Network optimization method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a network optimization method and device, computer equipment, a readable storage medium and a program product. The method comprises the steps of performing intention recognition on a task creation instruction input by a user for a network optimization task, and determining network optimization demand information of the task creation instruction; searching network optimization knowledge data matched with the network optimization demand information, and generating a network optimization plan according to the network optimization knowledge data; the network optimization plan comprises at least one network optimization measure for completing a network optimization task and operation steps in the network optimization measure; according to the network optimization plan, arranging operation tasks of a plurality of RPA intelligent agents to obtain an operation task arrangement result; and according to a task arrangement result, controlling each RPA intelligent agent to execute a corresponding operation task so as to jointly complete a network optimization task. By adopting the method, the network optimization efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a network optimization method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] Wireless network refers to a network that can interconnect various communication devices without wiring. Wireless network technology covers a wide range, and more and more people use various Internet services through wireless networks.

[0003] In the related art, network optimization of wireless networks often takes a lot of time, and when problems occur in the wireless network, it is not possible to quickly determine an optimization solution for the target area, resulting in a problem of low network optimization efficiency in the related art. Summary of the invention

[0004] Based on this, it is necessary to provide a network optimization method, device, computer equipment, computer-readable storage medium and computer program product that can improve network optimization efficiency in response to the above technical problems.

[0005] In a first aspect, the present application provides a network optimization method, the method comprising:

[0006] Performing intent recognition on a task creation instruction input by a user for a network optimization task, and determining network optimization requirement information of the task creation instruction;

[0007] Retrieving network optimization knowledge data matching the network optimization demand information, and generating a network optimization plan based on the network optimization knowledge data; the network optimization plan includes at least one network optimization measure for completing the network optimization task, and operation steps in the network optimization measure;

[0008] According to the network optimization plan, the operation tasks of multiple RPA agents are arranged to obtain the operation task arrangement results; the operation tasks are used for execution by the RPA agents, so that the RPA agents complete at least part of the operation steps in the network optimization measures;

[0009] According to the task scheduling results, each of the RPA agents is controlled to perform corresponding operation tasks to jointly complete the network optimization task.

[0010] In one embodiment, generating a network optimization plan according to the network optimization knowledge data includes:

[0011] Predicting network operating condition prediction information of the network optimization task according to the network optimization demand information; the network operating condition prediction information includes network load and user demand;

[0012] Determining at least one network optimization strategy that meets the network load and the user demand according to the network optimization knowledge data;

[0013] The network optimization plan is generated according to the at least one network optimization strategy.

[0014] In one embodiment, the network optimization demand information includes a network optimization location and a network optimization time, and predicting the network working condition information of the network optimization task according to the network optimization demand information includes:

[0015] Querying historical network performance data matching the network optimization location from a network monitoring system;

[0016] Training a network optimization model for the network optimization task based on the historical network performance data;

[0017] Through the network optimization model, the network load and user demand corresponding to the network optimization location at the network optimization time are output to obtain the network operating condition prediction information.

[0018] In one embodiment, the method further comprises:

[0019] In response to a command input operation performed on a user interaction interface of the RPA agent, obtaining a natural language text input by the command input operation;

[0020] In the case where the natural language text passes the validity verification, the natural language text is determined as the task creation instruction, and the step of performing intent recognition on the task creation instruction input by the user for the network optimization task and determining the network optimization requirement information carried by the task creation instruction is performed;

[0021] The validity verification is used to verify whether the natural language text meets the preset intent recognition requirements.

[0022] In one embodiment, the performing of intent recognition on a task creation instruction input by a user for a network optimization task and determining network optimization requirement information of the task creation instruction includes:

[0023] Inputting the task creation instruction into a pre-trained natural language processing model to obtain an intent recognition result output by the natural language processing model; the intent recognition result includes task intent information, key entity information, and semantic understanding information in the task creation instruction;

[0024] The network optimization requirement information of the task creation instruction is determined according to the task intent information, the key entity information and the semantic understanding information.

[0025] In one embodiment, arranging the operation tasks of multiple RPA agents according to the network optimization plan to obtain the operation task arrangement result includes:

[0026] Converting the network optimization plan into task instructions for the multiple RPA agents; different task instructions correspond to different operation tasks;

[0027] For any of the operation tasks, the operation task is decomposed to obtain at least two subtasks to be executed; each of the subtasks to be executed is independently executed by at least one of the RPA agents;

[0028] According to the characteristic information of each subtask to be executed and the capability information of each RPA agent, each subtask to be executed is assigned to the corresponding RPA agent to obtain the operation task orchestration result.

[0029] In a second aspect, the present application further provides a network optimization device, the device comprising:

[0030] An identification module, used to identify the intention of a task creation instruction input by a user for a network optimization task, and determine network optimization requirement information of the task creation instruction;

[0031] A retrieval module, used to retrieve network optimization knowledge data matching the network optimization demand information, and generate a network optimization plan according to the network optimization knowledge data; the network optimization plan includes at least one network optimization measure for completing the network optimization task, and operation steps in the network optimization measure;

[0032] An orchestration module, configured to orchestrate the operation tasks of multiple RPA agents according to the network optimization plan, and obtain an operation task orchestration result; the operation tasks are used for execution by the RPA agents, so that the RPA agents complete at least part of the operation steps in the network optimization measures;

[0033] The scheduling module is used to control each of the RPA agents to perform corresponding operation tasks according to the task scheduling results, so as to jointly complete the network optimization task.

[0034] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0035] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0036] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements the steps of the above method when executed by a processor.

[0037] The above-mentioned network optimization method, device, computer equipment, computer-readable storage medium and computer program product determine the network optimization requirement information of the task creation instruction by identifying the intention of the task creation instruction input by the user for the network optimization task; and retrieve the network optimization knowledge data matching the network optimization requirement information, and generate a network optimization plan according to the network optimization knowledge data; the network optimization plan includes at least one network optimization measure for completing the network optimization task, and the operation steps in the network optimization measure; then, by arranging the operation tasks of multiple RPA agents according to the network optimization plan, the operation task arrangement result is obtained; the operation task is used for the RPA agent to execute so that the RPA agent completes at least part of the operation steps in the network optimization measure, and according to the task arrangement result, each RPA agent is controlled to perform the corresponding operation task to jointly complete the network optimization task, and through the collaboration between different RPA agents, more efficient and accurate wireless network optimization management is achieved, so that the user does not need to enter a series of complex and cumbersome network optimization operations, but only needs to enter the task creation instruction for the network optimization task to instruct the computer device to combine the network optimization knowledge to perform intelligent and comprehensive network optimization on the network environment to be optimized, significantly improving network performance and user experience, thereby effectively improving network optimization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 An application environment diagram of a network optimization method in an embodiment;

[0040] Figure 2 A schematic diagram of a flow chart of a network optimization method in an embodiment;

[0041] Figure 3 A schematic diagram of a user interface of a network optimization system in one embodiment;

[0042] Figure 4 A schematic diagram of a flow chart of a network optimization method in another embodiment;

[0043] Figure 5A system framework diagram of a network optimization system in one embodiment;

[0044] Figure 6 is a structural block diagram of a network optimization device in one embodiment;

[0045] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] In this application, the implementation of the data capture technical solution involved, when the above embodiments of this application are applied to specific products or technologies, the relevant data collection, use and processing processes should comply with the requirements of national laws and regulations, conform to the principles of legality, legitimacy and necessity, do not involve the acquisition of data types prohibited or restricted by laws and regulations, and will not hinder the normal operation of the target website.

[0048] The network optimization method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the terminal 102 communicates with the server 104 through a network. The data storage system can store data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers.

[0049] The terminal 102 receives a task creation instruction input by a user for a network optimization task; the terminal 102 performs intent recognition on the task creation instruction input by the user for the network optimization task, and determines the network optimization requirement information of the task creation instruction; the terminal 102 retrieves network optimization knowledge data matching the network optimization requirement information from the knowledge base in the server 104, and generates a network optimization plan based on the network optimization knowledge data; the network optimization plan includes at least one network optimization measure for completing the network optimization task, and the operation steps in the network optimization measure; the terminal 102 arranges the operation tasks of multiple RPA agents according to the network optimization plan, and obtains the operation task arrangement results; the operation tasks are used for the RPA agents to execute, so that the RPA agents complete at least part of the operation steps in the network optimization measures; the terminal 102 controls each RPA agent to perform the corresponding operation tasks according to the task arrangement results, so as to jointly complete the network optimization task.

[0050] The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. Head-mounted devices may be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server 104 may be an independent physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides cloud computing services.

[0051] In an exemplary embodiment, Figure 2 As shown, a network optimization method is provided, which is applied to Figure 1 The terminal 102 in the example is used as an example to illustrate, including the following steps S202 to S208. Among them:

[0052] Step S202: performing intent recognition on a task creation instruction input by a user for a network optimization task, and determining network optimization requirement information of the task creation instruction.

[0053] The network optimization task may refer to a task of optimizing network performance to meet specified requirements. In practical applications, the network optimization task may be a task of ensuring the performance of a wireless network in a certain place and during a certain period of time.

[0054] The task creation instruction may refer to an instruction for creating the network optimization task. In practical applications, the task creation instruction may be in the form of natural language. For example, the task creation instruction may refer to "Please help me create a key security task for Park A during the holiday period of Park B and execute it."

[0055] The network optimization requirement information may refer to information used to represent the task objective of the network optimization task. In practical applications, the network optimization requirement information may include but is not limited to the network optimization location, network optimization time, network optimization type, network optimization objective, etc.

[0056] In a specific implementation, the terminal displays a user interface of a network optimization system as a front-end interface for users to interact with the RPA multi-agent system; the terminal can provide an interface for users to interact with RPA agents through the user interface, including command input, result display, and feedback collection. Specifically, the user interface can have the following functions:

[0057] 1) Data input function: collect user queries or instructions through forms, text boxes or other input controls. In this embodiment, the user can enter "Please help me create the key security tasks for Park A during the holiday period of Park B and execute them." in the text box.

[0058] 2) Request verification function: verify validity and ensure that user input meets the system processing requirements.

[0059] 3) Request forwarding function: forwards user requests to the backend processing module, such as the search enhancement generation module.

[0060] 4) Result display function: Receive the results of backend processing and display the execution results to the user on the user interface.

[0061] 5) User interaction function: allows users to perform further operations through the interface, such as viewing detailed reports, adjusting optimization strategies, etc. In order to facilitate the understanding of technical personnel, Figure 3 A schematic diagram of a user interface of a network optimization system is shown by way of example.

[0062] The user can input task creation instructions for network optimization tasks into the terminal through the data input function, such as "Please help me create the key guarantee task for Park A during the holiday period of Park B and execute it." The terminal then obtains the task creation instruction, and can perform intent recognition on the task creation instruction to determine the network optimization requirement information of the task creation instruction, so as to determine the network optimization location, network optimization time, network optimization type, network optimization target, etc. of the network optimization task.

[0063] In actual applications, the terminal can use the user intention understanding module in the network optimization system to realize the intention recognition of the task creation instruction, accurately parse the user's input, and convert it into a task that the network optimization system can understand and execute. In actual applications, the user intention understanding module can have the following functions:

[0064] 1) Input receiving function: receiving text information input by the user through the user interface module;

[0065] 2) Text preprocessing function: clean the user input, including removing irrelevant characters, standardizing text format, etc.

[0066] 3) Intent recognition function: Use the big model to prompt engineering technology and determine the user's main purpose. Continuing with the previous example, "Please help me create a key security task for Park A during the holiday period of Park B and execute it." The intention of this task creation instruction is to "create a key security task."

[0067] 4) Entity extraction function: Using the large model slot recognition capability, identify and extract key information, such as "A park", "B holiday". This information is the specific parameters for task execution.

[0068] 5) Semantic understanding function: Analyze the user's request and understand the deeper meaning behind it. For example, the user may need to ensure the stability of network communication in Park A during holidays.

[0069] Step S204, retrieving network optimization knowledge data that matches the network optimization demand information, and generating a network optimization plan based on the network optimization knowledge data.

[0070] The network optimization plan includes at least one network optimization measure for completing the network optimization task, and the operation steps in the network optimization measure.

[0071] In a specific implementation, after the terminal obtains the network optimization requirement information of the network optimization task, the terminal can retrieve the network optimization knowledge data matching the network optimization requirement information from the network optimization knowledge base by accessing the network optimization knowledge base accumulated in the process of pre-building and updating the wireless network optimization. Continuing with the above example, the network optimization knowledge data can be the knowledge data in the historical key security task records of Park A, the network layout documents, and the special event arrangement data of Holidays B.

[0072] After acquiring the above knowledge data, the terminal can use the document loader and OCR technology to convert the non-text data in the above knowledge data into text format for subsequent processing.

[0073] Then, the terminal can generate a network optimization plan based on the network optimization knowledge data. In practical applications, the terminal can use pre-set prompt words to instruct the large language model to generate a network optimization plan based on the network optimization knowledge data, including at least one network optimization measure for completing the network optimization task and the operation steps in the network optimization measure, such as a key guarantee task plan for the wireless network service of Park A during the holidays of Park B.

[0074] The key security task plan may include a series of network optimization measures for optimizing the wireless network of Park A during the holidays of Park B, as well as the operational steps in the network optimization measures.

[0075] In practical applications, the network optimization knowledge base can be maintained by the knowledge management and learning module in the network optimization system. The knowledge management and learning module is responsible for managing and updating the knowledge base accumulated in the wireless network optimization process, including historical data, lessons learned, best practices, etc., and improving the performance of the system through continuous learning. In practical applications, the knowledge management and learning module can include the following functions:

[0076] 1) Knowledge collection function: The input of this knowledge collection function includes historical optimization data, real-time optimization results and external knowledge sources. Historical optimization data includes records of past network optimization activities, including measures taken, effect evaluation and user feedback. Real-time optimization results include real-time data and results from current network optimization activities. External data sources include external knowledge resources such as industry reports, research papers, technical standards, etc. Before the task is executed, the knowledge collection function collects relevant knowledge from historical data, external research and previous experience.

[0077] 2) Vector knowledge base construction function: store vectorized knowledge in the vector database to optimize retrieval efficiency.

[0078] 3) Retrieval execution function: According to the user's request, the most relevant knowledge fragments or documents are retrieved through the vector database.

[0079] 4) Knowledge application function: Combine the retrieved knowledge to generate a detailed key support task plan, including network optimization strategies, resource allocation, expected risks and response measures.

[0080] 5) Knowledge updating function: Update the content in the knowledge base based on the results of task execution and user feedback, including adding new knowledge points and correcting outdated or inaccurate information.

[0081] Then, the network optimization system can also retrieve network optimization knowledge data that matches the network optimization demand information from the network optimization knowledge base through the search enhancement generation module, and generate a network optimization plan based on the network optimization knowledge data. In actual applications, the working process of the search enhancement generation module can include the following stages:

[0082] 1) Retrieval stage: First, the user's request is received and converted into a searchable query. The query is converted into a high-dimensional vector representation using a pre-trained embedding model. The documents or information fragments that are most similar to the query vector are retrieved in the knowledge base.

[0083] 2) Information fusion stage: Combine the retrieved information with the user’s original request to form a rich context.

[0084] 3) Generation phase: Use the pre-trained large language model and the fused context to generate responses. Generate key support mission plans through the generation model, including specific network optimization measures and execution steps.

[0085] Step S206: Orchestrate the operation tasks of multiple RPA agents according to the network optimization plan and obtain the operation task orchestration results.

[0086] Among them, RPA agent can refer to the smallest execution unit in a multi-agent system (Multi-Agent System, MAS) based on the RPA system.

[0087] A multi-agent system can refer to a system consisting of multiple interacting agents, which can be software programs, robots, sensors, or human participants. Each agent has a certain degree of autonomy and can independently perceive the environment, make decisions, and perform actions.

[0088] Among them, the RPA system, or Robotic Process Automation, is a system that uses software robots (also called automation scripts or "robots") to automatically execute repetitive and rule-based business processes. These software robots can simulate the behavior of human users interacting with digital systems, such as data input, file processing, web page interaction, etc.

[0089] Network optimization RPA system: refers to a system that uses Robotic Process Automation (RPA) technology to optimize network performance.

[0090] It should be noted that the above information is only used to enhance the understanding of the background of the present invention, and therefore may include information that is not known to a person of ordinary skill in the art.

[0091] Among them, the operation task is used for the RPA agent to execute, so that the RPA agent can complete at least part of the operation steps in the network optimization measures.

[0092] After the terminal obtains the network optimization plan for the network optimization task, it can convert the network optimization measures in the network optimization plan and the operation steps in each network optimization measure into task instructions for execution by each RPA agent. Then, the terminal can break down the task instructions into a series of subtask instructions, and assign the subtask instructions to the corresponding RPA agent for execution according to the characteristics of the subtask instructions and the capabilities of the RPA agent.

[0093] Step S208, according to the task scheduling results, control each RPA agent to perform the corresponding operation task to jointly complete the network optimization task.

[0094] In the specific implementation, the terminal can distribute the task instructions of each RPA agent to perform the corresponding operation tasks to the corresponding RPA agent through RPA technology according to the task scheduling results, thereby scheduling the corresponding RPA agent to perform their respective operation tasks, and realize the coordination and cooperation of each RPA agent to jointly complete the network optimization tasks instructed by the user.

[0095] In actual applications, the terminal can also monitor the process of RPA agents executing subtasks to ensure that the tasks are carried out according to the predetermined plan and make timely adjustments to deal with any deviations. If an exception occurs during the execution process, the process engine module will trigger the exception handling mechanism and call the corresponding error handling strategy. In actual applications, the terminal can coordinate multiple RPA agents to jointly perform wireless network optimization tasks through the collaborative RPA agent module in the network optimization system to ensure efficient and accurate completion of the tasks. Among them, the collaborative RPA agent module can include the following functions:

[0096] 1) Automated execution function: The RPA agent executes assigned tasks, such as adjusting base station parameters, optimizing network configuration, etc., while recording key data and status during the execution process.

[0097] 2) Collaborative work function: In tasks that require collaboration among multiple RPA agents, RPA agents exchange information through predefined communication protocols.

[0098] In the process of performing corresponding operation tasks, the RPA agent in the network optimization system often needs to interact with external systems or services. In practical applications, the RPA agent can exchange data and functions through API (application programming interface). Optionally, the network optimization system can maintain the integration of the network optimization system with external systems or services through the integration and API management module. Among them, the integration and API management module can have the following functions:

[0099] 1) API definition and development: Based on the characteristics of wireless network optimization tasks, define API endpoints to receive requests for creating key security tasks. Develop API interfaces that can receive requests containing parameters such as location, time, and event type, and trigger the corresponding RPA task process.

[0100] 2) API Document Management: Create API documents and use them to explain in detail how to call the API, including request parameters, response formats, and examples. Clearly indicate the API call restrictions, error codes, and possible exceptions in the API documents.

[0101] 3) Workflow automation: Use API management tools to automate API calls and RPA workflows, and automatically start RPA tasks after API calls. Design workflows so that after a successful API call, the network optimization system can perform subsequent network optimization tasks.

[0102] In addition, the terminal can also collect feedback information about the system operation through the feedback and iterative learning module of the network optimization system, and use this information for iterative learning to continuously improve the effect of wireless network optimization. In actual applications, the feedback and iterative learning module may include the following functions:

[0103] 1) Feedback collection function: Collect feedback information from multiple sources.

[0104] 2) Iterative learning function: Use machine learning algorithms to adjust and optimize the models in the network optimization system based on feedback information.

[0105] 3) Strategy adjustment function: adjust network optimization strategies and execution plans based on learning results.

[0106] 4) Implement improvement functions: Apply updated strategies and plans to the network optimization system, and execute corresponding improvement measures through the collaborative RPA agent module.

[0107] 5) Knowledge base update function: record the feedback analysis results and improvement measures in the above-mentioned knowledge base.

[0108] In the above network optimization method, the network optimization requirement information of the task creation instruction input by the user for the network optimization task is determined by performing intent recognition; and the network optimization knowledge data matching the network optimization requirement information is retrieved, and a network optimization plan is generated according to the network optimization knowledge data; the network optimization plan includes at least one network optimization measure for completing the network optimization task, and the operation steps in the network optimization measure; then, the operation tasks of multiple RPA agents are arranged according to the network optimization plan to obtain the operation task arrangement result; the operation task is used for the RPA agent to execute, so that the RPA agent completes at least part of the operation steps in the network optimization measure, and according to the task arrangement result, each RPA agent is controlled to perform the corresponding operation task to jointly complete the network optimization task, and through the collaboration between different RPA agents, more efficient and accurate wireless network optimization management is achieved, so that the user does not need to enter a series of complex and cumbersome network optimization operations, but only needs to enter the task creation instruction for the network optimization task to instruct the computer device to combine the network optimization knowledge to perform intelligent and comprehensive network optimization on the network environment to be optimized, significantly improving network performance and user experience, thereby effectively improving network optimization efficiency.

[0109] In an exemplary embodiment, a network optimization plan is generated based on network optimization knowledge data, including: predicting network operating condition prediction information of the network optimization task based on network optimization demand information; determining at least one network optimization strategy that meets network load and user needs based on the network optimization knowledge data; and generating a network optimization plan based on at least one network optimization strategy.

[0110] The network condition prediction information may refer to the network condition data of the network environment optimized by the network optimization task. In practical applications, the network condition prediction information may include the network load and user demand of the network optimization location within the network optimization time.

[0111] In the specific implementation, the terminal generates a network optimization plan based on the network optimization knowledge data, and the terminal can predict the network working condition prediction information of the network optimization task based on the network optimization demand information; specifically, the terminal can query the historical network performance data of the network optimization location based on the network optimization demand information; then, the terminal can use the above-mentioned historical network performance data to predict the network performance data of the network optimization location within the network optimization time, and obtain the network working condition prediction information of the network optimization task.

[0112] Then, the terminal determines at least one network optimization strategy that meets the network load and user needs according to the network optimization knowledge data. Specifically, the terminal can generate at least one network optimization strategy according to the preset template, network working condition prediction information and network optimization knowledge data, such as adjusting base station power, adding temporary base stations, optimizing frequency allocation, etc.

[0113] The terminal can generate a network optimization plan based on at least one network optimization strategy. Specifically, the terminal can use the pre-trained large language model, combined with the fused context, to fuse various network optimization strategies for network optimization tasks and generate responses, that is, generate a network optimization plan through the generation model, such as a key guarantee task plan for the wireless network service of Park A during the holidays of Park B. The key guarantee task plan may include a series of network optimization measures for optimizing the wireless network of Park A during the holidays of Park B, as well as the operation steps in the network optimization measures.

[0114] The technical solution of this embodiment generates a network optimization plan based on network optimization knowledge data, predicts network operating condition prediction information of the network optimization task based on network optimization demand information; determines at least one network optimization strategy that meets network load and user needs based on the network optimization knowledge data; generates a network optimization plan based on at least one network optimization strategy, so that the network optimization plan can more effectively and scientifically optimize the wireless network at the network optimization location indicated by the network optimization task, so as to facilitate the subsequent effective coordination of multiple RPA agents based on the network optimization plan to coordinate work and jointly achieve the task objectives of the network optimization task.

[0115] In an exemplary embodiment, network operating condition information of a network optimization task is predicted based on network optimization demand information, including: querying historical network performance data matching a network optimization location from a network monitoring system; training a network optimization model for the network optimization task based on the historical network performance data; and outputting the network load and user demand corresponding to the network optimization location at the network optimization time through the network optimization model to obtain network operating condition prediction information.

[0116] The network optimization demand information includes the network optimization location and the network optimization time.

[0117] In a specific implementation, when the terminal predicts the network operating condition information of the network optimization task based on the network optimization demand information, the terminal can query the historical network performance data that matches the network optimization location from the network monitoring system. Among them, the historical network performance data may include past network load, signal quality, user satisfaction data, etc. Then, the terminal can perform data analysis on the historical network performance data to identify network performance trends and anomalies. Specifically, the terminal can use historical network performance data to train a network optimization model for network optimization tasks; use the network optimization model to extract network performance trends and anomalies, that is, use the network optimization model to output the network load and user demand corresponding to the network optimization location at the network optimization time, and obtain network operating condition prediction information. Among them, the network optimization model can be a time series model.

[0118] In practical applications, after the terminal queries the historical network performance data that matches the network optimization location in the network monitoring system, the terminal can use the data preprocessing and enhancement module in the network optimization system to preprocess the collected historical network performance data to improve data quality and availability, and at the same time expand the data set through data enhancement technology to enhance the robustness and generalization ability of the model. In practical applications, the terminal can also use the multimodal data processing module in the network optimization system to process and analyze historical network performance data from different sources and formats. The formats can include text, images, sound and video, etc., to support the decision-making process of wireless network optimization. In practical applications, the data preprocessing and enhancement module can include the following functions:

[0119] 1) Data collection function: Collect network performance data of network optimization locations from multiple sources such as network management system, user feedback, and historical optimization records.

[0120] 2) Data cleaning function: Identify and process missing values, outliers, and duplicate records to ensure data consistency and accuracy.

[0121] 3) Data conversion function: Convert data into a format suitable for analysis, such as converting a timestamp to a date and time, or converting categorical data to a numerical representation.

[0122] 4) Data standardization function: Normalize or standardize the data to make them on the same scale for easy model processing.

[0123] 5) Data augmentation: Apply data augmentation techniques, such as randomly rotating, flipping, and scaling images, or applying synonym replacement in text data to expand the dataset.

[0124] 6) Feature engineering function: Create new features based on the optimization goal, such as extracting holidays, weather conditions and other factors that may affect network performance from timestamps.

[0125] 7) Data labeling function: Assign appropriate labels to each sample in the data set, such as network congestion, weak signal, etc., for supervised learning.

[0126] 8) Data splitting function: Divide the data set into training set, validation set and test set to facilitate model training and evaluation.

[0127] In addition, the multimodal data processing module may include the following functions:

[0128] 1) Data collection function: text data includes user feedback, network logs, technical documents, etc.; image data includes base station geographic location maps, network coverage heat maps, etc.; user data includes user voice; video data includes base station monitoring evaluation or field test videos uploaded by users, etc.;

[0129] 2) Data preprocessing function: cleaning, formatting and standardizing data of different modalities for subsequent analysis, including removing irrelevant characters, standardizing text format, image denoising, audio denoising, etc.;

[0130] 3) Feature extraction: Use deep learning models such as convolutional neural networks (CNN) to extract image features and recurrent neural networks (RNN) to extract text and audio features;

[0131] 4) Data fusion function: Integrate features from different modalities to form a unified feature representation for easy analysis and decision-making.

[0132] The terminal can also use the adaptive learning module in the network optimization system to train the network optimization model for the network optimization task based on the historical network performance data; and output the network load and user demand corresponding to the network optimization location at the network optimization time through the network optimization model. In this way, the network configuration and strategy can be automatically adjusted and optimized based on the historical performance data and real-time monitoring data of the network. Among them, the adaptive learning module can include the following functions:

[0133] 1) Data collection function: The module first receives the user's request from the user interface module and collects network performance data from the network monitoring system. Historical network performance data includes past network load, signal quality, user satisfaction data, etc.

[0134] 2) Data analysis function: Analyze input data and identify network performance trends and anomalies. In this embodiment, the data analysis module first receives user requests from the user interface module and collects network performance data from the network monitoring system;

[0135] 3) Model training function: Use machine learning algorithms to train network optimization models. In this embodiment, a deep learning model is trained to predict the network load and user demand corresponding to the target time;

[0136] 4) Strategy generation function: Based on the training results and preset templates, generate optimization strategies, such as adjusting base station power, adding temporary base stations, optimizing frequency allocation, etc.

[0137] The technical solution of this embodiment queries the historical network performance data matching the network optimization location from the network monitoring system, and trains a network optimization model for the network optimization task based on the historical network performance data; through the network optimization model, the network load and user demand corresponding to the network optimization location at the network optimization time are output to obtain network condition prediction information, so that the network optimization model can be used to effectively analyze the network performance trends and anomalies in the historical network performance data matching the network optimization location, so as to use the network performance trends and anomalies to predict the network condition or network performance demand corresponding to the network optimization location at the network optimization time in a reasonable manner.

[0138] In an exemplary embodiment, the method also includes: in response to a command input operation implemented on the user interaction interface of the RPA agent, obtaining the natural language text input by the command input operation; when the natural language text passes the validity verification, determining the natural language text as a task creation instruction, and performing intent recognition on the task creation instruction input by the user for the network optimization task, and determining the network optimization requirement information carried by the task creation instruction; wherein the validity verification is used to verify whether the natural language text meets the preset intent recognition requirements.

[0139] In a specific implementation, when the terminal obtains the task creation instructions input by the user for the network optimization task, the terminal can receive the user's command input operation through the user interaction interface of the RPA agent. For example, the user enters the natural language text "Please help me create the key guarantee task of Park A during the B holiday period and execute it." in the dialog box in the user interaction interface; then, the terminal can verify the validity of the user input, that is, verify the validity of the user input, such as whether the number of words and the amount of information meet the preset conditions, and determine whether the natural language text input by the user meets the preset intention recognition requirements.

[0140] For example, when "Please create a task" is entered in the dialog box in the user interaction interface, the terminal begins to verify the validity of "Please create a task". If it is determined that the natural language "Please create a task" entered by the user at this time cannot clarify the task parameters (such as the task target, task location or task requirements, etc.), the terminal will display a prompt message to prompt the user to re-enter a natural language text that can clarify the network optimization task in the dialog box in the user interaction interface. If the natural language text entered by the user meets the preset intent recognition requirements, the terminal can use the natural language text entered by the user as the task creation instruction input for the network optimization task, and begin to perform intent recognition on the task creation instruction input for the network optimization task.

[0141] The technical solution of this embodiment verifies the validity of the natural language text input by the user before identifying the intention of the task creation instruction input by the user for the network optimization task. Only when the natural language text passes the validity verification, the natural language text is determined as a task creation instruction and transmitted to the back-end for subsequent processing. This can effectively perform preliminary validity judgment on the natural language text input by the user, avoid the back-end processing of too much redundant content, and improve the operating performance of the network optimization system.

[0142] In an exemplary embodiment, intent recognition is performed on a task creation instruction input by a user for a network optimization task to determine network optimization requirement information of the task creation instruction, including: inputting the task creation instruction into a pre-trained natural language processing model to obtain an intent recognition result output by the natural language processing model; the intent recognition result includes task intent information, key entity information, and semantic understanding information in the task creation instruction; and determining the network optimization requirement information of the task creation instruction based on the task intent information, key entity information, and semantic understanding information.

[0143] In the specific implementation, the terminal recognizes the intent of the task creation instruction input by the user for the network optimization task and determines the network optimization requirement information of the task creation instruction. Specifically, the terminal can input the task creation instruction into a pre-trained natural language processing model to obtain the intent recognition result output by the natural language processing model; the intent recognition result includes the task intent information, key entity information and semantic understanding information in the task creation instruction.

[0144] Taking the task creation instruction of "Please help me create a key guarantee task for Park A during Holiday B and execute it." as an example, the task intent information may refer to "create a key guarantee task", and the key entity information may include "Park A" and "Holiday B"; the semantic understanding information may refer to "the user may need to ensure the stability of the network communication of Park A during Holiday B". Then, the terminal can determine the network optimization requirement information of the task creation instruction based on the task intent information, key entity information and semantic understanding information.

[0145] The technical solution of this embodiment, in the process of performing intent recognition on the task creation instruction input by the user for the network optimization task and determining the network optimization requirement information of the task creation instruction, obtains the intent recognition result output by the natural language processing model by inputting the task creation instruction into a pre-trained natural language processing model; the intent recognition result includes the task intent information, key entity information and semantic understanding information in the task creation instruction; based on the task intent information, key entity information and semantic understanding information, the network optimization requirement information of the task creation instruction is determined, thereby realizing intent recognition of the task creation instruction from multiple aspects of task intent, key entities and semantic understanding, and effectively mining the user's network optimization requirement information from the task creation instruction.

[0146] In an exemplary embodiment, according to a network optimization plan, operation tasks of multiple RPA agents are orchestrated to obtain operation task orchestration results, including: converting the network optimization plan into task instructions for multiple RPA agents; different task instructions correspond to different operation tasks; for any operation task, decomposing the operation task to obtain at least two subtasks to be executed; each subtask to be executed is independently executed by at least one RPA agent; according to the characteristic information of each subtask to be executed and the capability information of each RPA agent, each subtask to be executed is assigned to the corresponding RPA agent to obtain the operation task orchestration result.

[0147] In the specific implementation, when the terminal arranges the operation tasks of multiple RPA agents according to the network optimization plan and obtains the operation task arrangement results, the terminal can convert the network optimization plan into task instructions for multiple RPA agents; different task instructions correspond to different operation tasks. Specifically, the terminal can use the planner module in the network optimization system to design the task flow, task execution sequence, required resources, expected goals, etc. using RPA technology, and convert the network optimization plan into task instructions for multiple RPA agents; different task instructions correspond to different operation tasks. Among them, the planner module includes at least the following functions:

[0148] 1) Rule definition function: Based on historical data and business rules, define the execution rules of key security tasks, such as increasing the frequency of network monitoring during the National Day holiday, raising the thresholds of base station operating parameters, etc.

[0149] 2) Task planning function: Use RPA technology to design task processes, including task execution sequence, required resources, and expected goals.

[0150] 3) Strategy generation function: Based on the analysis results, generate optimization strategies, such as adjusting base station configuration, adding temporary base stations, optimizing frequency allocation, etc., and form a detailed network optimization plan.

[0151] 4) Task scheduling function: Convert the generated plans and strategies into executable task instructions, and dispatch them to the corresponding RPA agents for execution through RPA technology.

[0152] Then, the terminal can send the task instructions of the RPA agent to the process engine module in the network optimization system through the planner module in the network optimization system; the process engine module decomposes the task instructions corresponding to different operation tasks to obtain at least two subtasks to be executed; each subtask to be executed is independently executed by at least one RPA agent; then, the process engine module can assign each subtask to be executed to the corresponding RPA agent according to the characteristic information of each subtask to be executed and the capability information of each RPA agent, and obtain the operation task scheduling result. In actual applications, the process engine module is responsible for managing and executing the automated process of wireless network optimization tasks; the process engine module can include the following functions:

[0153] 1) Task reception and analysis function: Receive task instructions from the planner module, analyze task requirements, and determine network optimization goals and execution details.

[0154] 2) Task decomposition function: Decompose complex tasks into a series of executable subtasks, each of which can be independently executed by one or more RPA agents.

[0155] 3) RPA agent allocation function: Assign tasks to suitable agents for execution based on the characteristics of subtasks and the capabilities of RPA agents.

[0156] 4) Execution monitoring RPA: Monitor the process of RPA agents executing subtasks to ensure that tasks are carried out according to the predetermined plan and make timely adjustments to address any deviations.

[0157] 5) Exception handling RPA: If an exception occurs during the execution process, the process engine module will trigger the exception handling mechanism and call the corresponding error handling strategy.

[0158] 6) Task coordination RPA: For tasks that require collaboration among multiple agents, the process engine module is responsible for coordinating the work between different RPA agents to jointly complete network optimization tasks.

[0159] The technical solution of this embodiment is to convert the network optimization plan into task instructions for multiple RPA agents; different task instructions correspond to different operation tasks; for any operation task, the operation task is decomposed to obtain at least two subtasks to be executed; each subtask to be executed is independently executed by at least one RPA agent; according to the characteristic information of each subtask to be executed and the capability information of each RPA agent, each subtask to be executed is assigned to the corresponding RPA agent to obtain the operation task scheduling result, so that the operation task scheduling result can be used to orderly and efficiently schedule the collaboration between RPA agents in the multi-agent system, realize more efficient and accurate wireless network optimization management, intelligently complete the network optimization tasks issued by users, and effectively improve the network optimization efficiency.

[0160] In another embodiment, Figure 4 As shown, a network optimization method is provided, which is applied to Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0161] Step S402: Obtain a task creation instruction input by a user for a network optimization task.

[0162] Step S404, input the task creation instruction into the pre-trained natural language processing model to obtain the intent recognition result output by the natural language processing model; the intent recognition result includes the task intent information, key entity information and semantic understanding information in the task creation instruction.

[0163] Step S406, determining the network optimization requirement information of the task creation instruction according to the task intent information, key entity information and semantic understanding information; the network optimization requirement information includes the network optimization location and network optimization time.

[0164] Step S408, querying the network monitoring system for historical network performance data that matches the network optimization location.

[0165] Step S410: training a network optimization model for a network optimization task based on historical network performance data.

[0166] Step S412, outputting the network load and user demand corresponding to the network optimization location at the network optimization time through the network optimization model, and obtaining network working condition prediction information; the network working condition prediction information includes network load and user demand.

[0167] Step S414, retrieving network optimization knowledge data that matches the network optimization demand information, and determining at least one network optimization strategy that meets the network load and user demand based on the network optimization knowledge data.

[0168] Step S416, generating a network optimization plan according to at least one network optimization strategy; the network optimization plan includes at least one network optimization measure for completing the network optimization task, and the operation steps in the network optimization measure.

[0169] Step S418, converting the network optimization plan into task instructions for multiple RPA agents; different task instructions correspond to different operation tasks; the operation tasks are used for the RPA agents to execute, so that the RPA agents can complete at least part of the operation steps in the network optimization measures.

[0170] Step S420: for any operation task, decompose the operation task to obtain at least two subtasks to be executed; each subtask to be executed is independently executed by at least one RPA agent.

[0171] Step S422, according to the characteristic information of each subtask to be executed and the capability information of each RPA agent, each subtask to be executed is assigned to the corresponding RPA agent to obtain the operation task scheduling result.

[0172] Step S424, according to the task scheduling results, control each RPA agent to perform the corresponding operation task to jointly complete the network optimization task.

[0173] It should be noted that the specific limitations of the above steps can refer to the specific limitations of a network optimization method above.

[0174] In order to facilitate the understanding of those skilled in the art, Figure 5A system framework diagram of a network optimization system is exemplarily provided; wherein, the network optimization system may include a user interface module, a user intent understanding module, a data preprocessing and enhancement module, a multimodal data processing module, a knowledge management and learning module, an adaptive learning module, an integration and API management module, a search enhancement generation module, a planner module, an assisted RPA agent module, and a process engine module; wherein the user interface module may obtain a task creation instruction input by a user for a network optimization task; then, the user interface module sends the task creation instruction to the user intent understanding module; the user intent understanding module performs intent recognition on the task creation instruction input by the user for the network optimization task, and determines the network optimization requirement information of the task creation instruction.

[0175] Then, based on the network optimization demand information, the terminal obtains historical network performance data that matches the network optimization location from the network monitoring system. A series of data processing is performed on the historical network performance data through the data preprocessing and enhancement module and the multimodal data processing module; and through the adaptive learning module, the network optimization model for the network optimization task is trained based on the historical network performance data; through the network optimization model, the network load and user demand corresponding to the network optimization location at the network optimization time are output.

[0176] Through the knowledge management and learning module, as well as the search enhancement generation module, network optimization knowledge data matching the network optimization demand information is retrieved, and at least one network optimization strategy that meets the network load and user needs is determined based on the network optimization knowledge data; based on the at least one network optimization strategy, a network optimization plan is generated.

[0177] The planner module and the process engine module are responsible for converting the agent decisions specified in the network optimization plan into specific operation steps. The collaborative RPA agent module, the process engine module, the knowledge management and learning module, the integration and API management module, and the feedback and iterative learning module jointly ensure the stability, security, performance and continuous improvement of the system. The design of the entire network optimization system ensures that the interfaces between modules are standardized, the data flow is clear, and the control process is reasonable, so as to achieve efficient automation and intelligent operation. It should be noted that the specific definitions of the above-mentioned user interface module, user intent understanding module, data preprocessing and enhancement module, multimodal data processing module, knowledge management and learning module, adaptive learning module, integration and API management module, search enhancement generation module, planner module, assisted RPA agent module, and process engine module can refer to the specific definition of a network optimization method above, which will not be repeated here.

[0178] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0179] Based on the same inventive concept, the embodiment of the present application also provides a network optimization device for implementing the network optimization method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more network optimization device embodiments provided below can refer to the limitations on the network optimization method above, and will not be repeated here.

[0180] In an exemplary embodiment, Figure 6 As shown, a network optimization device is provided, comprising:

[0181] An identification module 610 is used to identify the intention of a task creation instruction input by a user for a network optimization task, and determine network optimization requirement information of the task creation instruction;

[0182] A retrieval module 620 is used to retrieve network optimization knowledge data that matches the network optimization demand information, and generate a network optimization plan based on the network optimization knowledge data; the network optimization plan includes at least one network optimization measure for completing the network optimization task, and the operation steps in the network optimization measure;

[0183] The orchestration module 630 is used to orchestrate the operation tasks of multiple RPA agents according to the network optimization plan, and obtain the operation task orchestration results; the operation tasks are used for the RPA agents to execute, so that the RPA agents complete at least part of the operation steps in the network optimization measures;

[0184] The scheduling module 640 is used to control each of the RPA agents to perform corresponding operation tasks according to the task scheduling results, so as to jointly complete the network optimization task.

[0185] In one embodiment, the retrieval module 620 is specifically used to predict the network condition prediction information of the network optimization task based on the network optimization demand information; the network condition prediction information includes network load and user demand; based on the network optimization knowledge data, determine at least one network optimization strategy that meets the network load and the user demand; based on the at least one network optimization strategy, generate the network optimization plan.

[0186] In one embodiment, the retrieval module 620 is specifically used for the network optimization demand information including the network optimization location and the network optimization time, and predicting the network working condition information of the network optimization task based on the network optimization demand information, including: querying historical network performance data matching the network optimization location from the network monitoring system; training a network optimization model for the network optimization task based on the historical network performance data; and outputting the network load and user demand corresponding to the network optimization location at the network optimization time through the network optimization model to obtain the network working condition prediction information.

[0187] In one of the embodiments, the device also includes: a verification module, which is used to obtain the natural language text input by the command input operation in response to the command input operation implemented on the user interaction interface of the RPA agent; when the natural language text passes the validity verification, the natural language text is determined as the task creation instruction, and the step of performing intent recognition on the task creation instruction input by the user for the network optimization task and determining the network optimization requirement information carried by the task creation instruction is executed; wherein the validity verification is used to verify whether the natural language text meets the preset intent recognition requirements.

[0188] In one embodiment, the recognition module 610 is specifically used to input the task creation instruction into a pre-trained natural language processing model to obtain an intent recognition result output by the natural language processing model; the intent recognition result includes task intent information, key entity information and semantic understanding information in the task creation instruction; and determine the network optimization requirement information of the task creation instruction based on the task intent information, the key entity information and the semantic understanding information.

[0189] In one embodiment, the orchestration module 630 is specifically used to convert the network optimization plan into task instructions for the multiple RPA agents; different task instructions correspond to different operation tasks; for any operation task, the operation task is decomposed to obtain at least two subtasks to be executed; each subtask to be executed is independently executed by at least one RPA agent; based on the characteristic information of each subtask to be executed and the capability information of each RPA agent, each subtask to be executed is assigned to the corresponding RPA agent to obtain the operation task orchestration result.

[0190] Each module in the above network optimization device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module above.

[0191] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 7 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a network optimization method is implemented.

[0192] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 7As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (Near Field Communication, NFC) or other technologies. When the computer program is executed by the processor, a network optimization method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0193] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0194] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above-mentioned network optimization method. The steps of the network optimization method here may be the steps of the network optimization method in each of the above-mentioned embodiments.

[0195] In one embodiment, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the above-mentioned network optimization method. The steps of the network optimization method here can be the steps of the network optimization method in each of the above-mentioned embodiments.

[0196] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, causes the processor to execute the steps of the above-mentioned network optimization method. The steps of the network optimization method here may be the steps of the network optimization method in each of the above-mentioned embodiments.

[0197] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0198] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0199] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0200] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A network optimization method, characterized in that: The method comprises: Performing intent recognition on a task creation instruction input by a user for a network optimization task, and determining network optimization requirement information of the task creation instruction; Retrieving network optimization knowledge data matching the network optimization demand information, and generating a network optimization plan based on the network optimization knowledge data; the network optimization plan includes at least one network optimization measure for completing the network optimization task, and operation steps in the network optimization measure; According to the network optimization plan, the operation tasks of multiple RPA agents are arranged to obtain the operation task arrangement results; the operation tasks are used for execution by the RPA agents, so that the RPA agents complete at least part of the operation steps in the network optimization measures; According to the task scheduling results, each of the RPA agents is controlled to perform corresponding operation tasks to jointly complete the network optimization task.

2. The method according to claim 1, characterized in that Generating a network optimization plan according to the network optimization knowledge data includes: Predicting network operating condition prediction information of the network optimization task according to the network optimization demand information; the network operating condition prediction information includes network load and user demand; Determining at least one network optimization strategy that meets the network load and the user demand according to the network optimization knowledge data; The network optimization plan is generated according to the at least one network optimization strategy.

3. The method according to claim 2, characterized in that The network optimization demand information includes a network optimization location and a network optimization time. The network working condition information of the network optimization task is predicted based on the network optimization demand information, including: Querying historical network performance data matching the network optimization location from a network monitoring system; Training a network optimization model for the network optimization task based on the historical network performance data; Through the network optimization model, the network load and user demand corresponding to the network optimization location at the network optimization time are output to obtain the network operating condition prediction information.

4. The method according to claim 1, characterized in that: The method further comprises: In response to a command input operation performed on a user interaction interface of the RPA agent, obtaining a natural language text input by the command input operation; In the case where the natural language text passes the validity verification, the natural language text is determined as the task creation instruction, and the step of performing intent recognition on the task creation instruction input by the user for the network optimization task and determining the network optimization requirement information carried by the task creation instruction is performed; The validity verification is used to verify whether the natural language text meets the preset intent recognition requirements.

5. The method according to claim 1, characterized in that The performing of intention recognition on a task creation instruction input by a user for a network optimization task and determining network optimization requirement information of the task creation instruction includes: Inputting the task creation instruction into a pre-trained natural language processing model to obtain an intent recognition result output by the natural language processing model; the intent recognition result includes task intent information, key entity information, and semantic understanding information in the task creation instruction; The network optimization requirement information of the task creation instruction is determined according to the task intent information, the key entity information and the semantic understanding information.

6. The method according to any one of claims 1 to 5, characterized in that: The step of arranging the operation tasks of multiple RPA agents according to the network optimization plan and obtaining the operation task arrangement result includes: Converting the network optimization plan into task instructions for the multiple RPA agents; different task instructions correspond to different operation tasks; For any of the operation tasks, the operation task is decomposed to obtain at least two subtasks to be executed; each of the subtasks to be executed is independently executed by at least one of the RPA agents; According to the characteristic information of each subtask to be executed and the capability information of each RPA agent, each subtask to be executed is assigned to the corresponding RPA agent to obtain the operation task orchestration result.

7. A network optimization device, characterized in that: The device comprises: An identification module, used to identify the intention of a task creation instruction input by a user for a network optimization task, and determine network optimization requirement information of the task creation instruction; A retrieval module, used to retrieve network optimization knowledge data matching the network optimization demand information, and generate a network optimization plan according to the network optimization knowledge data; the network optimization plan includes at least one network optimization measure for completing the network optimization task, and operation steps in the network optimization measure; An orchestration module, configured to orchestrate the operation tasks of multiple RPA agents according to the network optimization plan, and obtain an operation task orchestration result; the operation tasks are used for execution by the RPA agents, so that the RPA agents complete at least part of the operation steps in the network optimization measures; The scheduling module is used to control each of the RPA agents to perform corresponding operation tasks according to the task scheduling results, so as to jointly complete the network optimization task.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images