System operation and maintenance method, device, equipment, medium and program product

Abstract

The invention provides a system operation and maintenance method which can be applied to the technical field of artificial intelligence and financial science and technology. The system operation and maintenance method comprises the steps that in response to a first operation and maintenance problem input by a user, prompt word enhancement is conducted on the first operation and maintenance problem, and a second operation and maintenance problem is obtained; processing the second operation and maintenance question by adopting a pre-trained large language model, and outputting an answer to the first question, the pre-trained large language model being obtained by performing at least one of fine tuning and preference optimization based on a specific training sample; generating an operation and maintenance script according to the first operation and maintenance question and the first question answer; and executing the operation and maintenance script to realize operation and maintenance operation corresponding to the first operation and maintenance problem. The invention further provides a system operation and maintenance device, equipment, a storage medium and a program product.

Description

Technical Field

[0001] The present disclosure relates to the fields of artificial intelligence and financial technology, and specifically to automated processing of system operation and maintenance, and more specifically to a system operation and maintenance method, device, equipment, medium and program product. Background Art

[0002] As the digitalization process accelerates, the complexity of enterprise operation and maintenance environments continues to increase, and traditional operation and maintenance methods can no longer meet current needs. In recent years, the rise of large language models has brought new opportunities for change in the field of operation and maintenance automation.

[0003] However, the existing system's response speed and accuracy need to be improved when dealing with operation and maintenance issues with high real-time requirements, and it lacks the ability to generate and execute automated scripts in the operation and maintenance field. Summary of the invention

[0004] In view of the above problems, the present disclosure provides a system operation and maintenance method, apparatus, device, medium and program product for improving system operation and maintenance efficiency.

[0005] According to a first aspect of the present disclosure, a system operation and maintenance method is provided, including: in response to a user inputting a first operation and maintenance question, enhancing the first operation and maintenance question with prompt words to obtain a second operation and maintenance question; processing the second operation and maintenance question using a pre-trained large language model, and outputting an answer to the first question, the pre-trained large language model being obtained by at least one of fine-tuning and preference optimization based on specific training samples; generating an operation and maintenance script according to the first operation and maintenance question and the answer to the first question; and executing the operation and maintenance script to implement an operation and maintenance operation corresponding to the first operation and maintenance question.

[0006] According to an embodiment of the present disclosure, the training of a pre-trained large language model includes: obtaining operation and maintenance data, the operation and maintenance data including at least one of operation and maintenance documents, operation and maintenance logs, and web page operation and maintenance knowledge; constructing an operation and maintenance data set based on the operation and maintenance data, the operation and maintenance data set being characterized as at least one of an operation and maintenance knowledge question and answer pair and an operation and maintenance knowledge graph; and using the operation and maintenance data set to fine-tune the initial large language model to obtain a pre-trained large language model.

[0007] According to an embodiment of the present disclosure, the training of a pre-trained large language model includes: obtaining user evaluations corresponding to operation and maintenance data; constructing a preference data set based on the user evaluations and the operation and maintenance data; and using the preference data set to train an initial large language model or a fine-tuned initial large language model to obtain a pre-trained large language model.

[0008] According to an embodiment of the present disclosure, in response to a user raising a first operation and maintenance question, prompt words are enhanced on the first operation and maintenance question to obtain a second operation and maintenance question, including: obtaining a vector database, the vector database including multiple operation and maintenance knowledge question and answer pairs; matching the first operation and maintenance question with the multiple operation and maintenance knowledge question and answer pairs to obtain at least one target question, the vector distance corresponding to the target question and the first operation and maintenance question being less than a preset threshold; and determining the second operation and maintenance question based on the operation and maintenance knowledge question and answer pairs corresponding to the first operation and maintenance question and the target question.

[0009] According to an embodiment of the present disclosure, before using a pre-trained large language model to process the second operation and maintenance problem, the system operation and maintenance method also includes: retrieving background knowledge related to the second operation and maintenance problem in a pre-built operation and maintenance knowledge base; using the background knowledge to enhance the background knowledge of the second operation and maintenance problem to obtain a third operation and maintenance problem; using the pre-trained large language model to process the third operation and maintenance problem, and outputting the answer to the second question.

[0010] According to an embodiment of the present disclosure, the system operation and maintenance method further includes: updating a pre-built operation and maintenance knowledge base according to the first operation and maintenance question and the answer to the first question.

[0011] According to an embodiment of the present disclosure, the system operation and maintenance method also includes: obtaining historical conversations related to the user; when the user has conducted multiple rounds of conversations, using a pre-trained large language model to process the second operation and maintenance question and the historical conversations, and outputting the answer to the first question.

[0012] According to an embodiment of the present disclosure, the system operation and maintenance method further includes: determining an execution object of the operation and maintenance script; and continuously monitoring the execution object after executing the operation and maintenance script on the execution object.

[0013] A second aspect of the present disclosure provides an operation and maintenance device, including: an enhancement module, used to enhance the first operation and maintenance question with prompt words in response to a user inputting a first operation and maintenance question, so as to obtain a second operation and maintenance question; a processing module, used to process the second operation and maintenance question using a pre-trained large language model, and output an answer to the first question, wherein the pre-trained large language model is obtained by at least one of fine-tuning and preference optimization based on specific training samples; a generation module, used to generate an operation and maintenance script according to the first operation and maintenance question and the answer to the first question; and an execution module, used to execute the operation and maintenance script to implement the operation and maintenance operation corresponding to the first operation and maintenance question.

[0014] A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0015] The fourth aspect of the present disclosure further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the above computer program or instructions are executed by a processor.

[0016] The fifth aspect of the present disclosure further provides a computer program product, including a computer program or instructions, which implement the steps of the above method when the above computer program or instructions are executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above contents and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0018] Figure 1 The application scenario diagram of the system operation and maintenance method, apparatus, device, medium and program product according to the embodiment of the present disclosure is schematically shown;

[0019] Figure 2 A flowchart of a system operation and maintenance method according to an embodiment of the present disclosure is schematically shown;

[0020] Figure 3 A flowchart of fine-tuning of a large language model in a system operation and maintenance method according to an embodiment of the present disclosure is schematically shown;

[0021] Figure 4 A flowchart of preference optimization processing of a large language model in a system operation and maintenance method according to an embodiment of the present disclosure is schematically shown;

[0022] Figure 5 A flowchart of prompt word enhancement in a system operation and maintenance method according to an embodiment of the present disclosure is schematically shown;

[0023] Figure 6 A flowchart of background knowledge enhancement in a system operation and maintenance method according to an embodiment of the present disclosure is schematically shown;

[0024] Figure 7 A structural block diagram of a system operation and maintenance device according to an embodiment of the present disclosure is schematically shown; and

[0025] Figure 8 A block diagram of an electronic device suitable for implementing a system operation and maintenance method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0026] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0027] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0028] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0029] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0030] The embodiments of the present disclosure provide a system operation and maintenance method, device, equipment, medium and program product, which can be used in the field of financial technology or other fields. It should be noted that the system operation and maintenance method, device, equipment, medium and program product of the present disclosure can be used in the field of financial technology, and can also be used in any field other than the field of financial technology, and the application field of the system operation and maintenance method, device, equipment, medium and program product of the present disclosure is not limited.

[0031] In the technical solution of the present disclosure, the user information (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0032] In the scenario of using personal information for automated decision-making, the methods, devices, and systems provided by the embodiments of the present disclosure provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating a person's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0033] The embodiment of the present disclosure provides a system operation and maintenance method, including: in response to a user inputting a first operation and maintenance question, enhancing the prompt words of the first operation and maintenance question to obtain a second operation and maintenance question; using a pre-trained large language model to process the second operation and maintenance question and output an answer to the first question, the pre-trained large language model is obtained by at least one of fine-tuning and preference optimization based on a specific training sample; generating an operation and maintenance script according to the first operation and maintenance question and the answer to the first question; and executing the operation and maintenance script to implement the operation and maintenance operation corresponding to the first operation and maintenance question. By enhancing the prompt words of the first operation and maintenance question, the accuracy of the present implementation in handling problems is improved, and by using the large language model after fine-tuning and preference optimization, an operation and maintenance script can be automatically generated for the second operation and maintenance question, realizing automated operations in the field of operation and maintenance.

[0034] Figure 1 The application scenario diagram of the system operation and maintenance method according to an embodiment of the present disclosure is schematically shown.

[0035] like Figure 1 As shown, the application scenario 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the first terminal device 101, the second terminal device 102, the third terminal device 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0036] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0037] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0038] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0039] It should be noted that the system operation and maintenance method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the system operation and maintenance device provided in the embodiment of the present disclosure can generally be set in the server 105. The system operation and maintenance method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the system operation and maintenance device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0040] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0041] The following will be based on Figure 1 The scene described by Figure 2~Figure 4 The system operation and maintenance method of the disclosed embodiment is described in detail.

[0042] Figure 2 The flowchart of the system operation and maintenance method according to the embodiment of the present disclosure is schematically shown.

[0043] According to the embodiments of the present disclosure, Figure 2 As shown, the system operation and maintenance of this embodiment includes operations S210 to S240, and the transaction processing method can be executed by a server or any terminal.

[0044] In operation S210, in response to a user inputting a first operation and maintenance question, prompt word enhancement is performed on the first operation and maintenance question to obtain a second operation and maintenance question.

[0045] In some embodiments, the first operation and maintenance problem may include various operational problems in the field of operation and maintenance. The field of operation and maintenance refers to the daily operation and maintenance work of IT systems, including hardware and software management, troubleshooting, and performance optimization. Among them, troubleshooting problems may include, for example, how to solve system failures such as server failure to start and network interruption. Hardware and software management problems may include, for example, configuration problems such as how to correctly configure databases and network devices. Performance optimization problems may include, for example, how to optimize the system resources and load balancing.

[0046] For example, firstly, the user's natural language operation and maintenance question is semantically parsed using advanced natural language processing technology to extract key information. The key information is combined to obtain the first operation and maintenance question, and then the first operation and maintenance question is enhanced with prompt words to obtain the second operation and maintenance question. The second operation and maintenance question is a question text with prompt words obtained after relevant search for the first operation and maintenance question.

[0047] In operation S220, the second operation and maintenance question is processed using a pre-trained large language model, and an answer to the first question is output, where the pre-trained large language model is obtained by at least one of fine-tuning and preference optimization based on specific training samples.

[0048] In some embodiments, the base model of the pre-trained large language model can be, for example, a natural language processing model based on deep learning, and the model has been pre-trained on large-scale general expectations, has strong language understanding and generation capabilities, and can quickly understand and generate natural language text.

[0049] For example, the answer to the first question may be a natural language text used to describe the operation steps performed for an operation and maintenance problem.

[0050] In operation S230, an operation and maintenance script is generated according to the first operation and maintenance question and the answer to the first question.

[0051] In some embodiments, for example, the first operation and maintenance question and the answer to the first question may be input into a pre-trained large language model, and the large language model may be used to generate an operation and maintenance script. The operation and maintenance script is a program generated for solving the first operation and maintenance question.

[0052] In operation S240, the operation and maintenance script is executed to implement the operation and maintenance operation corresponding to the first operation and maintenance problem.

[0053] In some embodiments, the operation and maintenance operation is a series of actions performed to solve the first operation and maintenance problem. For example, the first operation and maintenance problem is how to solve the problem that the server cannot be started, and the corresponding operation and maintenance operation may be to reinstall the server system.

[0054] By enhancing the prompt words for the first operation and maintenance problem, the accuracy of this implementation in handling problems is improved, and through fine-tuning and preference-optimized large language models, operation and maintenance scripts can be automatically generated for the second operation and maintenance problem, realizing automated operations in the operation and maintenance field.

[0055] In operation S210, user information is obtained.

[0056] In some embodiments, before obtaining the user's information, the user's consent or authorization may be obtained. For example, before operation S210, a request to obtain the user's information may be issued to the user. When the user agrees or authorizes that the user's information may be obtained, the operation S210 is performed.

[0057] In operation S220, the user information is used to perform subsequent action execution processing to implement terminal operations or information acquisition.

[0058] In some embodiments, a corresponding operation entry may be provided for the user to choose to agree or reject the automated decision result. That is, before the user information is processed for subsequent actions, an instruction to agree or reject the processing input by the user through the corresponding operation entry may be obtained. If the user agrees to the processing, the user information is processed / decided for subsequent actions, i.e., step S220 is executed. If the user refuses to process, the expert decision process is entered.

[0059] Figure 3 The flowchart of the fine-tuning process of a large language model in the system operation and maintenance method according to an embodiment of the present disclosure is schematically shown.

[0060] According to the embodiments of the present disclosure, Figure 3 As shown, the fine-tuning process of the large language model in the system operation and maintenance method of this embodiment includes operations S321 to S323.

[0061] In operation S321, operation and maintenance data is acquired, where the operation and maintenance data includes at least one of operation and maintenance documents, operation and maintenance logs, and webpage operation and maintenance knowledge.

[0062] In some embodiments, operation and maintenance documents are, for example, a large number of paper documents related to operation and maintenance that are collected. Operation and maintenance logs are log files automatically generated during system operation, which represent the operation records and problem-solving processes of operation and maintenance personnel in actual work. Web operation and maintenance knowledge is relevant information related to operation and maintenance obtained from web pages. For example, discussions and questions and answers related to operation and maintenance collected from operation and maintenance technical forums and communities. In addition, operation and maintenance data may also include enterprise intranet data, for example, internal information used in the operation and maintenance process obtained from the enterprise's internal knowledge base.

[0063] In operation S322, an operation and maintenance data set is constructed based on the operation and maintenance data, and the operation and maintenance data set is characterized as at least one of an operation and maintenance knowledge question and answer pair and an operation and maintenance knowledge graph.

[0064] In some embodiments, for example, question-answer pairs can be constructed directly based on the collected operation and maintenance data, where the question-answer pairs represent common operation and maintenance problems and their solutions. It is also possible to generate question-answer pairs containing detailed steps and operation instructions by simulating actual operation and maintenance scenarios using operation and maintenance data. The content covered by the knowledge graph can include, for example, device configuration, troubleshooting, and performance optimization.

[0065] In operation S323, the initial large language model is fine-tuned using the operation and maintenance data set to obtain a pre-trained large language model.

[0066] In some embodiments, before fine-tuning the initial large language model, the method may further include data preprocessing.

[0067] Data preprocessing includes, for example: text cleaning, removing extra spaces, HTML tags, and special characters from the collected text data. Format standardization, standardizing the data formats from different sources to ensure the consistency of the data formats. Word segmentation and annotation, using entity recognition or keyword extraction to segment text data and perform necessary annotations.

[0068] In some embodiments, for example, a supervised learning strategy is used to fine-tune the initial large language model, and a data set dedicated to the operation and maintenance field is used for training iterations. After the training iteration, an optimization algorithm such as the Adam optimization algorithm is used to adjust the model parameters to ensure that the model can better adapt to the tasks in the operation and maintenance field. After the optimization and adjustment, a loss function such as the cross entropy loss function is selected to measure the direct difference between the model output and the expected answer, and the loss function is minimized during the training process. At this point, a fine-tuned large language model is obtained.

[0069] The dataset that is not used for training is used as the validation set. The fine-tuned large language model is cross-validated using the validation set, and the performance of the model is evaluated using indicators such as accuracy, recall, and F1 score to ensure that the model has high accuracy in answering questions in the operation and maintenance field.

[0070] The verified large language model can be used to answer actual operation and maintenance questions, and collect user feedback during actual use. Through feedback, we can identify the deficiencies of the model, update the data set used for fine-tuning according to the deficiencies, and regularly repeat fine-tuning the large language model with the updated data set to ensure that the model always adapts to the latest operation and maintenance requirements and technological developments.

[0071] By fine-tuning the initial large language model, the response speed of the large language model and the accuracy of the output results are improved.

[0072] Figure 4 The flowchart of the preference optimization processing of the large language model in the system operation and maintenance method according to the embodiment of the present disclosure is schematically shown.

[0073] According to the embodiments of the present disclosure, Figure 4 As shown, the preference optimization processing of the large language model in the system operation and maintenance method of this embodiment includes operations S421 to S423.

[0074] In operation S421, user evaluations corresponding to the operation and maintenance data are obtained.

[0075] In some embodiments, the user evaluation is the user's direct feedback on the operation and maintenance data, such as satisfaction and preference for the operation and maintenance data.

[0076] In operation S422, a preference data set is constructed based on user evaluation and operation and maintenance data.

[0077] In the embodiment of the present disclosure, for example, the user evaluation and operation and maintenance data are first cleaned to remove duplicate and invalid data, and then the user's preference information is annotated with a label according to each user evaluation and its corresponding operation and maintenance data. After the labeling is completed, the data is quality tested and the tested data is integrated into a unified preference data set.

[0078] In operation S423, the initial large language model or the fine-tuned initial large language model is trained using the preference data set to obtain a pre-trained large language model.

[0079] In some embodiments, for example, a direct preference optimization algorithm is used to train the initial large language model. During the training process, the direct preference optimization algorithm adjusts the parameters in the module according to the preference data set to maximize the probability of the preferred output and minimize the probability of the non-preferred output. The performance of the model is regularly evaluated during the training process, and an evaluation result is generated. The strategy of the language model is then updated according to the evaluation result. A pre-trained large language model is obtained.

[0080] By performing preference optimization training on the initial large language model, the model can quickly adapt to operation and maintenance scenarios and generate outputs with compound user preferences, and the output process summary can maintain stable performance.

[0081] Figure 5 The flowchart of prompt word enhancement in the system operation and maintenance method according to an embodiment of the present disclosure is schematically shown.

[0082] According to the embodiments of the present disclosure, Figure 5 As shown, the prompt word enhancement in the system operation and maintenance method of this embodiment includes operations S511 to S513.

[0083] In operation S511, a vector database is obtained, where the vector database includes a plurality of operation and maintenance knowledge question and answer pairs.

[0084] In some embodiments, the vector database is, for example, a database that uses vector storage and query techniques to store and process coordinate data.

[0085] In operation S512, the first operation and maintenance question is matched with a plurality of operation and maintenance knowledge question and answer pairs to obtain at least one target question, and a vector distance between the target question and the first operation and maintenance question is less than a preset threshold.

[0086] In some embodiments, keywords of the first operation and maintenance question are extracted, and keywords are matched with questions in the operation and maintenance knowledge question and answer pair to obtain a target question with at least one keyword that is the same as the keyword of the first operation and maintenance question. The vector distance between the target question and the first operation and maintenance question is less than a preset threshold, indicating that the distance between the keywords of the target question and the keywords of the first operation and maintenance question in the vector database is very close, that is, the semantics of the keywords of the two are roughly the same.

[0087] In operation S513, a second operation and maintenance problem is determined based on the operation and maintenance knowledge question and answer pair corresponding to the first operation and maintenance problem and the target problem.

[0088] In some embodiments, after the system retrieves multiple target questions and their answers that are closest to the user's question from the vector database, it integrates these target questions and the corresponding segmented knowledge texts with the first operation and maintenance question into a question text with prompt words as the second operation and maintenance question.

[0089] By enhancing the prompt words for the first operation and maintenance problem, the disclosed solution can more accurately analyze the first operation and maintenance problem in combination with the relevant target problem, thereby improving the accuracy of the disclosed solution.

[0090] Figure 6 A flowchart of background knowledge enhancement in a system operation and maintenance method according to an embodiment of the present disclosure is schematically shown.

[0091] According to the embodiments of the present disclosure, Figure 6 As shown, the background knowledge enhancement in the system operation and maintenance method of this embodiment includes operations S621 to S623.

[0092] In operation S621, background knowledge related to the second operation and maintenance problem is retrieved from a pre-built operation and maintenance knowledge base.

[0093] In some embodiments, the operation and maintenance knowledge base can, for example, build a comprehensive operation and maintenance knowledge base by collecting and organizing operation and maintenance related documents, and structure the operation and maintenance knowledge base and classify and store various types of documents to facilitate efficient retrieval and utilization.

[0094] For example, keyword information of the second operation and maintenance problem is extracted, and relevant information is retrieved from the operation and maintenance knowledge base according to the keyword information as background knowledge of the second operation and maintenance problem.

[0095] In operation S622, background knowledge is used to enhance the second operation and maintenance problem, thereby obtaining a third operation and maintenance problem.

[0096] In some embodiments, for example, the retrieved background knowledge may be integrated with the second operation and maintenance problem to obtain a third operation and maintenance problem.

[0097] In operation S623, the pre-trained large language model is used to process the third operation and maintenance question, and an answer to the second question is output.

[0098] By enhancing the background knowledge of the second operation and maintenance problem, the third operation and maintenance problem is obtained, so that the second operation and maintenance problem can more accurately reflect the operation requirements to be solved, thereby improving the accuracy of the disclosed solution in solving problems.

[0099] According to an embodiment of the present disclosure, the system operation and maintenance method further includes: updating a pre-built operation and maintenance knowledge base according to the first operation and maintenance question and the answer to the first question.

[0100] In some embodiments, the operation and maintenance knowledge base can, for example, combine the first operation and maintenance question and the answer to the first question into a question-answer pair and save it in the operation and maintenance knowledge base through real-time data collection and automatic update mechanism, thereby completing the data update of the operation and maintenance knowledge base.

[0101] By updating the operation and maintenance knowledge base, the data in the operation and maintenance database can always have the latest operation and maintenance information, improving the accuracy and timeliness of the information.

[0102] According to an embodiment of the present disclosure, the system operation and maintenance method also includes: obtaining historical conversations related to the user; when the user has conducted multiple rounds of conversations, using a pre-trained large language model to process the second operation and maintenance question and the historical conversations, and outputting the answer to the first question.

[0103] In some embodiments, when the user has multiple rounds of conversations, for example, the context tracking mechanism is used to obtain context information in the multiple rounds of conversations of the user. Then, the associated information with the second operation and maintenance problem is determined based on the keywords in the online information, and finally the historical conversation and the second operation and maintenance problem are input into the pre-trained large language model together with the associated information for processing.

[0104] By tracking and processing historical conversations, the disclosed solution can process complex multi-round conversations according to the context, thereby improving the accuracy of the processing of the disclosed solution. According to an embodiment of the present disclosure, the system operation and maintenance method also includes: determining an execution object of the operation and maintenance script; and after executing the operation and maintenance script on the execution object, continuously monitoring the execution object.

[0105] In some embodiments, the execution object can be determined according to the operations processed by the operation and maintenance script.

[0106] For example, if the operation and maintenance script is used to configure the database, the execution object is the database. After the operation and maintenance script is executed on the database, the database is monitored. For example, the server status and system logs of the database system can be monitored, and the operation and maintenance personnel can be notified immediately if an abnormality is found.

[0107] For another example, a preset threshold can be set for the number of server accesses to the database. When the number of server accesses exceeds the preset threshold, it will be determined as a suspected risk access, and an alarm will be sent to the operation and maintenance personnel.

[0108] By continuously monitoring the execution object, the disclosed solution can timely discover and handle potential problems, thereby improving the stability of operation.

[0109] Based on the above system operation and maintenance method, the present disclosure also provides a system operation and maintenance device. Figure 7 The device is described in detail.

[0110] Figure 7 The structural block diagram of the system operation and maintenance device according to an embodiment of the present disclosure is schematically shown.

[0111] like Figure 7 As shown, the system operation and maintenance device 700 of this embodiment includes an enhancement module 710 , a processing module 720 , a generation module 730 and an execution module 740 .

[0112] The enhancement module 710 is used to enhance the prompt word of the first operation and maintenance question in response to the user inputting the first operation and maintenance question to obtain the second operation and maintenance question. In one embodiment, the enhancement module 710 can be used to perform the operation S210 described above, which will not be repeated here.

[0113] The processing module 720 is used to process the second operation and maintenance problem using a pre-trained large language model and output an answer to the first question, where the pre-trained large language model is obtained by at least one of fine-tuning and preference optimization based on specific training samples. In one embodiment, the processing module 720 can be used to perform the operation S220 described above, which will not be repeated here.

[0114] The generation module 730 is used to generate an operation and maintenance script according to the first operation and maintenance question and the answer to the first question. In one embodiment, the generation module 730 can be used to perform the operation S230 described above, which will not be described in detail here.

[0115] The execution module 740 is used to execute the operation and maintenance script to implement the operation and maintenance operation corresponding to the first operation and maintenance problem. In one embodiment, the execution module 740 can be used to execute the operation S240 described above, which will not be described in detail here.

[0116] According to an embodiment of the present disclosure, any multiple modules among the enhancement module 710, the processing module 720, the generation module 730 and the execution module 740 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the enhancement module 710, the processing module 720, the generation module 730 and the execution module 740 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in any appropriate combination of any of them. Alternatively, at least one of the enhancement module 710, the processing module 720, the generation module 730 and the execution module 740 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be executed.

[0117] According to an embodiment of the present disclosure, the processing module 720 includes a first acquisition submodule, a first construction submodule and a fine-tuning submodule.

[0118] The first acquisition submodule is used to acquire operation and maintenance data, where the operation and maintenance data includes at least one of operation and maintenance documents, operation and maintenance logs, and webpage operation and maintenance knowledge.

[0119] The first construction submodule is used to construct an operation and maintenance data set based on the operation and maintenance data, and the operation and maintenance data set is characterized by at least one of an operation and maintenance knowledge question and answer pair and an operation and maintenance knowledge graph.

[0120] The fine-tuning submodule is used to fine-tune the initial large language model using the operation and maintenance dataset to obtain a pre-trained large language model.

[0121] According to an embodiment of the present disclosure, the processing module 720 further includes a second acquisition submodule, a second construction submodule and a training submodule.

[0122] The second acquisition submodule is used to acquire user evaluations corresponding to the operation and maintenance data.

[0123] The second construction submodule is used to construct a preference data set based on user evaluation and operation and maintenance data.

[0124] The training submodule is used to train the initial large language model or the fine-tuned initial large language model using the preferred data set to obtain a pre-trained large language model.

[0125] According to an embodiment of the present disclosure, the enhancement module 710 includes a third acquisition submodule, a matching submodule, and a determination submodule.

[0126] The third acquisition submodule is used to acquire a vector database, which includes multiple operation and maintenance knowledge question and answer pairs.

[0127] The matching submodule is used to match the first operation and maintenance question with multiple operation and maintenance knowledge question and answer pairs to obtain at least one target question, and the vector distance corresponding to the target question and the first operation and maintenance question is less than a preset threshold.

[0128] The determination submodule is used to determine the second operation and maintenance problem based on the operation and maintenance knowledge question and answer pair corresponding to the first operation and maintenance problem and the target problem.

[0129] According to an embodiment of the present disclosure, the processing module 720 further includes a retrieval submodule, an enhancement submodule and a processing submodule.

[0130] The retrieval submodule is used to retrieve background knowledge related to the second operation and maintenance problem in a pre-built operation and maintenance knowledge base.

[0131] The enhancement submodule is used to enhance the background knowledge of the second operation and maintenance problem by using the background knowledge to obtain the third operation and maintenance problem.

[0132] The processing submodule is used to process the third operation and maintenance question using a pre-trained large language model and output an answer to the second question.

[0133] According to an embodiment of the present disclosure, the retrieval submodule includes an updating unit.

[0134] The updating unit is used to update the pre-built operation and maintenance knowledge base according to the first operation and maintenance question and the answer to the first question.

[0135] According to an embodiment of the present disclosure, the system operation and maintenance device 700 further includes an acquisition module and an output module.

[0136] The acquisition module is used to obtain historical conversations related to the user.

[0137] The output module is used to process the second operation and maintenance question and the historical conversations using a pre-trained large language model when the user has conducted multiple rounds of conversations, and output the answer to the first question.

[0138] According to an embodiment of the present disclosure, the system operation and maintenance device 700 also includes a determination module and a monitoring module.

[0139] The determination module is used to determine the execution object of the operation and maintenance script.

[0140] The monitoring module is used to continuously monitor the execution object after the operation and maintenance script is executed on the execution object.

[0141] Figure 8 A block diagram of an electronic device suitable for implementing a system operation and maintenance method according to an embodiment of the present disclosure is schematically shown.

[0142] like Figure 8 As shown, the electronic device 800 according to an embodiment of the present disclosure includes a processor 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage part 808 to a random access memory (RAM) 803. The processor 801 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a dedicated microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 801 may also include an onboard memory for caching purposes. The processor 801 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0143] In RAM 803, various programs and data required for the operation of electronic device 800 are stored. Processor 801, ROM 802 and RAM 803 are connected to each other via bus 804. Processor 801 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 802 and / or RAM 803. It should be noted that the program can also be stored in one or more memories other than ROM 802 and RAM 803. Processor 801 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0144] According to an embodiment of the present disclosure, the electronic device 800 may further include an input / output (I / O) interface 805, which is also connected to the bus 804. The electronic device 800 may further include one or more of the following components connected to the input / output (I / O) interface 805: an input portion 806 including a keyboard, a mouse, etc.; an output portion 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 808 including a hard disk, etc.; and a communication portion 809 including a network interface card such as a LAN card, a modem, etc. The communication portion 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the input / output (I / O) interface 805 as needed. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as needed, so that the computer program read therefrom is installed into the storage portion 808 as needed.

[0145] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0146] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present disclosure, the computer-readable storage medium may include the ROM 802 and / or RAM 803 described above and / or one or more memories other than ROM 802 and RAM 803.

[0147] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program includes a program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the system operation and maintenance method provided by the embodiment of the present disclosure.

[0148] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 801. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0149] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 809, and / or installed from a removable medium 811. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0150] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 809, and / or installed from the removable medium 811. When the computer program is executed by the processor 801, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0151] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0152] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions. In the above detailed description, various features are combined together in a single embodiment to simplify the present disclosure. This disclosure method should not be interpreted as reflecting such an intention, that is, the embodiment of the subject matter claimed for protection requires more features than the features clearly stated in each claim. Rather, as the following claims reflect, the disclosure lies in less than all features of a single disclosed embodiment.Thus, the following claims are hereby expressly incorporated into the Detailed Description, with each claim standing on its own as a separate preferred embodiment of the disclosure.

[0153] In addition, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present disclosure, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. With respect to the term "comprising" used in the specification or claims, the word is covered in a manner similar to the term "including", as explained in terms of "including," used as a transitional word in the claims. Any term "or" used in the specification of the claims is intended to mean "non-exclusive or".

[0154] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A system operation and maintenance method, characterized in that: The method comprises: In response to a user inputting a first operation and maintenance question, the first operation and maintenance question is enhanced with prompt words to obtain a second operation and maintenance question; Using a pre-trained large language model to process the second operation and maintenance problem and output an answer to the first question, the pre-trained large language model being obtained by at least one of fine-tuning and preference optimization based on specific training samples; Generate an operation and maintenance script according to the first operation and maintenance question and the answer to the first question; and Execute the operation and maintenance script to implement the operation and maintenance operation corresponding to the first operation and maintenance problem.

2. The method according to claim 1, characterized in that The training of the pre-trained large language model includes: Acquiring operation and maintenance data, wherein the operation and maintenance data includes at least one of operation and maintenance documents, operation and maintenance logs, and webpage operation and maintenance knowledge; Constructing an operation and maintenance data set based on the operation and maintenance data, wherein the operation and maintenance data set is characterized by at least one of an operation and maintenance knowledge question and answer pair and an operation and maintenance knowledge graph; The operation and maintenance data set is used to fine-tune the initial large language model to obtain the pre-trained large language model.

3. The method according to claim 2, characterized in that The training of the pre-trained large language model includes: Obtaining user evaluations corresponding to the operation and maintenance data; Building a preference data set based on the user evaluation and the operation and maintenance data; The initial large language model or the fine-tuned initial large language model is trained using the preference data set to obtain the pre-trained large language model.

4. The method according to claim 1, characterized in that: In response to the user raising a first operation and maintenance question, the first operation and maintenance question is enhanced with prompt words to obtain a second operation and maintenance question, including: Acquire a vector database, wherein the vector database includes a plurality of operation and maintenance knowledge question and answer pairs; Matching the first operation and maintenance question with the multiple operation and maintenance knowledge question and answer pairs to obtain at least one target question, wherein a vector distance between the target question and the first operation and maintenance question is less than a preset threshold; The second operation and maintenance problem is determined based on the operation and maintenance knowledge question and answer pair corresponding to the first operation and maintenance problem and the target problem.

5. The method according to claim 1, characterized in that Before using the pre-trained large language model to process the second operation and maintenance problem, the method further includes: Retrieving background knowledge related to the second operation and maintenance problem from a pre-built operation and maintenance knowledge base; Using the background knowledge to enhance the background knowledge of the second operation and maintenance problem, to obtain a third operation and maintenance problem; The third operation and maintenance question is processed using a pre-trained large language model, and an answer to the second question is output.

6. The method according to claim 5, characterized in that The method further comprises: The pre-built operation and maintenance knowledge base is updated according to the first operation and maintenance question and the answer to the first question.

7. The method according to claim 1, characterized in that The method further comprises: Obtaining historical conversations related to the user; In the case where the user has conducted multiple rounds of dialogue, a pre-trained large language model is used to process the second operation and maintenance question and the historical dialogue, and an answer to the first question is output.

8. The method according to claim 1, characterized in that The method further comprises: Determine the execution object of the operation and maintenance script; After the operation and maintenance script is executed on the execution object, the execution object is continuously monitored.

9. A system operation and maintenance device, characterized in that: The device comprises: an enhancement module, configured to, in response to a user inputting a first operation and maintenance question, enhance the first operation and maintenance question with prompt words to obtain a second operation and maintenance question; a processing module, configured to process the second operation and maintenance problem using a pre-trained large language model and output an answer to the first question, wherein the pre-trained large language model is obtained by at least one of fine-tuning and preference optimization based on specific training samples; A generation module, used to generate an operation and maintenance script according to the first operation and maintenance question and the answer to the first question; An execution module is used to execute the operation and maintenance script to implement the operation and maintenance operation corresponding to the first operation and maintenance problem.

10. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

12. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images