System operation and maintenance strategy determination method and device, equipment, storage medium and product

Abstract

The invention relates to a system operation and maintenance strategy determination method and device, equipment, a storage medium and a product. The method comprises the steps of obtaining problem description data of a to-be-detected system; inputting the problem description data into a trained operation and maintenance strategy determination model to obtain an operation and maintenance strategy corresponding to the problem description data; wherein the operation and maintenance strategy determination model is trained in the following mode: taking a pre-trained generative pre-training model as the operation and maintenance strategy determination model; and obtaining historical operation and maintenance data of the to-be-detected system, and according to the historical operation and maintenance data, performing fine adjustment on the network parameters of the operation and maintenance strategy determination model. By adopting the method, the determination speed of the operation and maintenance strategy can be improved, and the system operation and maintenance efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of information technology operation and maintenance technology, and in particular to a method, device, equipment, storage medium and product for determining a system operation and maintenance strategy. Background Art

[0002] In the field of traditional information technology operation and maintenance, system failure or exception handling is a major challenge for operation and maintenance engineers.

[0003] When faced with a problem, engineers need to spend a lot of time and energy to consult numerous documents, logs, and reference materials, or consult domain experts to find a solution. This information is relatively scattered, and the retrieval and integration process is cumbersome and time-consuming. In addition, the same problem may be expressed differently in different documents, making problem location complex and easy to misunderstand.

[0004] Therefore, in an emergency, over-reliance on manual work will not only prolong response time, but may also threaten business continuity and cause adverse effects. Therefore, there is an urgent need for a method to increase the speed of determining operation and maintenance strategies to improve system operation and maintenance efficiency. Summary of the invention

[0005] Based on this, it is necessary to provide a system operation and maintenance strategy determination method, device, equipment, storage medium and product that can improve the determination speed of operation and maintenance strategies and thereby improve the system operation and maintenance efficiency in response to the above technical problems.

[0006] In a first aspect, the present application provides a method for determining a system operation and maintenance strategy, comprising:

[0007] Obtain problem description data of the system to be tested;

[0008] Input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data;

[0009] The operation and maintenance strategy determination model is trained in the following ways:

[0010] Use the pre-trained generative pre-trained model as the operation and maintenance strategy determination model;

[0011] Obtain the historical operation and maintenance data of the system to be tested, and fine-tune the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data.

[0012] In one embodiment, obtaining historical operation and maintenance data of the system to be detected includes:

[0013] Obtain historical faults that occurred during the historical use of the system to be tested, as well as historical operation and maintenance records corresponding to each historical fault;

[0014] Obtain historical log data recorded during historical use of the system to be tested;

[0015] The server system information of the system to be detected, the historical operation and maintenance records and the historical log data are used as the historical operation and maintenance data of the system to be detected.

[0016] In one embodiment, fine-tuning the network parameters of the operation and maintenance strategy determination model based on historical operation and maintenance data includes:

[0017] Extract problem features from the historical operation and maintenance data of the system to be tested to obtain sample problem description data;

[0018] Input the sample problem description data into the operation and maintenance strategy determination model to obtain the operation and maintenance strategy prediction result corresponding to the sample problem description data;

[0019] According to the operation and maintenance strategy prediction results and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data, the network parameters of the operation and maintenance strategy determination model are fine-tuned.

[0020] In one embodiment, problem features are extracted from historical operation and maintenance data of the system to be detected to obtain sample problem description data, including:

[0021] Perform text segmentation processing on the historical operation and maintenance data of the detection system to obtain the text segmentation results;

[0022] Construct a text vocabulary based on the frequency of occurrence of each word in the text segmentation results;

[0023] Based on the information of each word in the text vocabulary, sample question description data is determined.

[0024] In one embodiment, text segmentation processing is performed on the historical operation and maintenance data of the system to be detected to obtain text segmentation results, including:

[0025] Perform data preprocessing on the historical operation and maintenance data of the detection system to obtain preprocessed operation and maintenance information;

[0026] Wherein, the preprocessing includes at least one of text consistency processing, text deduplication processing and text standardization processing;

[0027] Perform text segmentation on the preprocessed operation and maintenance information to obtain text segmentation results.

[0028] In one embodiment, the method further comprises:

[0029] Evaluate the fine-tuned operation and maintenance strategy determination model;

[0030] The evaluation process includes:

[0031] Determine the number of true positives, false positives, and false negatives based on the operation and maintenance strategy prediction results corresponding to the sample problem description data and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data;

[0032] Evaluate the fine-tuned O&M strategy determination model based on the number of true positives, false positives, and false negatives.

[0033] Accordingly, the method further includes:

[0034] When the evaluation result indicates that the fine-tuned operation and maintenance strategy determination model meets the preset requirements, a trained operation and maintenance strategy determination model is obtained.

[0035] In a second aspect, the present application also provides a system operation and maintenance strategy determination device, including:

[0036] A data acquisition module is used to obtain problem description data of the system to be tested;

[0037] A strategy determination module is used to input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data;

[0038] The operation and maintenance strategy determination model is trained in the following ways:

[0039] A model determination module is used to use the pre-trained generative pre-trained model as an operation and maintenance strategy determination model;

[0040] The model fine-tuning module is used to obtain the historical operation and maintenance data of the system to be tested, and fine-tune the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data.

[0041] 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 when the processor executes the computer program, the following steps are implemented:

[0042] Obtain problem description data of the system to be tested;

[0043] Input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data;

[0044] The operation and maintenance strategy determination model is trained in the following ways:

[0045] Use the pre-trained generative pre-trained model as the operation and maintenance strategy determination model;

[0046] Obtain the historical operation and maintenance data of the system to be tested, and fine-tune the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data.

[0047] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0048] Obtain problem description data of the system to be tested;

[0049] Input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data;

[0050] The operation and maintenance strategy determination model is trained in the following ways:

[0051] Use the pre-trained generative pre-trained model as the operation and maintenance strategy determination model;

[0052] Obtain the historical operation and maintenance data of the system to be tested, and fine-tune the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data.

[0053] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0054] Obtain problem description data of the system to be tested;

[0055] Input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data;

[0056] The operation and maintenance strategy determination model is trained in the following ways:

[0057] Use the pre-trained generative pre-trained model as the operation and maintenance strategy determination model;

[0058] Obtain the historical operation and maintenance data of the system to be tested, and fine-tune the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data.

[0059] The above-mentioned system operation and maintenance strategy determination method, device, equipment, storage medium and product pre-train a generative pre-trained operation and maintenance strategy determination model, and based on the operation and maintenance strategy determination model, process the problem description data of the system to be detected to obtain the operation and maintenance strategy corresponding to the problem description data. Compared with the traditional technology, when the system to be detected has a problem and needs operation and maintenance, the operation and maintenance engineer obtains relevant information of the system to be detected from different channels to determine the operation and maintenance strategy. The system operation and maintenance strategy determination method provided in this application can significantly improve the speed of determining the operation and maintenance strategy, thereby achieving the effect of improving the system operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] 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.

[0061] Figure 1 A flowchart of a method for determining a system operation and maintenance strategy in one embodiment;

[0062] Figure 2 A flowchart of the steps of obtaining historical operation and maintenance data in one embodiment;

[0063] Figure 3 A schematic diagram of a flow chart of a network parameter fine-tuning step in an embodiment;

[0064] Figure 4 A schematic diagram of a flow chart of a model evaluation step in one embodiment;

[0065] Figure 5A A flowchart of a method for determining a system operation and maintenance strategy in another embodiment;

[0066] Figure 5B A flowchart of a method for determining a system operation and maintenance strategy in another embodiment;

[0067] Figure 6 It is a structural block diagram of a system operation and maintenance strategy determination device in one embodiment;

[0068] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0069] 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.

[0070] Before introducing the method for determining the system operation and maintenance strategy provided by the present application, it should be noted that in traditional information technology operation and maintenance (IT Operations), once a system failure or anomaly occurs, operation and maintenance engineers usually need to spend a lot of time and energy to consult a large number of documents, logs and reference materials, or consult experts to find solutions. However, this key information is often scattered in various resources, such as technical documents, configuration files, monitoring system outputs, event logs and reports, making the retrieval and integration of information cumbersome and time-consuming. In addition, the same problem or information may appear in different forms in different documents, which not only increases the complexity of problem location, but may also lead to misunderstandings. Solving complex operation and maintenance problems requires in-depth knowledge of systems, networks, applications and related technologies, which limits the effective participation of non-professional or junior personnel. Over-reliance on manual retrieval and problem solving not only prolongs response time, but may also have an adverse impact on business continuity in an emergency.

[0071] With the development of Generative Pre-trained Transformer (GPT) technology, GPT has powerful language understanding and generation capabilities and efficient information processing capabilities, and can be applied to different application fields. Therefore, GPT can be seen in more and more scenarios. In order to solve the problem of low speed of determining the above-mentioned operation and maintenance strategy, the present invention proposes a system operation and maintenance strategy determination method combined with GPT.

[0072] In one embodiment, Figure 1 As shown, a method for determining a system operation and maintenance strategy is provided. This embodiment uses the method applied to a terminal as an example for illustration. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0073] S110, obtaining problem description data of the system to be detected.

[0074] The system to be tested is a system that may have problems at present. For example, the system to be tested may be an information management system. The problem description data is used to describe the problems currently exhibited by the system to be tested, for example, the system to be tested is stuck, or the system to be tested cannot submit data.

[0075] Exemplarily, the problem description data may be discovered by a user and uploaded by the user through a client. The problem description data may also be obtained through automatic detection by a terminal. This application does not impose any restrictions on the source of the problem description data or the form of the problem description data.

[0076] S120, input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data.

[0077] The operation and maintenance strategy determination model is trained in the following ways:

[0078] The pre-trained generative pre-trained model is used as the operation and maintenance strategy determination model; the historical operation and maintenance data of the system to be tested is obtained, and the network parameters of the operation and maintenance strategy determination model are fine-tuned based on the historical operation and maintenance data.

[0079] The historical operation and maintenance data includes historical problems that occurred during the use of the system to be tested, historical solutions to the historical problems, and system parameters of the system to be tested. The historical operation and maintenance data can be used as a basis for determining the operation and maintenance strategy.

[0080] In the above-mentioned system operation and maintenance strategy determination method, a generative pre-trained operation and maintenance strategy determination model is pre-trained, and based on the operation and maintenance strategy determination model, the problem description data of the system to be detected is processed to obtain the operation and maintenance strategy corresponding to the problem description data. Compared with the traditional technology, when a problem occurs in the system to be detected and operation and maintenance is required, the operation and maintenance engineer obtains relevant information of the system to be detected from different channels to determine the operation and maintenance strategy. The system operation and maintenance strategy determination method provided in this application can significantly improve the speed of determining the operation and maintenance strategy, thereby achieving the effect of improving the system operation and maintenance efficiency.

[0081] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the process of obtaining the historical operation and maintenance data of the system to be detected is refined.

[0082] See also Figure 2 The steps for obtaining historical operation and maintenance data shown include:

[0083] S210, obtaining historical faults that occurred during historical use of the system to be detected, and historical operation and maintenance records corresponding to each historical fault.

[0084] For example, in this embodiment, historical faults that occurred during historical use of the system to be detected, as well as historical operations and maintenance corresponding to each historical fault, can be stored in a historical operation and maintenance database, and the historical operation and maintenance database can be used as a historical operation and maintenance manual, wherein the historical faults in the historical operation and maintenance manual are stored in association with the corresponding historical operation and maintenance records.

[0085] It should be noted that the operation and maintenance manual is a document used by an organization or team to record and convey operation and maintenance related information, processes and best practices. It is a concentration of the operation and maintenance team's experience. Its goal is to ensure the reliability, availability and stability of systems and services, as well as to speed up the diagnosis and resolution of problems. All operation and maintenance experience-related materials are tracked and managed through Git to ensure the consistency and traceability of the environment. The contents of the operation and maintenance manual include but are not limited to: a system overview describing the organization's core systems and services, the main functions and uses of each system, and a description of the system's architecture or topology diagram. Provide detailed system and service documents for each system and service, including system architecture and components, deployment and configuration information, monitoring and alarm settings, troubleshooting guides, performance tuning recommendations, security policies and best practices, common problems and solutions, etc.

[0086] In some scenarios, the operation and maintenance manual also includes operation and maintenance tools, such as operation and maintenance tools and scripts used in operation and maintenance work, including listing the tools and scripts used for monitoring, automation and maintenance, and providing the purpose of the tools and scripts, document links and example usage. The operation and maintenance manual also includes emergency plans, including describing the response process in emergency situations, providing disaster recovery and recovery plans, and determining backup strategies for key systems and data. Documentation that describes security measures and strategies and ensures compliance with compliance standards and regulations.

[0087] In addition, the operation and maintenance manual also includes the following: Describes how to perform capacity planning and expansion, and provides recommendations for performance optimization. Defines the update strategy for systems and services, provides maintenance windows and plans, and documents that explain how to perform emergency repairs and upgrades. Provides troubleshooting processes and guidelines, lists troubleshooting and problem solutions for common problems and solutions. Describes the logging strategy, explains how to access and analyze log data, and provides monitoring and alarm setting parameter instructions. References such as other related documents, resources, and links. Other materials, including appendices such as glossaries and abbreviation glossaries, record the version history and update instructions of the operation and maintenance manual.

[0088] S220, obtaining historical log data recorded during historical use of the system to be detected.

[0089] Among them, historical log data includes all usage records recorded by the system to be tested during historical use, including user activities, transactions, errors, and exceptions. Based on historical log data, operation and maintenance personnel can ensure stable operation of applications, discover and solve problems in a timely manner, and meet security and compliance requirements.

[0090] Exemplarily, historical log data can be stored in JSON structure, with fields including but not limited to: timestamp, log level, log message, event identifier, user identifier, request / response information, stack trace, and environment information. In some scenarios, log collection tools can also be used to centrally store logs generated by related applications in the same location to make it easier to manage and query historical log data.

[0091] S230, taking the server system information of the system to be detected, the historical operation and maintenance records and the historical log data as the historical operation and maintenance data of the system to be detected.

[0092] Among them, server system information is the basis of operation and maintenance work, which can provide data for controlling, diagnosing, maintaining and optimizing servers. By actively managing and analyzing server system information, operation and maintenance personnel can ensure the high availability, stability and security of the server to meet business needs.

[0093] Exemplarily, in this embodiment, the host resource utilization is collected through the system monitoring and alarm tools, including but not limited to the central processing unit (CPU), memory, disk utilization, network traffic, application performance indicators, network indicators, database performance and health status of hardware devices. Among them, application performance indicators include request response time, error rate, etc. Network indicators include the number of Transmission Control Protocol (TCP) connections, Hypertext Transfer Protocol (HTTP) request rate, Domain Name System (DNS) query latency. Database performance includes query execution time, connection pool usage, cache hit rate, etc. The health status of hardware devices includes temperature, fan speed, power supply, etc.

[0094] In this embodiment, the server system information, historical operation and maintenance records and historical log data are taken together as the historical operation and maintenance data of the system to be detected.

[0095] In the above embodiment, different types of data included in the historical operation and maintenance data of the system to be detected are given, making the data content of the historical operation and maintenance data clearer. Furthermore, the data of the system to be detected in different dimensions are all used as historical operation and maintenance data, which facilitates the model training of the operation and maintenance strategy determination model.

[0096] Furthermore, the present application can also use the network platform data as the historical operation and maintenance data of the system to be tested. The network platform data is helpful in solving technical problems, sharing knowledge, keeping learning and establishing connections, and is a valuable resource in operation and maintenance work.

[0097] It should be noted that in order to maintain the timeliness and effectiveness of historical operation and maintenance data, in this embodiment, the historical operation and maintenance data can be updated and maintained regularly. For example, newly generated problems and the operation and maintenance strategies corresponding to the problems are added to the historical operation and maintenance database; incorrect or outdated information in the historical operation and maintenance database is modified and corrected; information that is no longer applicable to the system to be tested or inaccurate is removed from the historical operation and maintenance database.

[0098] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the process of fine-tuning the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data is described in detail.

[0099] See also Figure 3 The network parameter fine-tuning steps shown include:

[0100] S310, extracting problem features from historical operation and maintenance data of the system to be detected to obtain sample problem description data.

[0101] In an optional implementation, the historical operation and maintenance data of the system to be tested can be input into the feature extraction model to obtain the problem features corresponding to each data in the historical operation and maintenance data. Furthermore, according to the problem features corresponding to each data, at least one data is combined to obtain sample problem description data. It should be noted that the above-mentioned feature extraction model can be a common feature extraction network, and this application does not impose any restrictions on the model construction and training process of the feature extraction model.

[0102] In an optional implementation, the historical operation and maintenance data of the system to be tested may be subjected to text segmentation processing to obtain text segmentation results; a text vocabulary may be constructed based on the frequency of occurrence of each word in the text segmentation results; and sample problem description data may be determined based on the information of each word in the text vocabulary.

[0103] Exemplarily, the process of determining the text segmentation result may be as follows: performing data preprocessing on the historical operation and maintenance data of the system to be detected to obtain preprocessed operation and maintenance information; wherein the preprocessing includes at least one of text consistency processing, text deduplication processing and text standardization processing; performing text segmentation processing on the preprocessed operation and maintenance information to obtain a text segmentation result.

[0104] Optionally, in this embodiment, a natural language processing toolkit may be used to pre-process the text data to remove special characters, punctuation marks, HyperText Markup Language (HTML) tags, and other noises that may interfere with model learning.

[0105] Optionally, in this embodiment, a text segmentation tool can be used to segment English and Chinese texts and convert the texts into a format that the model can understand. In addition, the text is further normalized by converting all letters to lowercase and removing stop words to eliminate the impact of natural language variability on model training.

[0106] Furthermore, the occurrence frequency of each word in each text segmentation result is determined, and the words are arranged in descending order of frequency to construct a vocabulary. Then, based on the information of each word in the text vocabulary, sample question description data is determined. For example, the vocabulary can be input into a pre-trained question generation model to obtain sample question description data.

[0107] S320, inputting the sample problem description data into the operation and maintenance strategy determination model to obtain the operation and maintenance strategy prediction result corresponding to the sample problem description data.

[0108] It should be noted that in this step, the operation and maintenance strategy determination model can vectorize the sample problem description data to obtain the word vector of the sample problem description data, and the operation and maintenance strategy determination model can further process the word vector.

[0109] Optionally, in order to reduce the data processing pressure of the operation and maintenance strategy determination model, in this embodiment, the sample problem description data can also be first input into the word vector determination model to obtain the word vector corresponding to the sample problem description data, and then the word vector can be input into the operation and maintenance strategy determination model to obtain the operation and maintenance strategy prediction result corresponding to the sample problem description data.

[0110] S330, fine-tuning the network parameters of the operation and maintenance strategy determination model according to the operation and maintenance strategy prediction results and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data.

[0111] In the above embodiment, problem features are extracted from the historical operation and maintenance data of the system to be detected to obtain sample problem description data, so that the sample problem description data can cover various problems in the historical operation and maintenance data. Rich sample problem description data can make the training effect of the operation and maintenance strategy determination model better.

[0112] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the steps after fine-tuning the operation and maintenance strategy determination model are introduced, specifically, including evaluating the fine-tuned operation and maintenance strategy determination model.

[0113] See also Figure 4 The model evaluation steps shown include:

[0114] S410, determining the number of true positive examples, the number of false positive examples, and the number of false negative examples according to the operation and maintenance strategy prediction results corresponding to the sample problem description data and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data.

[0115] S420 , evaluating the fine-tuned operation and maintenance strategy determination model based on the number of true positive examples, the number of false positive examples, and the number of false negative examples.

[0116] Exemplarily, the model precision can be determined based on the number of true positive examples and the number of false positive examples; the model recall can be determined based on the number of true positive examples and the number of false negative examples.

[0117] For example, the model accuracy can be determined by the following formula:

[0118] ;

[0119] In the formula, is the model accuracy; TP (True Positive) indicates the number of true positive examples; FP (False Positive) indicates the number of false positive examples.

[0120] For example, the model recall rate can be determined by the following formula:

[0121] ;

[0122] In the formula, is the model recall rate; TP (True Positive) represents the number of true positive examples; FN (False Negative) represents the number of false negative examples.

[0123] Furthermore, the comprehensive score of the model is determined based on the model accuracy and the model recall rate. For example, the comprehensive score of the model can be determined by the following formula:

[0124] ;

[0125] In the formula, F1 is the comprehensive score of the model, which takes into account both the model accuracy and the model recall rate, and can more effectively describe the overall performance of the model rather than just a single indicator.

[0126] The step shown in S420 above is only one method of evaluating the operation and maintenance strategy determination model provided by the present application. In some embodiments, the operation and maintenance strategy determination model can also be evaluated in the following manner.

[0127] The model accuracy is determined based on the number of accurate operation and maintenance strategy prediction results corresponding to the sample problem description data and the total number of sample problem description data. For example, it can be determined by the following formula:

[0128] ;

[0129] In the formula, total_sample represents the total number of sample problem description data, and correct_sample represents the number of accurate prediction results.

[0130] According to the prediction results and operation and maintenance strategy labels corresponding to each sample problem description data, the mean square error, mean absolute error and determination coefficient are determined. Exemplarily, they can be determined by the following formula:

[0131] ;

[0132] Where, MSE (Mean Squared Error) is the mean square error; n is the total number of sample problem description data; i is the i-th sample problem description data; y i Describe the prediction results of the data for the i-th sample problem; Describes the operational strategy label of the data for the i-th sample problem.

[0133] ;

[0134] Where, MAE (Mean Absolute Error) is the mean absolute error; n is the total number of sample problem description data; i is the i-th sample problem description data; y i Describe the prediction results of the data for the i-th sample problem; Describes the operational strategy label of the data for the i-th sample problem.

[0135] ;

[0136] In the formula, y i Describe the prediction results of the data for the i-th sample problem; Describe the operation and maintenance strategy label of the data for the i-th sample problem; The average value of the operation and maintenance strategy label describing the data for the i-th sample problem.

[0137] Accordingly, when the evaluation result indicates that the fine-tuned operation and maintenance strategy determination model meets the preset requirements, a trained operation and maintenance strategy determination model is obtained. The preset requirements can be requirements for evaluating the model evaluation results based on manual experience, or requirements for constraining the number of training times; this application does not impose any restrictions on this.

[0138] In the above embodiment, a specific process of training the operation and maintenance strategy determination model is provided.

[0139] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the system operation and maintenance strategy determination method provided by the present application is introduced in detail.

[0140] See also Figure 5A The system operation and maintenance strategy determination method shown includes:

[0141] S510A, construction and training of operation and maintenance strategy determination model;

[0142] S520A, input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data.

[0143] See also Figure 5B The operational strategy shown determines the steps for building and training the model, including:

[0144] S501B, using the pre-trained generative pre-trained model as an operation and maintenance strategy determination model;

[0145] S502B, obtaining historical faults that occurred during the historical use of the system to be detected, and historical operation and maintenance records corresponding to each historical fault;

[0146] S503B, obtaining historical log data recorded during the historical use of the system to be detected;

[0147] S504B, taking the server system information of the system to be detected, the historical operation and maintenance records and the historical log data as the historical operation and maintenance data of the system to be detected;

[0148] S505B, preprocessing the historical operation and maintenance data of the system to be detected to obtain preprocessed operation and maintenance information;

[0149] Wherein, the preprocessing includes at least one of text consistency processing, text deduplication processing and text standardization processing;

[0150] S506B, performing text segmentation processing on the pre-processed operation and maintenance information to obtain a text segmentation result;

[0151] S507B, constructing a text vocabulary table according to the occurrence frequency of each word in the text segmentation result;

[0152] S508B, determining sample question description data based on each word information in the text vocabulary table;

[0153] S509B, inputting the sample problem description data into the operation and maintenance strategy determination model to obtain the operation and maintenance strategy prediction result corresponding to the sample problem description data;

[0154] S510B, fine-tuning the network parameters of the operation and maintenance strategy determination model according to the operation and maintenance strategy prediction result and the operation and maintenance strategy label corresponding to the corresponding sample problem description data;

[0155] S511B, determining the number of true positive examples, the number of false positive examples, and the number of false negative examples according to the operation and maintenance strategy prediction results corresponding to the sample problem description data and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data;

[0156] S512B, evaluate the fine-tuned operation and maintenance strategy determination model based on the number of true positives, false positives, and false negatives;

[0157] S513B, when the evaluation result indicates that the fine-tuned operation and maintenance strategy determination model meets the preset requirements, a trained operation and maintenance strategy determination model is obtained.

[0158] 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.

[0159] Based on the same inventive concept, the embodiment of the present application also provides a system operation and maintenance strategy determination device for implementing the system operation and maintenance strategy determination 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 the embodiments of one or more system operation and maintenance strategy determination devices provided below can refer to the limitations of the system operation and maintenance strategy determination method above, and will not be repeated here.

[0160] In an exemplary embodiment, Figure 6 As shown, a system operation and maintenance strategy determination device is provided, including: a model determination module 610, a model fine-tuning module 620, a data acquisition module 630 and a strategy determination module 640, wherein:

[0161] The data acquisition module 630 is used to acquire the problem description data of the system to be detected;

[0162] A strategy determination module 640 is used to input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data;

[0163] The operation and maintenance strategy determination model is trained in the following ways:

[0164] A model determination module 610 is used to use the pre-trained generative pre-trained model as an operation and maintenance strategy determination model;

[0165] The model fine-tuning module 620 is used to obtain the historical operation and maintenance data of the system to be detected, and fine-tune the network parameters of the operation and maintenance strategy determination model according to the historical operation and maintenance data.

[0166] In one embodiment, the model fine-tuning module 620 includes a first acquisition unit, which is used to acquire historical faults that occurred during the historical use of the system to be detected, and historical operation and maintenance records corresponding to each historical fault; a second acquisition unit, which is used to acquire historical log data recorded during the historical use of the system to be detected; and a data determination unit, which is used to use the server system information of the system to be detected, the historical operation and maintenance records and the historical log data as the historical operation and maintenance data of the system to be detected.

[0167] In one embodiment, the model fine-tuning module 620 includes a first determination unit, which is used to extract problem features from the historical operation and maintenance data of the system to be detected to obtain sample problem description data; a second determination unit, which is used to input the sample problem description data into the operation and maintenance strategy determination model to obtain the operation and maintenance strategy prediction results corresponding to the sample problem description data; and a model training unit, which is used to fine-tune the network parameters of the operation and maintenance strategy determination model according to the operation and maintenance strategy prediction results and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data.

[0168] In one embodiment, the first determination unit includes a text segmentation subunit, which is used to perform text segmentation processing on the historical operation and maintenance data of the system to be detected to obtain a text segmentation result; a vocabulary construction subunit, which is used to construct a text vocabulary according to the frequency of occurrence of each word in the text segmentation result; and a data determination subunit, which is used to determine the sample problem description data based on the information of each word in the text vocabulary.

[0169] In one embodiment, the text segmentation subunit is specifically used to perform data preprocessing on the historical operation and maintenance data of the system to be detected to obtain preprocessed operation and maintenance information; wherein the preprocessing includes at least one of text consistency processing, text deduplication processing and text standardization processing; and text segmentation processing is performed on the preprocessed operation and maintenance information to obtain a text segmentation result.

[0170] In one embodiment, the system operation and maintenance strategy determination device also includes a model evaluation module, which is used to evaluate the fine-tuned operation and maintenance strategy determination model; the model evaluation module includes a quantity determination unit, which is used to determine the number of true positive examples, the number of false positive examples, and the number of false negative examples based on the operation and maintenance strategy prediction results corresponding to the sample problem description data, and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data; the model evaluation unit is used to evaluate the fine-tuned operation and maintenance strategy determination model based on the number of true positive examples, the number of false positive examples, and the number of false negative examples; accordingly, the system operation and maintenance strategy determination device also includes a model determination module, which is used to obtain a trained operation and maintenance strategy determination model when the evaluation result indicates that the fine-tuned operation and maintenance strategy determination model meets the preset requirements.

[0171] Each module in the above-mentioned system operation and maintenance strategy determination device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0172] 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 7 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, 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. 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 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 method for determining a system operation and maintenance strategy 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.

[0173] 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.

[0174] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0175] Obtain problem description data of the system to be tested;

[0176] Input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data;

[0177] The operation and maintenance strategy determination model is trained in the following ways:

[0178] Use the pre-trained generative pre-trained model as the operation and maintenance strategy determination model;

[0179] Obtain the historical operation and maintenance data of the system to be tested, and fine-tune the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data.

[0180] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0181] Obtain historical faults that occurred during the historical use of the system to be tested, as well as historical operation and maintenance records corresponding to each historical fault;

[0182] Obtain historical log data recorded during historical use of the system to be tested;

[0183] The server system information of the system to be detected, the historical operation and maintenance records and the historical log data are used as the historical operation and maintenance data of the system to be detected.

[0184] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0185] Extract problem features from the historical operation and maintenance data of the system to be tested to obtain sample problem description data;

[0186] Input the sample problem description data into the operation and maintenance strategy determination model to obtain the operation and maintenance strategy prediction result corresponding to the sample problem description data;

[0187] According to the operation and maintenance strategy prediction results and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data, the network parameters of the operation and maintenance strategy determination model are fine-tuned.

[0188] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0189] Perform text segmentation processing on the historical operation and maintenance data of the detection system to obtain the text segmentation results;

[0190] Construct a text vocabulary based on the frequency of occurrence of each word in the text segmentation results;

[0191] Based on the information of each word in the text vocabulary, sample question description data is determined.

[0192] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0193] Perform data preprocessing on the historical operation and maintenance data of the detection system to obtain preprocessed operation and maintenance information;

[0194] Wherein, the preprocessing includes at least one of text consistency processing, text deduplication processing and text standardization processing;

[0195] Perform text segmentation on the preprocessed operation and maintenance information to obtain text segmentation results.

[0196] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0197] Evaluate the fine-tuned operation and maintenance strategy determination model;

[0198] The evaluation process includes:

[0199] Determine the number of true positives, false positives, and false negatives based on the operation and maintenance strategy prediction results corresponding to the sample problem description data and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data;

[0200] Evaluate the fine-tuned O&M strategy determination model based on the number of true positives, false positives, and false negatives.

[0201] Accordingly, the method further includes:

[0202] When the evaluation result indicates that the fine-tuned operation and maintenance strategy determination model meets the preset requirements, a trained operation and maintenance strategy determination model is obtained.

[0203] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0204] Obtain problem description data of the system to be tested;

[0205] Input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data;

[0206] The operation and maintenance strategy determination model is trained in the following ways:

[0207] Use the pre-trained generative pre-trained model as the operation and maintenance strategy determination model;

[0208] Obtain the historical operation and maintenance data of the system to be tested, and fine-tune the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data.

[0209] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0210] Obtain historical faults that occurred during the historical use of the system to be tested, as well as historical operation and maintenance records corresponding to each historical fault;

[0211] Obtain historical log data recorded during historical use of the system to be tested;

[0212] The server system information of the system to be detected, the historical operation and maintenance records and the historical log data are used as the historical operation and maintenance data of the system to be detected.

[0213] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0214] Extract problem features from the historical operation and maintenance data of the system to be tested to obtain sample problem description data;

[0215] Input the sample problem description data into the operation and maintenance strategy determination model to obtain the operation and maintenance strategy prediction result corresponding to the sample problem description data;

[0216] According to the operation and maintenance strategy prediction results and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data, the network parameters of the operation and maintenance strategy determination model are fine-tuned.

[0217] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0218] Perform text segmentation processing on the historical operation and maintenance data of the detection system to obtain the text segmentation results;

[0219] Construct a text vocabulary based on the frequency of occurrence of each word in the text segmentation results;

[0220] Based on the information of each word in the text vocabulary, sample question description data is determined.

[0221] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0222] Perform data preprocessing on the historical operation and maintenance data of the detection system to obtain preprocessed operation and maintenance information;

[0223] Wherein, the preprocessing includes at least one of text consistency processing, text deduplication processing and text standardization processing;

[0224] Perform text segmentation on the preprocessed operation and maintenance information to obtain text segmentation results.

[0225] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0226] Evaluate the fine-tuned operation and maintenance strategy determination model;

[0227] The evaluation process includes:

[0228] Determine the number of true positives, false positives, and false negatives based on the operation and maintenance strategy prediction results corresponding to the sample problem description data and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data;

[0229] Evaluate the fine-tuned O&M strategy determination model based on the number of true positives, false positives, and false negatives.

[0230] Accordingly, the method further includes:

[0231] When the evaluation result indicates that the fine-tuned operation and maintenance strategy determination model meets the preset requirements, a trained operation and maintenance strategy determination model is obtained.

[0232] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0233] Obtain problem description data of the system to be tested;

[0234] Input the problem description data into the trained operation and maintenance strategy determination model to obtain the operation and maintenance strategy corresponding to the problem description data;

[0235] The operation and maintenance strategy determination model is trained in the following ways:

[0236] Use the pre-trained generative pre-trained model as the operation and maintenance strategy determination model;

[0237] Obtain the historical operation and maintenance data of the system to be tested, and fine-tune the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data.

[0238] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0239] Obtain historical faults that occurred during the historical use of the system to be tested, as well as historical operation and maintenance records corresponding to each historical fault;

[0240] Obtain historical log data recorded during historical use of the system to be tested;

[0241] The server system information of the system to be detected, the historical operation and maintenance records and the historical log data are used as the historical operation and maintenance data of the system to be detected.

[0242] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0243] Extract problem features from the historical operation and maintenance data of the system to be tested to obtain sample problem description data;

[0244] Input the sample problem description data into the operation and maintenance strategy determination model to obtain the operation and maintenance strategy prediction result corresponding to the sample problem description data;

[0245] According to the operation and maintenance strategy prediction results and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data, the network parameters of the operation and maintenance strategy determination model are fine-tuned.

[0246] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0247] Perform text segmentation processing on the historical operation and maintenance data of the detection system to obtain the text segmentation results;

[0248] Construct a text vocabulary based on the frequency of occurrence of each word in the text segmentation results;

[0249] Based on the information of each word in the text vocabulary, sample question description data is determined.

[0250] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0251] Perform data preprocessing on the historical operation and maintenance data of the detection system to obtain preprocessed operation and maintenance information;

[0252] Wherein, the preprocessing includes at least one of text consistency processing, text deduplication processing and text standardization processing;

[0253] Perform text segmentation on the preprocessed operation and maintenance information to obtain text segmentation results.

[0254] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0255] Evaluate the fine-tuned operation and maintenance strategy determination model;

[0256] The evaluation process includes:

[0257] Determine the number of true positives, false positives, and false negatives based on the operation and maintenance strategy prediction results corresponding to the sample problem description data and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data;

[0258] Evaluate the fine-tuned O&M strategy determination model based on the number of true positives, false positives, and false negatives.

[0259] Accordingly, the method further includes:

[0260] When the evaluation result indicates that the fine-tuned operation and maintenance strategy determination model meets the preset requirements, a trained operation and maintenance strategy determination model is obtained.

[0261] 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.

[0262] 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.

[0263] 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.

[0264] 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 construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications 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 method for determining a system operation and maintenance strategy, characterized in that: The method comprises: Obtain problem description data of the system to be tested; Inputting the problem description data into a trained operation and maintenance strategy determination model to obtain an operation and maintenance strategy corresponding to the problem description data; The operation and maintenance strategy determination model is trained in the following manner: Use the pre-trained generative pre-trained model as the operation and maintenance strategy determination model; The historical operation and maintenance data of the system to be detected is obtained, and the network parameters of the operation and maintenance strategy determination model are fine-tuned according to the historical operation and maintenance data.

2. The method according to claim 1, characterized in that The obtaining of historical operation and maintenance data of the system to be detected includes: Obtain historical faults that occurred during historical use of the system to be tested, as well as historical operation and maintenance records corresponding to each of the historical faults; Obtaining historical log data recorded during historical use of the system to be detected; The server system information of the system to be detected, the historical operation and maintenance records and the historical log data are used as the historical operation and maintenance data of the system to be detected.

3. The method according to claim 1, characterized in that The fine-tuning of the network parameters of the operation and maintenance strategy determination model according to the historical operation and maintenance data includes: Extracting problem features from historical operation and maintenance data of the system to be tested to obtain sample problem description data; Inputting the sample problem description data into the operation and maintenance strategy determination model to obtain the operation and maintenance strategy prediction result corresponding to the sample problem description data; According to the operation and maintenance strategy prediction results and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data, the network parameters of the operation and maintenance strategy determination model are fine-tuned.

4. The method according to claim 3, characterized in that The problem feature extraction of the historical operation and maintenance data of the system to be detected to obtain sample problem description data includes: Performing text segmentation processing on the historical operation and maintenance data of the system to be detected to obtain a text segmentation result; Constructing a text vocabulary table according to the occurrence frequency of each word in the text segmentation result; Based on the information of each word in the text vocabulary, sample question description data is determined.

5. The method according to claim 4, characterized in that The performing text segmentation processing on the historical operation and maintenance data of the system to be detected to obtain a text segmentation result includes: Performing data preprocessing on the historical operation and maintenance data of the system to be detected to obtain preprocessed operation and maintenance information; Wherein, the preprocessing includes at least one of text consistency processing, text deduplication processing and text standardization processing; The preprocessed operation and maintenance information is subjected to text segmentation processing to obtain a text segmentation result.

6. The method according to any one of claims 1 to 5, characterized in that The method further comprises: Evaluate the fine-tuned operation and maintenance strategy determination model; The evaluation process includes: Determine the number of true positives, false positives, and false negatives based on the operation and maintenance strategy prediction results corresponding to the sample problem description data and the operation and maintenance strategy labels corresponding to the corresponding sample problem description data; Based on the number of true positive examples, the number of false positive examples, and the number of false negative examples, the fine-tuned operation and maintenance strategy determination model is evaluated; Accordingly, the method further includes: When the evaluation result indicates that the fine-tuned operation and maintenance strategy determination model meets the preset requirements, a trained operation and maintenance strategy determination model is obtained.

7. A system operation and maintenance strategy determination device, characterized in that: The device comprises: A data acquisition module is used to obtain problem description data of the system to be tested; A strategy determination module, used to input the problem description data into a trained operation and maintenance strategy determination model to obtain an operation and maintenance strategy corresponding to the problem description data; The operation and maintenance strategy determination model is trained in the following manner: A model determination module is used to use the pre-trained generative pre-trained model as an operation and maintenance strategy determination model; The model fine-tuning module is used to obtain historical operation and maintenance data of the system to be detected, and fine-tune the network parameters of the operation and maintenance strategy determination model based on the historical operation and maintenance data.

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