example, the access management agent reports that the most recent authentication record is from two hours ago, the network monitoring agent reports an alarm indicating a problem with the link between the access point and the upstream switch, and the alarm processing agent reports that the access point was rebooted yesterday and that the serial number of the relevant device has changed, indicating a replacement. Based on this information, the user interaction agent invokes the large language model, which outputs the following processing result: The uplink is faulty. Please ask the administrator to check and replace it as soon as possible. The user interaction agent generates a report representing the processing results for the user to review and perform subsequent processing.

[0144] As can be seen from the above, in the embodiment of the present application, the first agent and the second agent can interact collaboratively, and the second agent can determine whether it needs to participate in the processing of business needs as a target agent based on the memory information. This enables the first agent and the target agent to collaboratively complete the processing of business needs. This process is automatically implemented by the agent, which can greatly reduce the complexity of the network business processing flow, thereby conveniently realizing the processing of network business. In addition, whether each agent needs to process business needs is not pre-set, so the staff does not need to pre-configure the agents that participate in the business demand processing for various possible business needs, which can reduce labor costs. In addition, compared with the fixed agents that participate in the business demand processing pre-configured for business needs, the target agent determined based on the actual network environment represented by the memory information and / or the situation of the agent itself can make the selected target agent more compatible with the network environment and / or the situation of the agent itself, thereby improving the effect of the target agent in completing business needs.

[0145] In another embodiment of the present application, the following step C is further performed after step S204.

[0146] Step C: Generate a result report indicating the processing result of the above business requirement according to a preset report template and the processing result of the above business requirement.

[0147] As can be seen from the above, in the embodiment of the present application, the first agent can also generate a result report so that the user can view the result report to determine the processing effect of the business needs.

[0148] The following describes the process of using the above-mentioned global memory information and short-term memory information with a specific embodiment.

[0149] In a complex enterprise network environment, multiple agents (such as user interaction agents, access management agents, network monitoring agents, alarm processing agents, and configuration management agents) work together to perform automated troubleshooting and repair. The scenario is that the fault has not occurred in the past month (not stored in short-term memory information) and a similar fault has occurred three times in the past year (stored in global memory information). The following is the process of using the global memory and short-term memory in this scenario:

[0150] A company's network system experienced a malfunction. A network monitoring agent detected the anomaly and, as the first agent, triggered a troubleshooting process. In this process, multiple agents collaborated to determine the cause of the malfunction and implement appropriate remedial measures.

[0151] 1. Details of the usage process of global memory information:

[0152] Problem Detection: If the network monitoring agent fails to retrieve information about the same fault scenario from its short-term memory, it will instead retrieve fault records and their handling procedures from its global memory to reference similar fault handling methods. This global memory contains historical fault information for the enterprise network, including the fault type, the devices involved, the historical handling procedures, and the historical results.

[0153] Capability query: Each agent accesses global memory information to query the currently available agent capabilities and network environment configuration. For example, the alarm processing agent uses global memory information to understand the currently available alarm handling strategies, available tools, network topology, and network device configuration.

[0154] Process execution: With the support of global memory information, each intelligent agent works together to formulate and execute the best fault workflow through comprehensive judgment of historical data and capability information. Finally, the configuration management intelligent agent sends relevant maintenance scripts and configuration policies to network devices to complete the fault handling.

[0155] As the network multi-agent processes the fault, it will record the key information of the processing process into the short-term memory information.

[0156] 2. Use of short-term memory information:

[0157] Process details:

[0158] Original business requirement record: Short-term memory information first records the original business requirements detected during this troubleshooting, such as the specific network anomaly type and the network devices involved.

[0159] Workflow Recording: As troubleshooting progresses, short-term memory dynamically records the specific steps each agent performs, the tools used, and the intermediate results. For example, a network monitoring agent may first detect an anomaly in traffic flow on a specific switch and record this information in short-term memory.

[0160] Result Storage: After completing its task, each agent stores intermediate and final results in short-term memory. For example, after attempting to reconfigure a faulty switch, the configuration management agent stores the success or failure status and time information in short-term memory.

[0161] Quick response: When it is necessary to review recent operations or make the next decision, the agent can quickly retrieve relevant information from short-term memory to ensure the consistency and efficiency of the processing process.

[0162] The role of short-term memory information: Short-term memory information supports the agent to quickly access recent operation records and results in the current task, enhances the real-time response capability of the system, and avoids the delay caused by repeated query of the global memory library.

[0163] See also Figure 4 , is a flow chart of the second network service processing method provided in the embodiment of the present application, which is similar to the aforementioned Figure 2 Compared with the embodiment shown, when the memory information includes the first business requirement and the first process information in the short-term memory information, the following step S205 is further included after step S201.

[0164] S205: If the business requirement is the same as the first business requirement in the short-term memory information, the second agent included in the first workflow represented by the first process information corresponding to the first business requirement is controlled to perform tasks according to the first workflow.

[0165] Specifically, the above-mentioned short-term memory information records the first business demand processed within the most recent preset time period, that is, the first business demand processed recently. If the business demand obtained this time is the same as the first business demand, it means that the business demand has been processed recently. The first intelligent agent can directly process the first business demand according to the first workflow represented by the first process information without having to determine the workflow again.

[0166] In addition, after receiving the service demand, the first agent can determine whether the processing of the service demand is related to the actual situation of the network environment. If it is related to the actual situation of the network environment, since the actual situation of the network environment will change over time, even if the service demand belongs to the first service demand, the processing result of the first service demand may not be applicable to the current service demand. In this case, steps S202 to S204 need to be executed again for processing. For example, the service demand related to the actual situation of the network environment is a fault handling service demand.

[0167] However, if the business requirement is a question-and-answer type business requirement, for example, a business requirement asking how to configure a switch, because the way to configure a switch is basically fixed, if the business requirement belongs to the first business requirement in the short-term memory information, it can be directly processed according to the first workflow represented by the first process information corresponding to the first business requirement.

[0168] As can be seen above, in this embodiment, if the short-term memory information records a first business requirement that is identical to the newly acquired business requirement, it indicates that the business requirement has been repeatedly raised within a short period of time. In this embodiment, the first workflow indicated by the first process information corresponding to the first business requirement can be directly processed without re-determining the workflow. Therefore, in this embodiment, the timing information of the business requirements is used to identify the business requirement that is identical to the current business requirement from among the recently processed first business requirements. This first business requirement is then used to guide the processing of the current business requirement, thereby saving computing resources.

[0169] See also Figure 5 , is a flow chart of the third network service processing method provided by the embodiment of the present application, in which the above-mentioned memory information includes the above-mentioned historical process information and the above-mentioned first process information, and Figure 2 Compared with the embodiment shown, the above method further includes the following steps S206-S207.

[0170] S206: If the first workflow represented by the first process information can achieve the corresponding first business requirement, then based on the first process information and historical process information of the preset business process, determine whether the execution time of the first workflow is less than the execution time of the above preset workflow.

[0171] The preset workflow is a historical workflow with the shortest execution time for achieving the first business requirement.

[0172] If the first workflow represented by the first process information can achieve the first business requirement, then the first workflow represented by the first process information is a reasonable process for achieving the first business requirement. In this case, if the execution time of the first workflow is less than the working time of the preset workflow, then the first workflow is superior to the preset workflow and is more efficient than the preset workflow. Then, step S207 is executed.

[0173] S207: If the execution time of the first workflow is shorter than the execution time of the preset workflow, the first process information of the first workflow is recorded as historical process information in the global memory information, and the first workflow is used as a new preset workflow.

[0174] That is, in this embodiment, after a more optimal first workflow is found, the first workflow information is recorded in the global memory information, so that it is retained in the global memory information for a long time and has a long-term impact on the subsequent business processing of the agent. It is prevented from being deleted after being stored in the short-term memory information for a preset period of time.

[0175] It should be noted that the initial preset workflow may be set by experts based on experience.

[0176] In addition, steps S206 and S207 can be executed when there are idle computing resources in the electronic device to avoid affecting the normal operation of the intelligent agent of the electronic device and to fully utilize the computing resources of the electronic device. Any intelligent agent can execute steps S206 and S207 to compare the first workflow related to itself with the preset workflow.

[0177] From the above, it can be seen that in this application, after completing the processing of the first business demand, not only the first process information of the first business demand is recorded, but also the first workflow of the first business demand is reflected. If it is found to be better than the current preset workflow, the global memory information is updated, so that the global memory information is continuously optimized and enriched, thereby enabling the intelligent body to achieve better results when processing business needs based on the global memory information.

[0178] See also Figure 6 , is a flow chart of the fourth network service processing method provided in the embodiment of the present application, which is similar to the aforementioned Figure 2 Compared with the embodiment shown, the following steps S208-S209 are also included.

[0179] S208: Generate a simulation workflow based on the above-mentioned memory information in response to preset simulation business requirements.

[0180] Among them, the above-mentioned simulated business requirements are business requirements that are expected to be generated.

[0181] S209: When the above-mentioned simulated workflow cannot realize the above-mentioned simulated business requirements, if the business requirements actually obtained are the above-mentioned simulated business requirements, other workflows outside the simulated workflow are generated, and the intelligent agents contained in the generated workflow are controlled to perform tasks according to the generated workflow.

[0182] In this embodiment, the intelligent agent related to the preset simulated business demand can simulate the simulated business demand before actually receiving the simulated business demand to generate a simulated workflow. If the simulation result is that a simulated workflow that cannot realize the simulated business demand is found, it will be recorded, so that when the business demand is actually processed later, the simulated workflow can be avoided and other workflows can be tried to be used for processing, thereby achieving better processing results.

[0183] This embodiment can be executed when there are idle computing resources in the electronic device to avoid affecting the normal operation of the intelligent body of the electronic device and to fully utilize the computing resources of the electronic device.

[0184] This embodiment can be executed by any agent that participates in simulating business needs.

[0185] From the above, it can be seen that in the embodiment of the present application, the intelligent agent can also simulate the processing of business needs in advance before the business needs are generated, thereby achieving self-optimization and learning, and thus achieving better business processing results.

[0186] See also Figure 7 , which is a flowchart of the fifth network service processing method provided in an embodiment of the present application. It includes five steps: problem analysis, workflow arrangement, workflow execution, result reflection, and report output.

[0187] The goal of problem analysis is to classify problems and obtain parameters. The input is the business requirement, and the output is parameter information. Problem analysis relies on parameter templates and small classification models. Parameter information refers to the parameters that need to be configured, extracted, or processed to implement the business requirement. The small classification model is used to categorize business requirements, such as question-and-answer requirements and those requiring automated processing.

[0188] The goal of workflow orchestration is to decompose tasks and orchestrate agent workflows. The inputs are business requirements and parameter templates. The output is an agent workflow. The steps rely on the business requirements and corresponding workflow information in global and short-term memory, a large language model, and agent capability information.

[0189] The goal of workflow execution is for an agent to execute a workflow, invoke functions, and output processing results. Inputs include business requirements, parameter templates, and workflows. Workflow execution relies on the agent's ability to invoke interface functions.

[0190] The goal of result reflection is to determine whether the processing results meet expectations. If so, the report output is continued; otherwise, the process is re-processed. The inputs are business requirements and processing results. The output is whether the execution results meet expectations. Result reflection relies on global memory information and a large language model.

[0191] The goal of report output is to return a report to the user. The input is business requirements and processing results. The output is a report of the processing results. Report output relies on a large language model.

[0192] Corresponding to the aforementioned network service processing method, an embodiment of the present application also provides an electronic device.

[0193] See also Figure 8 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application, the electronic device comprising:

[0194] Processor 801;

[0195] transceiver 804;

[0196] A machine-readable storage medium 802 stores machine-executable instructions that can be executed by the processor 801, and the machine-executable instructions cause the processor 801 to control the first agent to perform the following steps:

[0197] Obtain pending business requirements in the network environment;

[0198] Forwarding the business demand to multiple second agents, so that each second agent determines whether it can perform tasks to meet the business demand based on memory information, wherein each second agent is used to perform different tasks in the network environment, and the memory information includes environmental memory information and / or exclusive memory information representing business demands that have been processed in the network environment. Each set of exclusive memory information corresponds to a first agent or a second agent, indicating the business processing capability of the corresponding first agent or second agent and / or its historical task execution process;

[0199] Determining a target agent for executing the business requirement based on the judgment results of each second agent;

[0200] Control the target agent to perform tasks to meet the business needs.

[0201] like Figure 8 As shown, the network device may further include a communication bus 803. The processor 801, the machine-readable storage medium 802, and the transceiver 804 communicate with each other via the communication bus 803. The communication bus 803 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus 803 may be divided into an address bus, a data bus, a control bus, and the like.

[0202] The transceiver 804 may be a wireless communication module. Under the control of the processor 801 , the transceiver 804 exchanges data with other devices.

[0203] The machine-readable storage medium 802 may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Alternatively, the machine-readable storage medium 802 may be at least one storage device located remote from the processor.

[0204] The processor 801 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0205] As can be seen from the above, in the embodiment of the present application, the first agent and the second agent can interact collaboratively, and the second agent can determine whether it needs to participate in the processing of business needs as a target agent based on the memory information. This enables the first agent and the target agent to collaboratively complete the processing of business needs. This process is automatically implemented by the agent, which can greatly reduce the complexity of the network business processing flow, thereby conveniently realizing the processing of network business. In addition, whether each agent needs to process business needs is not pre-set, so the staff does not need to pre-configure the agents that participate in the business demand processing for various possible business needs, which can reduce labor costs. In addition, compared with the fixed agents that participate in the business demand processing pre-configured for business needs, the target agent determined based on the actual network environment represented by the memory information and / or the situation of the agent itself can make the selected target agent more compatible with the network environment and / or the situation of the agent itself, thereby improving the effect of the target agent in completing business needs.

[0206] In one embodiment of the present application, the environmental memory information includes global memory information and / or short-term memory information;

[0207] The global memory information includes at least one of the following information: historical process information of historical workflows for processing historical business needs, capability information of each agent, environmental information of the network environment, knowledge information required for each agent to perform tasks, and a report template for a result report representing the result of processing business needs;

[0208] The short-term memory information includes at least one of the following information: a first business requirement processed within a recent preset time period, a time when the first business requirement was generated, first process information of a first workflow for processing the first business requirement, a processing result of the first business requirement, and event information of an event occurring in the network environment within a recent preset time period;

[0209] Each group of exclusive memory information includes at least one of the following information: capability information of the corresponding intelligent agent, tool information of the tools that the corresponding intelligent agent can dispatch, task processing flow information of the corresponding intelligent agent performing tasks, scheduling information indicating other intelligent agents dispatched by the corresponding intelligent agent and / or dispatched by other intelligent agents, and knowledge information required for the corresponding intelligent agent to perform tasks.

[0210] From the above, it can be seen that the memory information in the embodiments of the present application may include global memory information, short-term memory information, and exclusive memory information. The usage scope and storage time of different information are different, that is, the memory information may be composed of rich information of multiple different dimensions. Based on the rich information, the intelligent agent can accurately understand complex network problems and provide more accurate solutions.

[0211] In one embodiment of the present application, when the memory information includes the first service requirement and the first process information in the short-term memory information, after obtaining the service requirement to be processed in the network environment, the method further includes:

[0212] If the business requirement is the same as the first business requirement in the short-term memory information, the second agent included in the first workflow represented by the first process information corresponding to the first business requirement is controlled to perform the task according to the first workflow.

[0213] As can be seen above, in this embodiment, if the short-term memory information records a first business requirement that is identical to the newly acquired business requirement, it indicates that the business requirement has been repeatedly raised within a short period of time. In this embodiment, the first workflow indicated by the first process information corresponding to the first business requirement can be directly processed without re-determining the workflow. Therefore, in this embodiment, the timing information of the business requirements is used to identify the business requirement that is identical to the current business requirement from among the recently processed first business requirements. This first business requirement is then used to guide the processing of the current business requirement, thereby saving computing resources.

[0214] In one embodiment of the present application, when the environment memory information includes the historical process information and the first process information, the machine executable instructions further cause the processor to execute the following steps:

[0215] If the first workflow indicated by the first process information is capable of achieving the corresponding first business requirement, determining, based on the first process information and historical process information of a preset business process, whether the execution time of the first workflow is less than the execution time of the preset workflow, wherein the preset workflow is the historical workflow with the shortest execution time for achieving the first business requirement;

[0216] If the execution time of the first workflow is shorter than the execution time of the preset workflow, the first process information of the first workflow is recorded as historical process information in the global memory information, and the first workflow is used as a new preset workflow.

[0217] From the above, it can be seen that in this application, after completing the processing of the first business demand, not only the first process information of the first business demand is recorded, but also the first workflow of the first business demand is reflected. If it is found to be better than the current preset workflow, the global memory information is updated, so that the global memory information is continuously optimized and enriched, thereby enabling the intelligent body to achieve better results when processing business needs based on the global memory information.

[0218] In one embodiment of the present application, the machine executable instructions further cause the processor to perform the following steps:

[0219] For a preset simulated business requirement, generating a simulated workflow based on the memory information, wherein the simulated business requirement is a business requirement expected to be generated;

[0220] When it is determined that the simulated workflow cannot achieve the simulated business requirements, if the business requirements actually obtained are the simulated business requirements, other workflows outside the simulated workflow are generated, and the intelligent agents contained in the generated workflow are controlled to perform tasks according to the generated workflow.

[0221] From the above, it can be seen that in the embodiment of the present application, the intelligent agent can also simulate the processing of business needs in advance before the business needs are generated, thereby achieving self-optimization and learning, and thus achieving better business processing results.

[0222] In one embodiment of the present application, controlling the target agent to perform tasks to meet the business needs includes:

[0223] Based on the environmental memory information and / or the exclusive memory information corresponding to the target agent, a workflow is generated to implement the business requirement, wherein the workflow represents the tasks to be performed by the target agent and the order in which the tasks are performed;

[0224] Control the target agent to perform the corresponding tasks in the order indicated by the workflow.

[0225] In one embodiment of the present application, after controlling the target agent to perform tasks to meet the business requirements, the method further includes:

[0226] Generate a result report indicating the processing result of the business requirement according to a preset report template and the processing result of the business requirement.

[0227] As can be seen from the above, in the embodiment of the present application, the first agent can also generate a result report so that the user can view the result report to determine the processing effect of the business needs.

[0228] Corresponding to the aforementioned network service processing method, an embodiment of the present application also provides a network service processing device.

[0229] See also Figure 9 , is a schematic diagram of the structure of a network service processing device provided in an embodiment of the present application, the device is applied to a first agent, and the device includes:

[0230] The demand acquisition module 901 is used to acquire the business demand to be processed in the network environment;

[0231] A demand forwarding module 902 is configured to forward the business demand to multiple second agents, so that each second agent determines whether it can perform tasks to meet the business demand based on memory information. Each second agent is configured to perform different tasks in the network environment. The memory information includes environmental memory information and / or exclusive memory information representing business demands that have been processed in the network environment. Each set of exclusive memory information corresponds to a first agent or a second agent, indicating the business processing capability of the corresponding first agent or second agent and / or its historical task execution process.

[0232] A target agent determination module 903 is configured to determine a target agent for executing the business requirement based on the judgment results of each second agent;

[0233] The first task execution module 904 is used to control the target agent to execute tasks to meet the business needs.

[0234] As can be seen from the above, in the embodiment of the present application, the first agent and the second agent can interact collaboratively, and the second agent can determine whether it needs to participate in the processing of business needs as a target agent based on the memory information. This enables the first agent and the target agent to collaboratively complete the processing of business needs. This process is automatically implemented by the agent, which can greatly reduce the complexity of the network business processing flow, thereby conveniently realizing the processing of network business. In addition, whether each agent needs to process business needs is not pre-set, so the staff does not need to pre-configure the agents that participate in the business demand processing for various possible business needs, which can reduce labor costs. In addition, compared with the fixed agents that participate in the business demand processing pre-configured for business needs, the target agent determined based on the actual network environment represented by the memory information and / or the situation of the agent itself can make the selected target agent more compatible with the network environment and / or the situation of the agent itself, thereby improving the effect of the target agent in completing business needs.

[0235] In one embodiment of the present application, the environmental memory information includes global memory information and / or short-term memory information;

[0236] The global memory information includes at least one of the following information: historical process information of historical workflows for processing historical business needs, capability information of each agent, environmental information of the network environment, knowledge information required for each agent to perform tasks, and a report template for a result report representing the result of processing business needs;

[0237] The short-term memory information includes at least one of the following information: a first business requirement processed within a recent preset time period, a time when the first business requirement was generated, first process information of a first workflow for processing the first business requirement, a processing result of the first business requirement, and event information of an event occurring in the network environment within a recent preset time period;

[0238] Each group of exclusive memory information includes at least one of the following information: capability information of the corresponding intelligent agent, tool information of the tools that the corresponding intelligent agent can dispatch, task processing flow information of the corresponding intelligent agent performing tasks, scheduling information indicating other intelligent agents dispatched by the corresponding intelligent agent and / or dispatched by other intelligent agents, and knowledge information required for the corresponding intelligent agent to perform tasks.

[0239] From the above, it can be seen that the memory information in the embodiments of the present application may include global memory information, short-term memory information, and exclusive memory information. The usage scope and storage time of different information are different, that is, the memory information may be composed of rich information of multiple different dimensions. Based on the rich information, the intelligent agent can accurately understand complex network problems and provide more accurate solutions.

[0240] In one embodiment of the present application, when the memory information includes the first business requirement and the first process information in the short-term memory information, the apparatus further includes:

[0241] The second task execution module is used to control the second agent included in the first workflow represented by the first process information corresponding to the first business requirement to execute the task according to the first workflow if the above-mentioned business requirement is the same as the first business requirement in the short-term memory information.

[0242] As can be seen above, in this embodiment, if the short-term memory information records a first business requirement that is identical to the newly acquired business requirement, it indicates that the business requirement has been repeatedly raised within a short period of time. In this embodiment, the first workflow indicated by the first process information corresponding to the first business requirement can be directly processed without re-determining the workflow. Therefore, in this embodiment, the timing information of the business requirements is used to identify the business requirement that is identical to the current business requirement from among the recently processed first business requirements. This first business requirement is then used to guide the processing of the current business requirement, thereby saving computing resources.

[0243] In one embodiment of the present application, when the environment memory information includes the historical process information and the first process information, the apparatus further includes:

[0244] a duration determination module configured to determine, if the first workflow indicated by the first process information is capable of achieving the corresponding first business requirement, whether the execution duration of the first workflow is less than the execution duration of the preset workflow based on the first process information and historical process information of the preset business process, wherein the preset workflow is the historical workflow with the shortest execution duration for achieving the first business requirement;

[0245] The workflow determination module is used to record the first process information of the first workflow as historical process information in the global memory information if the execution time of the first workflow is shorter than the execution time of the preset workflow, and to use the first workflow as a new preset workflow.

[0246] From the above, it can be seen that in this application, after completing the processing of the first business demand, not only the first process information of the first business demand is recorded, but also the first workflow of the first business demand is reflected. If it is found to be better than the current preset workflow, the global memory information is updated, so that the global memory information is continuously optimized and enriched, thereby enabling the intelligent body to achieve better results when processing business needs based on the global memory information.

[0247] In one embodiment of the present application, the device further comprises:

[0248] A simulation process generation module, configured to generate a simulation workflow based on the memory information for a preset simulation business requirement, wherein the simulation business requirement is a business requirement expected to be generated;

[0249] The third task execution module is used to generate other workflows outside the simulated workflow if the simulated workflow cannot achieve the simulated business requirements, and control the intelligent agents contained in the generated workflow to perform tasks according to the generated workflow if the business requirements actually obtained are the simulated business requirements.

[0250] From the above, it can be seen that in the embodiment of the present application, the intelligent agent can also simulate the processing of business needs in advance before the business needs are generated, thereby achieving self-optimization and learning, and thus achieving better business processing results.

[0251] In one embodiment of the present application, the first task execution module 904 is specifically configured to:

[0252] Based on the environmental memory information and / or the exclusive memory information corresponding to the target agent, a workflow is generated to implement the business requirement, wherein the workflow represents the tasks to be performed by the target agent and the order in which the tasks are performed;

[0253] Control the target agent to perform the corresponding tasks in the order indicated by the workflow.

[0254] In one embodiment of the present application, the device further comprises:

[0255] The result report generating module is used to generate a result report indicating the processing result of the business requirement according to a preset report template and the processing result of the business requirement.

[0256] As can be seen from the above, in the embodiment of the present application, the first agent can also generate a result report so that the user can view the result report to determine the processing effect of the business needs.

[0257] In another embodiment provided by the present application, a computer-readable storage medium is further provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned network service processing methods are implemented.

[0258] In another embodiment provided by the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute any one of the network service processing methods in the above embodiments.

[0259] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. 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 a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0260] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0261] Each embodiment in this specification is described in a related manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, since the electronic device, device, computer-readable storage medium, and computer program product embodiments are generally similar to the method embodiments, their descriptions are relatively simple. For related portions, reference can be made to the descriptions of the method embodiments.

[0262] The above description is only a preferred embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application are included in the scope of protection of the present application.

Claims

1. A network service processing method, characterized in that: The method is applied to a first agent, and the method comprises: Obtain pending business requirements in the network environment; Forwarding the business demand to multiple second agents, so that each second agent determines whether it can perform tasks to meet the business demand based on memory information, wherein each second agent is used to perform different tasks in the network environment, and the memory information includes environmental memory information and / or exclusive memory information representing business demands that have been processed in the network environment. Each set of exclusive memory information corresponds to a first agent or a second agent, indicating the business processing capability of the corresponding first agent or second agent and / or its historical task execution process; Determining a target agent for executing the business requirement based on the judgment results of each second agent; Control the target agent to perform tasks to meet the business needs.

2. The method according to claim 1, characterized in that The environmental memory information includes global memory information and / or short-term memory information; The global memory information includes at least one of the following information: historical process information of historical workflows for processing historical business needs, capability information of each agent, environmental information of the network environment, knowledge information required for each agent to perform tasks, and a report template for a result report representing the result of processing business needs; The short-term memory information includes at least one of the following information: a first business requirement processed within a recent preset time period, a time when the first business requirement was generated, first process information of a first workflow for processing the first business requirement, a processing result of the first business requirement, and event information of an event occurring in the network environment within a recent preset time period; Each group of exclusive memory information includes at least one of the following information: capability information of the corresponding first or second intelligent agent, tool information of the tools that the corresponding first or second intelligent agent can dispatch, task processing flow information of the corresponding first or second intelligent agent performing tasks, scheduling information indicating other intelligent agents dispatched by the corresponding first or second intelligent agent and / or dispatched by other intelligent agents, and knowledge information required for the corresponding first or second intelligent agent to perform tasks.

3. The method according to claim 2, characterized in that In a case where the memory information includes the first service requirement and the first process information in the short-term memory information, after obtaining the service requirement to be processed in the network environment, the method further includes: If the business requirement is the same as the first business requirement in the short-term memory information, the second agent included in the first workflow represented by the first process information corresponding to the first business requirement is controlled to perform the task according to the first workflow.

4. The method according to claim 2, characterized in that When the environment memory information includes the historical process information and the first process information, the method further includes: If the first workflow indicated by the first process information is capable of achieving the corresponding first business requirement, determining, based on the first process information and historical process information of a preset business process, whether the execution time of the first workflow is less than the execution time of the preset workflow, wherein the preset workflow is the historical workflow with the shortest execution time for achieving the first business requirement; If the execution time of the first workflow is shorter than the execution time of the preset workflow, the first process information of the first workflow is recorded as historical process information in the global memory information, and the first workflow is used as a new preset workflow.

5. The method according to claim 1, wherein The method further comprises: For a preset simulated business requirement, generating a simulated workflow based on the memory information, wherein the simulated business requirement is a business requirement expected to be generated; In the case that the simulated workflow cannot realize the simulated business requirements, if the business requirements actually obtained are the simulated business requirements, other workflows outside the simulated workflow are generated, and the intelligent agents contained in the generated workflow are controlled to perform tasks according to the generated workflow.

6. The method according to claim 1, characterized in that The controlling the target agent to perform tasks to meet the business requirements includes: Based on the environmental memory information and / or the exclusive memory information corresponding to the target agent, a workflow is generated to implement the business requirement, wherein the workflow represents the tasks to be performed by the target agent and the order in which the tasks are to be performed; Control the target agent to perform the corresponding tasks in the order indicated by the workflow.

7. The method according to any one of claims 1 to 6, characterized in that After controlling the target agent to perform tasks to meet the business needs, the method further includes: Generate a result report indicating the processing result of the business requirement according to a preset report template and the processing result of the business requirement.

8. An electronic device, characterized in that: The electronic device comprises: processor; transceiver; A machine-readable storage medium storing machine-executable instructions that can be executed by the processor, wherein the machine-executable instructions cause the processor to control the first agent to perform the following steps: Obtain pending business requirements in the network environment; Forwarding the business demand to multiple second agents, so that each second agent determines whether it can perform tasks to meet the business demand based on memory information, wherein each second agent is used to perform different tasks in the network environment, and the memory information includes environmental memory information and / or exclusive memory information representing business demands that have been processed in the network environment. Each set of exclusive memory information corresponds to a first agent or a second agent, indicating the business processing capability of the corresponding first agent or second agent and / or its historical task execution process; Determining a target agent for executing the business requirement based on the judgment results of each second agent; Control the target agent to perform tasks to meet the business needs.

9. A network service processing device, characterized in that: The device is applied to a first agent, and the device includes: Demand acquisition module, used to obtain business requirements to be processed in the network environment; a demand forwarding module, configured to forward the business demand to a plurality of second agents, so that each second agent determines whether it is capable of performing tasks to meet the business demand based on memory information, wherein each second agent is respectively configured to perform different tasks in the network environment, the memory information including environmental memory information and / or exclusive memory information representing business demands that have been processed in the network environment, each set of exclusive memory information corresponding to a first agent or a second agent, indicating the business processing capability of the corresponding first agent or second agent and / or its historical task execution process; A target agent determination module, configured to determine a target agent for executing the business requirement based on the judgment results of each second agent; The first task execution module is used to control the target agent to execute tasks to meet the business needs.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, controls the first agent to implement the method steps described in any one of claims 1-7.

Images