An operation and maintenance knowledge automatic updating method, device, equipment and program product

Abstract

The application discloses an operation and maintenance knowledge automatic updating method, device, equipment and program product. The method comprises the following steps: a vector knowledge base is constructed in advance; in response to a correction event of target operation and maintenance knowledge, correction information of the target operation and maintenance knowledge is acquired; the target operation and maintenance knowledge and the correction information are input into a large model to obtain associated knowledge and correction information of the associated knowledge; when the correction information is audited, knowledge tags corresponding to the target operation and maintenance knowledge and the associated knowledge are determined; and based on the knowledge tags corresponding to the target operation and maintenance knowledge and the associated knowledge and the correction information, the target operation and maintenance knowledge and the associated knowledge are updated in the vector knowledge base. The application combines the large model to acquire the correction information, so that the vector knowledge base of operation and maintenance knowledge including a plurality of operation and maintenance documents can be automatically updated, the user does not need to participate in the operation and maintenance updating process all the time, the automatic updating of the operation and maintenance knowledge is realized, the updating efficiency of the operation and maintenance knowledge is improved, the updating cost is reduced, and the stability and confidence of the operation and maintenance knowledge are improved.

Description

Technical Field

[0001] This application belongs to the field of data update technology, and in particular relates to an automatic update method, device, equipment and program product for operation and maintenance knowledge. Background Technology

[0002] Operation and maintenance documentation typically covers operational and maintenance knowledge such as application system emergency manuals, log analysis, and security protection, and can be updated manually.

[0003] However, manually updating operational knowledge is costly and inefficient. In particular, when errors frequently occur in operational knowledge, manual updates can also affect the confidence level of the operational knowledge.

[0004] Based on this, this application provides a method for automatically updating operation and maintenance knowledge. Summary of the Invention

[0005] This application provides a method, apparatus, device, computer-readable storage medium, and computer program product for automatically updating operation and maintenance knowledge, which can improve the efficiency of updating operation and maintenance knowledge, reduce the cost of updating, and improve the stability and confidence of operation and maintenance knowledge.

[0006] In a first aspect, embodiments of this application provide a method for automatically updating operation and maintenance knowledge, the method comprising: A vector knowledge base is pre-built to store operation and maintenance knowledge. Each piece of operation and maintenance knowledge has a corresponding knowledge tag, which includes source document information, knowledge location information within the document, and operation and maintenance entity information. In response to a correction event of the target operation and maintenance knowledge, obtain the correction information of the target operation and maintenance knowledge; The target operation and maintenance knowledge and the correction information are input into the large model, so that the large model generates associated knowledge related to the operation and maintenance knowledge and correction information of the associated knowledge; In response to receiving the approval information for the correction information, determine the knowledge tags corresponding to the target operation and maintenance knowledge and the associated knowledge; Based on the target operation and maintenance knowledge, the knowledge tags corresponding to the associated knowledge, and the correction information, the target operation and maintenance knowledge and the associated knowledge are updated in the vector knowledge base.

[0007] Secondly, embodiments of this application provide an automatic update device for operation and maintenance knowledge, the device comprising: A vector knowledge base construction module is used to pre-build a vector knowledge base, which is used to store operation and maintenance knowledge. Each operation and maintenance knowledge has a corresponding knowledge tag, which includes source document information, knowledge location information within the document, and operation and maintenance entity information. The first correction information acquisition module is used to acquire correction information of the target operation and maintenance knowledge in response to a correction event of the target operation and maintenance knowledge. The second correction information acquisition module is used to input the target operation and maintenance knowledge and the correction information into the large model, so that the large model generates associated knowledge related to the operation and maintenance knowledge and correction information of the associated knowledge; The knowledge tag determination module is used to determine the knowledge tags corresponding to the target operation and maintenance knowledge and the associated knowledge in response to the approval information received from the correction information. The vector knowledge base update module is used to update the target operation and maintenance knowledge and the related knowledge in the vector knowledge base based on the target operation and maintenance knowledge, the knowledge tags corresponding to the related knowledge, and the correction information.

[0008] Thirdly, embodiments of this application provide an electronic device, which includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements any of the possible implementations of the first aspect described above.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the method in any of the possible implementations of the first aspect described above.

[0010] Fifthly, embodiments of this application provide a computer program product in which instructions, when executed by a processor of an electronic device, cause the electronic device to perform a method as described in any of the possible implementations of the first aspect above.

[0011] In this embodiment, a large model is used to obtain correction information for automated updating of a vector knowledge base containing multiple operation and maintenance documents. This eliminates the need for full user involvement in the update process, enabling automated updates of operation and maintenance knowledge. It can also update other knowledge related to erroneous operation and maintenance knowledge, preventing content conflicts or gaps and improving the confidence level of the knowledge. Furthermore, the accuracy of the updated operation and maintenance knowledge is further improved through correction information review. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1A flowchart illustrating an automatic update method for operation and maintenance knowledge provided in an embodiment of this application; Figure 2 An updated schematic diagram provided for an embodiment of this application; Figure 3 A schematic diagram of system interaction provided for an embodiment of this application; Figure 4 A schematic diagram of the structure of an automatic update device for operation and maintenance knowledge provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0014] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0015] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0016] Furthermore, the acquisition, storage, use, and processing of data in this application comply with relevant national laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0017] Current operations and maintenance (O&M) knowledge updates are typically done manually, which is costly and inefficient. For example, multiple O&M documents may need to be manually modified. When there are many modifications or frequent changes, it consumes significant human resources. Furthermore, since O&M knowledge is usually integrated into documents, modifications require manual reading of these documents. When the O&M documents are extensive, manual modification becomes even more inconvenient and inefficient. Similarly, if multiple O&M documents require simultaneous updates, it necessitates manually querying and updating all documents.

[0018] In addition, frequent manual updates may also affect the confidence level of operation and maintenance knowledge. This is because manual updates inevitably involve updating only a single piece of operation and maintenance knowledge, neglecting to update other related operation and maintenance knowledge. This can lead to conflicts or gaps in the content of different operation and maintenance knowledge, thus affecting the confidence level of the operation and maintenance knowledge.

[0019] Therefore, to address the problems of the prior art, embodiments of this application provide a method, apparatus, device, computer-readable storage medium, and computer program product for automatically updating operation and maintenance knowledge. The method for automatically updating operation and maintenance knowledge can be applied to any scenario where there is a need for automatic updates of operation and maintenance knowledge.

[0020] The execution entity in this application embodiment can be any electronic device, such as a server or personal computer. For ease of explanation, the following description will use a server as the execution entity. This server can deploy large models. A large model refers to a deep learning model with a large number of model parameters, typically containing hundreds of millions, tens of billions, hundreds of billions, trillions, or even tens of trillions of model parameters. Large models can also be called foundation models. Pre-training a large model on a large corpus produces a pre-trained model with hundreds of millions of parameters. Such models can adapt to a wide range of downstream tasks and have good generalization ability, such as Large Language Models (LLMs).

[0021] In practical applications, large models only require a small number of samples to fine-tune the pre-trained model before they can be applied to different tasks. Large models can be widely used in fields such as Natural Language Processing (NLP) and Computer Vision. Specifically, they can be applied to computer vision tasks such as Visual Question Answering (VQA), Image Captioning (IC), and Image Generation, as well as natural language processing tasks such as text-based sentiment classification, text summarization, and machine translation. The main application scenarios for large models include digital assistants, intelligent robots, search, online education, office software, e-commerce, and intelligent design.

[0022] In this embodiment, the large model can be used to acquire various operation and maintenance documents, break down these documents to obtain operation and maintenance knowledge, and predict changes to this knowledge. Alternatively, the large model can also be deployed on terminals used by operation and maintenance personnel to obtain relevant data through data interaction with a server.

[0023] The automatic update method for operation and maintenance knowledge provided in the embodiments of this application will be introduced first.

[0024] Figure 1 This is a flowchart illustrating an automatic update method for operation and maintenance knowledge provided in an embodiment of this application. Figure 1 As shown in the embodiments of this application, the automatic updating method for operation and maintenance knowledge includes the following steps: S110: Pre-build a vector knowledge base.

[0025] S120: In response to a correction event of the target operation and maintenance knowledge, obtain the correction information of the target operation and maintenance knowledge.

[0026] S130: Input the target operation and maintenance knowledge and correction information into the large model, so that the large model generates related knowledge associated with the operation and maintenance knowledge and correction information of the related knowledge.

[0027] S140: In response to the approval information received regarding the correction information, determine the knowledge tags corresponding to the target operation and maintenance knowledge and related knowledge.

[0028] S150: Based on the knowledge tags and correction information corresponding to the target operation and maintenance knowledge and related knowledge, update the target operation and maintenance knowledge and related knowledge in the vector knowledge base.

[0029] In this embodiment, a large model is used to obtain correction information for automated updating of a vector knowledge base containing multiple operation and maintenance documents. This eliminates the need for full user involvement in the update process, enabling automated updates of operation and maintenance knowledge. It can also update other knowledge related to erroneous operation and maintenance knowledge, preventing content conflicts or gaps and improving the confidence level of the knowledge. Furthermore, the accuracy of the updated operation and maintenance knowledge is further improved through correction information review.

[0030] It should be noted that the vector knowledge base constructed in step S110 is used to store operation and maintenance knowledge. Each piece of operation and maintenance knowledge has a corresponding knowledge tag, which includes source document information, knowledge location information within the document, and operation and maintenance entity information. More specifically, the vector knowledge base stores operation and maintenance knowledge vectors and knowledge tags obtained by vectorizing the original data of operation and maintenance knowledge. Operation and maintenance knowledge can be represented by operation and maintenance knowledge vectors. Furthermore, the vector knowledge base can also store the original data of operation and maintenance knowledge. Each piece of operation and maintenance knowledge has a corresponding operation and maintenance knowledge vector and knowledge tag. The knowledge tag can also be understood as metadata used to describe and manage operation and maintenance knowledge. It is bound to the operation and maintenance knowledge and stored in the vector knowledge base to facilitate knowledge retrieval and traceability. An operation and maintenance entity is a specific object or component that is monitored, managed, and maintained. For example, an operation and maintenance entity can be an entity representing the application system associated with the operation and maintenance knowledge, such as a transaction processing system or a clearing system.

[0031] In some embodiments, for S110, the vector knowledge base is constructed in the following manner: Multiple operation and maintenance documents of the application system are collected through multiple data collection interfaces; Each operation and maintenance document is broken down to obtain the operation and maintenance knowledge of each document and the knowledge tags of each operation and maintenance knowledge. Multiple operational and maintenance knowledge items and their corresponding knowledge tags are stored in a vector knowledge base.

[0032] For example, the application system could be an SMS sending system, etc. Depending on the different data sources of the application system, corresponding data acquisition interfaces can be used.

[0033] Specifically, operations and maintenance (O&M) documentation can be various types of data obtained from application systems through various channels in the production environment, including but not limited to emergency group chat data, event ticket data, problem ticket data, and change ticket data. O&M documentation obtained from different data sources can be collected by calling different data collection interfaces.

[0034] For example, group chat data can be obtained by calling the API of instant messaging software, and event tickets, problem tickets, and change tickets can be obtained by calling the API of operation and maintenance big data.

[0035] After obtaining the operations and maintenance (O&M) documentation, it can be input into a large model to break it down, yielding O&M knowledge and knowledge tags. The source document information in the knowledge tags represents the storage location of the O&M documentation within the application system, which can be represented by a file storage path, etc. The knowledge location information within the document represents the position of the O&M knowledge within the document, which can be represented by specific chapters, paragraphs, etc. For example, the knowledge tag may include the original file path, a unique chapter identifier, a hierarchical path, a block number, a version number, an O&M entity, and an audit object, etc. Here, the source document information is specified as the original file path, and the knowledge location information within the document is specified as the hierarchical path, the unique chapter identifier, and the block number.

[0036] Understandably, the operation and maintenance documents are broken down into knowledge points according to chapters, etc. After the knowledge points are vectorized to obtain knowledge vectors, the knowledge vectors are stored in the vector knowledge base, and knowledge tags such as the original file path are recorded so that operation and maintenance entity recognition and document restoration can be performed based on the knowledge tags.

[0037] Table 1 is an example of a knowledge tag and operation and maintenance knowledge provided in the embodiments of this application, as shown in Table 1.

[0038] Table 1

[0039] The original file path represents the path of the corresponding operations and maintenance (O&M) document. The unique chapter identifier represents the chapter of the O&M document within the O&M document. The hierarchical path is a further refinement of the chapter, representing the position of the O&M document within the O&M document. Based on this hierarchical path, the O&M document can be queried within the O&M document. The block number can be a further refinement of the hierarchical path, representing the specific line of the specific paragraph of the O&M document within the O&M document. The content represents the specific text content of the O&M document, i.e., the original data of the O&M document. The timestamp can represent the splitting time and the knowledge update time. When the O&M document is first obtained, the timestamp represents the splitting time; when the O&M document is subsequently updated, the timestamp represents the knowledge update time.

[0040] Based on the examples of knowledge tags and operation and maintenance knowledge mentioned above, a document correction event may include the source document information of the knowledge to be updated, the review object of the knowledge to be updated, the knowledge update time, and the updated version. The review object can be any person capable of reviewing operation and maintenance knowledge. The review object for each piece of operation and maintenance knowledge can be a fixed person or a person randomly selected from the list of review objects; this embodiment of the application does not impose any restrictions on this.

[0041] In addition, the collected operation and maintenance documents may include both non-text and structured operation and maintenance data. When splitting non-text operation and maintenance data, data type conversion can be performed first to obtain operation and maintenance knowledge. Structured operation and maintenance data, such as event tickets, allows for direct identification of the corresponding operation and maintenance entity based on the information. Furthermore, the relevant fields of the structured operation and maintenance data can be integrated to obtain text-based operation and maintenance knowledge.

[0042] In this embodiment, incremental knowledge is automatically collected through multiple data acquisition interfaces to obtain richer operation and maintenance (O&M) documents, thereby improving the comprehensiveness of the O&M knowledge included in the vector knowledge base. The O&M documents are segmented using preset categories and a large model to construct the vector knowledge base. This facilitates updating the O&M knowledge in the vector knowledge base when erroneous knowledge is reported, ensuring the accuracy of the O&M knowledge in the vector knowledge base and improving the accuracy of the O&M documents subsequently presented to O&M personnel based on this vector knowledge base. In some embodiments, O&M personnel can interact with the large model through a terminal. For example, O&M personnel can send O&M knowledge retrieval text through the terminal, and the large model retrieves the O&M knowledge by querying the vector knowledge base and displays it to the O&M personnel.

[0043] In some embodiments, before obtaining the correction information of the target operation and maintenance knowledge in response to a correction event of the target operation and maintenance knowledge in step S120, the method may further include: In response to the operation and maintenance knowledge retrieval operation, determine the operation and maintenance entity to be retrieved; In the vector knowledge base, identify at least one piece of operational knowledge that matches the operational entity to be retrieved; Determine the search results from at least one operational knowledge.

[0044] The operation and maintenance entity to be retrieved is the operation and maintenance entity associated with the operation and maintenance knowledge retrieval text entered by the operation and maintenance personnel.

[0045] For example, if an operations and maintenance (O&M) personnel input the O&M knowledge retrieval text "query single transaction limit rules" through the terminal, they can identify the O&M entity to be retrieved based on this O&M knowledge retrieval text.

[0046] Using the previous example, the operations and maintenance entity may include transaction limit rules, specifically transaction limit rules for different regions, transaction limit rules for different transaction users, etc.

[0047] When determining the operation and maintenance entity to be retrieved, all operation and maintenance entities related to the operation and maintenance knowledge retrieval text can be retrieved. Specifically, this can be achieved through text similarity detection or through large model derivation. This application embodiment does not limit this.

[0048] For example, in a pre-built list of operations and maintenance entities, the operations and maintenance entity with the highest text similarity to the operations and maintenance knowledge retrieval text is matched. Multiple operations and maintenance entities can also be identified; this embodiment of the application does not limit this.

[0049] Then, at least one piece of operational knowledge that matches the operational entity can be identified in the vector knowledge base.

[0050] Specifically, since operational knowledge can be obtained by decomposing operational documents into a large model, and during decomposition, a knowledge tag can be obtained for each piece of operational knowledge, the knowledge tag includes source document information, knowledge location information within the document, and operational entity information. Therefore, each piece of operational knowledge can have at least one tag representing the operational entity to which it belongs. Thus, during matching, at least one piece of operational knowledge belonging to that operational entity can be determined based on the knowledge tag of each piece of operational knowledge and the corresponding operational entity.

[0051] For example, if the identified operation and maintenance entity is entity 1, and there are existing operation and maintenance knowledge 1~3, the operation and maintenance entity information of operation and maintenance knowledge 1~3 represents that operation and maintenance knowledge 1~3 is related to entity 1~3 respectively, then operation and maintenance knowledge 1 can be obtained by matching based on entity 1.

[0052] Subsequently, by retrieving the operation and maintenance knowledge text, for each identified operation and maintenance entity, at least one operation and maintenance knowledge matching that entity is used for similarity matching, avoiding interference from irrelevant knowledge, and obtaining operation and maintenance knowledge with high similarity to the operation and maintenance knowledge retrieval text as the retrieval result.

[0053] Of course, if multiple pieces of operational knowledge are acquired, in order to ensure the continuity of reading for operational personnel, a large model can be used to logically integrate the various pieces of operational knowledge and generate search results to be displayed to operational personnel.

[0054] Specifically, various operational and maintenance knowledge items and integration prompts can be input into the large model, which will then output logically integrated results as search results. The integration prompts can be set as needed, such as "Please integrate the logical relationships between various operational and maintenance knowledge items," etc.

[0055] Following the previous example, input the various operation and maintenance knowledge and integration prompts into the large model. The logical integration result output by the large model can be "The single transaction limit rules for region A are x and m, the single transaction limit rules for region B are x and y, the same rules in the single transaction limit rules for region A and region B include x, and the different rules include m and y".

[0056] In this embodiment, the operation and maintenance entity to be retrieved is first identified. The subsequent process of matching operation and maintenance knowledge based on this entity can be understood as Retrieval-Augmented Generation (RAG). RAG enables the large model to acquire more retrieval information, reducing illusions and resulting in more accurate operation and maintenance knowledge, thus improving the user experience for operation and maintenance personnel. Furthermore, this vector knowledge base can be updated in real time, and the included operation and maintenance knowledge is highly real-time, further improving the accuracy of the operation and maintenance knowledge obtained by the large model based on this vector knowledge base.

[0057] It should be noted that operations and maintenance (O&M) personnel can perform O&M knowledge retrieval operations at any time. For example, this can be performed before retrieving the correction information for the target O&M knowledge in response to a correction event. After the search results are displayed to the O&M personnel, they can provide feedback on the corrections to the O&M knowledge included in the displayed search results.

[0058] Correction events are events used to trigger the correction process of operational knowledge. Correction events can take various forms, such as correction feedback events from operations and maintenance personnel, monitoring alarm events from automated fault monitoring, etc.

[0059] In some embodiments, to ensure that the correction event that triggers the modification of a certain operation and maintenance knowledge is not accidental, the following may be included before step S120: In response to feedback on any operational knowledge, accumulate the number of feedback on any operational knowledge. When the number of correction feedbacks for any operation and maintenance knowledge exceeds a preset threshold, that operation and maintenance knowledge is taken as the target operation and maintenance knowledge, and a correction event for the target operation and maintenance knowledge is generated.

[0060] In this embodiment, the correction event manifests as a correction feedback event. For example, with a preset threshold of 3, if the operations and maintenance personnel determine that the displayed operations and maintenance knowledge contains an error, they can input correction feedback information via the terminal, such as "The single transaction limit rule for region A you replied to has an error." At this time, the number of correction feedbacks for this operations and maintenance knowledge can be accumulated. When the number of correction feedbacks exceeds 3, a correction event representing the displayed operations and maintenance knowledge is generated.

[0061] In this embodiment, when the number of correction feedbacks exceeds a preset threshold, it indicates that the errors in the maintenance knowledge previously reported by maintenance users are not isolated incidents, and the maintenance knowledge may indeed have problems and needs to be updated. By controlling the relationship between the number of correction feedbacks and the preset threshold, the necessity and timing of corrections are ensured.

[0062] In some embodiments, in S120, the correction information of the target operation and maintenance knowledge can be obtained by outputting a preset question statement.

[0063] For example, when the correction feedback information entered by the operations and maintenance personnel through the terminal does not include the correction information for the target operations and maintenance knowledge, the correction information for the target operations and maintenance knowledge can be obtained by displaying a preset question. The preset question can be set as needed. For example, the preset question could be "Please enter the correction information for the operations and maintenance knowledge that you believe is incorrect," etc.

[0064] Of course, if the correction feedback information entered by the operations and maintenance personnel includes correction information for the target operations and maintenance knowledge, then there is no need to output the preset question statement.

[0065] In some embodiments, when determining the associated knowledge and the correction information of the associated knowledge in S130, an associated query prompt template can be obtained; Based on the target operation and maintenance knowledge and the related query prompt template, generate related query prompt information; The related query prompts and the correction information of the target operation and maintenance knowledge are input into the large model. The large model then searches for related operation and maintenance documents in the database associated with the operation and maintenance entities of the target operation and maintenance knowledge, and searches for related knowledge associated with the target operation and maintenance knowledge in the operation and maintenance knowledge corresponding to the related operation and maintenance documents included in the vector knowledge base, and generates correction information for the related knowledge.

[0066] The related query suggestion template provides suggestions for generating related knowledge and correction information for the large model.

[0067] For example, the related query suggestion template may include a fixed suggestion statement and information to be replaced. The related query suggestion template can be set as needed, such as "In a given complete document, find all knowledge points with 'target knowledge point X' as the core, and find all knowledge points with causal, supplementary, pre- / post-, related, or dependent relationships with it."

[0068] For example, "Target Knowledge Point X" can be included as information to be replaced in the related query suggestion template. Then, by replacing the information to be replaced in the related query suggestion template with the target operation and maintenance knowledge, the related query suggestion information can be obtained.

[0069] Next, the associated query prompts and the corrected information for the target operations and maintenance knowledge are input into the large model, enabling the large model to query the associated knowledge from the database associated with the operations and maintenance entity of the target operations and maintenance knowledge. For example, this database can be a Configuration Management Database (CMDB), a core component of Information Technology Service Management (ITSM), used to centrally store and manage IT infrastructure, services, and their relationships.

[0070] Specifically, first, identify the operations and maintenance entity associated with the operation and maintenance knowledge. Then, determine the query scope in the vector knowledge base within the database associated with the operation and maintenance entity. Within this query scope, identify the related knowledge associated with the operation and maintenance knowledge. This query scope includes all related operation and maintenance documents and the operation and maintenance knowledge included in the operation and maintenance documents associated with the target operation and maintenance knowledge's operation and maintenance entity.

[0071] In other words, the query scope is the operation and maintenance knowledge collected from the operation and maintenance documents of the direct upstream and downstream systems of the operation and maintenance entity. The upstream and downstream systems can obtain the knowledge through the CMDB database.

[0072] In this embodiment of the application, by querying within a certain range, interference from the knowledge of other operation and maintenance entities is avoided, and more accurate association knowledge is obtained.

[0073] In some embodiments, for S130, the correction information of the associated knowledge can be inferred by the large model through the correction information of the target operation and maintenance knowledge. The correction information of the associated knowledge may be the same as or different from the correction information of the target operation and maintenance knowledge, depending on the degree of association between the associated knowledge and the target operation and maintenance knowledge. The stronger the association, the more similar the correction information of the associated knowledge and the correction information of the target operation and maintenance knowledge may be.

[0074] In some embodiments, regarding S140, the correction information for the target operation and maintenance knowledge and related knowledge can be reviewed by the review object. Specifically, the correction information for the target operation and maintenance knowledge and related knowledge is reviewed through the following operations: In the vector knowledge base, the review objects of target operation and maintenance knowledge and related knowledge are determined by the knowledge tags of target operation and maintenance knowledge and related knowledge respectively; The correction information for the target operation and maintenance knowledge and the correction information for related knowledge will be displayed to the corresponding review objects respectively.

[0075] Each auditing entity can audit each correction information and obtain the audit result. If the audit result is that the correction information of the target operation and maintenance knowledge and related knowledge has passed the audit, then step S150 can be executed. However, if the auditing entity reports that any correction information is incorrect, a command to obtain correction update information can be sent to the auditing entity. The auditing entity can send correction update information through the terminal. In response to the correction information update operation, step S150 is executed.

[0076] In this embodiment, through an audit process, reinforcement learning is introduced into the large model to proactively push updated solutions for erroneous knowledge points and their inferences to the audit recipient for confirmation of whether modifications are necessary. Based on the feedback from the audit recipient, the accuracy of subsequent predictions of other related knowledge can be improved autonomously, avoiding the need for offline fine-tuning of the large model and ensuring its continued usability.

[0077] In some embodiments, new operational knowledge may emerge due to modifications. Therefore, the target operational knowledge and related knowledge may also be newly added knowledge, and the corresponding knowledge to be updated will not exist in the vector knowledge base. For S150, based on the knowledge tags corresponding to the target operational knowledge and related knowledge, the knowledge to be updated that already exists in the vector knowledge base and / or the knowledge to be added that does not exist in the vector knowledge base can be determined. For each piece of knowledge to be updated, the knowledge in the vector knowledge base is updated based on the correction information of the knowledge to be updated; For each piece of knowledge to be added, based on the correction information of the knowledge to be added, the knowledge to be added and its corresponding knowledge tag are added to the vector knowledge base.

[0078] For example, the knowledge to be updated may include the target operational knowledge for which operational personnel provide correction feedback, as well as related knowledge associated with the target operational knowledge.

[0079] When determining the knowledge to be updated and the knowledge to be added, it can be determined by function calculation and combined with knowledge tags. This application embodiment does not limit this.

[0080] It is understandable that update operations can include two types. The first is replacing the correction information of the knowledge to be updated with the knowledge to be updated, which is a modification operation. The second is adding the correction information of the knowledge to be added to the vector knowledge base, so that the vector knowledge base includes the knowledge to be added and its corresponding knowledge tag, which is an addition operation.

[0081] In this embodiment, since the knowledge tags include in-document knowledge location information for operational knowledge, the knowledge to be added and updated can be quickly identified through these tags. Subsequently, two update operations are used to update both existing and previously non-existent operational knowledge, implementing an iterative update mechanism to ensure the stability of subsequent operational document content updated based on this vector knowledge base and to reflect changes in operational knowledge. Furthermore, since the large model can output response information to operational personnel based on this operational knowledge, updating the operational knowledge in the vector knowledge base can also improve the accuracy of the response information output by the large model.

[0082] It should be noted that the vector knowledge base can store both operation and maintenance knowledge vectors and knowledge tags at the same time, as well as the original data of operation and maintenance knowledge, operation and maintenance knowledge vectors, and knowledge tags at the same time.

[0083] When executing S150, for the former, the updated objects include the operation and maintenance knowledge vector and knowledge tags. For the latter, the updated objects include the original data of operation and maintenance knowledge, the operation and maintenance knowledge vector, and knowledge tags.

[0084] Taking updating target operation and maintenance knowledge as an example, updating the operation and maintenance knowledge vector of target operation and maintenance knowledge can be understood as determining the original data of the updated target operation and maintenance knowledge based on the correction information of the target operation and maintenance knowledge, and then performing vectorization processing on the original data of the updated target operation and maintenance knowledge to obtain the updated operation and maintenance knowledge vector.

[0085] Similarly, updating knowledge tags can be done by updating the content included in the knowledge tags using the original data of the updated target operation and maintenance knowledge.

[0086] For example, if the chapter unique identifier of the original data of the updated target operation and maintenance knowledge changes, then the original chapter unique identifier included in the knowledge tag will be replaced with the chapter unique identifier of the original data of the updated target operation and maintenance knowledge.

[0087] In some embodiments, the operation and maintenance knowledge in the vector knowledge base is obtained by splitting the operation and maintenance documents. Therefore, after updating the operation and maintenance knowledge in the vector knowledge base, the operation and maintenance documents can be updated at any time after executing S150.

[0088] For example, in response to an operation and maintenance document retrieval operation, the document information of the operation and maintenance document to be retrieved is determined; The document information is matched with the knowledge tags in the vector knowledge base to find multiple operation and maintenance knowledge related to the operation and maintenance document to be obtained. Based on the multiple operational and maintenance knowledge points found, the operational and maintenance documents to be obtained are reconstructed.

[0089] For example, the document information represents the source document information of the operation and maintenance document to be obtained, and the original file path can be specified.

[0090] For example, operations and maintenance (O&M) personnel can send a text message such as "Please show me O&M document A" via a terminal. This text message can then be used to identify the O&M document to be retrieved, i.e., O&M document A. Since the knowledge tags obtained from splitting the O&M document record its original path, the knowledge tags of each O&M knowledge stored in the vector knowledge base can be matched with the document information of the O&M document to be retrieved. This allows for the identification of multiple O&M knowledge items associated with the document, and ultimately, the O&M document to be retrieved can be reconstructed.

[0091] For example, if the original file path in the knowledge tag of each operation and maintenance knowledge stored in the vector knowledge base is consistent with the original file path of the document information of the operation and maintenance document to be obtained, then the operation and maintenance knowledge matches the operation and maintenance document to be obtained, and the operation and maintenance document to be obtained can be reconstructed based on the operation and maintenance knowledge.

[0092] For example, operations and maintenance personnel send a download command via terminal to "retriev document D1 containing knowledge point X". By querying the vector knowledge base, they obtain all the operations and maintenance knowledge contained in document D1. Then, based on the knowledge tags of each piece of operations and maintenance knowledge, i.e., the original file path, unique chapter identifier, hierarchical path, block number, version number, operations and maintenance entity, etc., the knowledge is rearranged into a new document D1'. At this point, document D1' contains the latest knowledge point X', thus ensuring the timely updating of the knowledge document.

[0093] In this embodiment, when operations and maintenance personnel need to access operations and maintenance documents, the documents are updated to ensure that the operational knowledge included in the documents is up-to-date. Updating based on a vector knowledge base avoids the instability caused by completely reconstructing the documents from templates. Iterative document updates are achieved through partial updates to the vector knowledge base.

[0094] It should be noted that the operation and maintenance documentation can also be updated immediately after the vector knowledge base is updated, without waiting for the operation and maintenance personnel to send a request to obtain the operation and maintenance documentation.

[0095] In some embodiments, after performing S150, a document correction event may be added to the event repository. The document correction event includes the source document information of the updated knowledge, the audit object of the updated knowledge, the knowledge update time, and the updated version. After reconstructing the maintenance documents to be retrieved, the method also includes: Based on document correction events, revision information is generated in the maintenance documents to be retrieved.

[0096] For example, the updated knowledge may be the aforementioned target operation and maintenance knowledge and related knowledge.

[0097] When updating operations and maintenance (O&M) documentation, correction records can also be recorded. Specifically, a document correction event is added to the event repository. This event represents the changes made to the O&M documentation to be retrieved. The correction record may include information such as the source document information and the reviewing entity for each updated piece of knowledge in the document. The source document information may include the original file path of the document. This document correction event can also be displayed to O&M personnel.

[0098] For example, revision information is used to characterize the revision history of each modification in the operations and maintenance document to be acquired. This revision history allows identification of which operations and maintenance knowledge has been modified in the document. Specifically, when generating revision information, knowledge tags and other information related to the modified operations and maintenance knowledge in the document can be recorded based on document correction events.

[0099] In this embodiment, a document correction event is added to synchronously save explicit revision records. If problems arise later, these document correction events can be used to trace responsibility or find the root cause of the problem. By generating revision information, changes in operational knowledge can be reflected.

[0100] Figure 2 An update diagram provided for an embodiment of this application, such as... Figure 2 As shown.

[0101] After obtaining the correction information for the target operation and maintenance knowledge and related knowledge, the vector knowledge base and event storage base can be updated respectively using the correction information for the target operation and maintenance knowledge and related knowledge.

[0102] Based on the automatic update method for operation and maintenance knowledge provided in the above embodiments, this application also provides a specific implementation of an automatic update system for operation and maintenance knowledge. Please refer to the following embodiments.

[0103] Figure 3 A system interaction diagram provided for an embodiment of this application, such as Figure 3 As shown.

[0104] An automated system for updating operational knowledge can include a knowledge update module, a document access layer, and a knowledge Q&A module. The document access layer can also be called a knowledge collection module.

[0105] The document access layer is used to split and reorganize the operation and maintenance documents collected by operation and maintenance personnel from other application systems, and add the split operation and maintenance knowledge and knowledge tags to the vector knowledge base.

[0106] Specifically, existing operation and maintenance documents in the production environment, such as application system encyclopedias, emergency handling manuals, and service handling manuals, are processed through a large model to extract a number of operation and maintenance knowledge as metadata. At the same time, knowledge tags for operation and maintenance knowledge are determined and stored in a vector knowledge base.

[0107] Upon receiving a request from operations and maintenance (O&M) personnel to export O&M documents (i.e., a download instruction for O&M documents) via their terminal, the document access layer can query the latest O&M knowledge from the vector knowledge base, generate the latest O&M documents, and provide them to the O&M personnel for download. Through this method, the storage and updating of O&M documents become more efficient and intelligent.

[0108] The knowledge-based question-answering module provides services through RAG and large-scale models. It collects feedback from operations and maintenance personnel regarding corrections to target operations and maintenance knowledge, along with related knowledge predicted by the large-scale model and its correction information, and transmits this information to the knowledge update module. The knowledge update module compares the incoming operations and maintenance knowledge with the original knowledge in the vector knowledge base based on knowledge tags, determining whether it is new or modified, and submits the correction information to reviewers for approval. After approval, the knowledge update module updates the vector knowledge base and regenerates the operations and maintenance documents.

[0109] The knowledge update module can also be used to pass updated operation and maintenance documents to the knowledge question and answer module, so as to realize the synchronous update of large model knowledge and effectively improve the accuracy of question answers.

[0110] In this embodiment, a complete closed loop of knowledge management is realized, including knowledge collection, document updates, knowledge Q&A, and feedback updates, while simultaneously realizing iterative updates of operation and maintenance documents and near real-time updates of the Q&A system.

[0111] Based on the automatic update method for operation and maintenance knowledge provided in the above embodiments, this application also provides a specific implementation of the automatic update device for operation and maintenance knowledge. Please refer to the following embodiments.

[0112] like Figure 4 As shown, the automatic update device 400 for operation and maintenance knowledge provided in this application embodiment includes the following modules: The vector knowledge base construction module 401 is used to pre-build a vector knowledge base, which is used to store operation and maintenance knowledge. Each operation and maintenance knowledge has a corresponding knowledge tag, which includes source document information, knowledge location information within the document, and operation and maintenance entity information. The first correction information acquisition module 402 is used to acquire correction information of the target operation and maintenance knowledge in response to a correction event of the target operation and maintenance knowledge. The second correction information acquisition module 403 is used to input the target operation and maintenance knowledge and the correction information into the large model, so that the large model generates associated knowledge related to the operation and maintenance knowledge and correction information of the associated knowledge; The knowledge tag determination module 404 is used to determine the knowledge tags corresponding to the target operation and maintenance knowledge and the associated knowledge in response to the approval information received from the correction information. The vector knowledge base update module 405 is used to update the target operation and maintenance knowledge and the related knowledge in the vector knowledge base based on the target operation and maintenance knowledge, the knowledge tags corresponding to the related knowledge, and the correction information.

[0113] The automatic update device 400 for the aforementioned maintenance knowledge is described in detail below: Optionally, the vector knowledge base update module 405 is specifically used to determine, based on the knowledge tags corresponding to the target operation and maintenance knowledge and the associated knowledge, the knowledge to be updated that already exists in the vector knowledge base and / or the knowledge to be added that does not exist in the vector knowledge base; For each piece of knowledge to be updated, the knowledge in the vector knowledge base is updated based on the correction information of the knowledge to be updated; For each piece of knowledge to be added, based on the correction information of the knowledge to be added, the knowledge to be added and the corresponding knowledge tag are added to the vector knowledge base.

[0114] Optionally, the device further includes: The document update module is used to respond to the operation and maintenance document retrieval operation and determine the document information of the operation and maintenance document to be retrieved; The document information is matched with the knowledge tags in the vector knowledge base to find multiple operation and maintenance knowledge related to the operation and maintenance document to be obtained; Based on the multiple operational and maintenance knowledge points found, the operational and maintenance documents to be obtained are reconstructed.

[0115] Optionally, the device further includes: The document correction event adding module is used to add a document correction event in the event repository after updating the target operation and maintenance knowledge and the related knowledge in the vector knowledge base based on the knowledge tags corresponding to the target operation and maintenance knowledge and the related knowledge and the correction information. The document correction event includes the source document information of the updated knowledge, the review object of the updated knowledge, the knowledge update time and the updated version. Based on this, the device further includes: The revision information generation module is used to generate revision information in the maintenance document to be acquired based on the document correction event after the document is reconstructed.

[0116] Optionally, the second correction information acquisition module 403 is specifically used to acquire the related query prompt template; Based on the target operation and maintenance knowledge and the associated query prompt template, generate associated query prompt information; The associated query prompt information and the correction information of the target operation and maintenance knowledge are input into the large model, so that the large model can find the associated operation and maintenance document in the database associated with the operation and maintenance entity of the target operation and maintenance knowledge, and find the associated knowledge associated with the target operation and maintenance knowledge in the operation and maintenance knowledge corresponding to the associated operation and maintenance document included in the vector knowledge base, and generate the correction information of the associated knowledge.

[0117] Optionally, the device further includes: The retrieval module is used to determine the operation and maintenance entity to be retrieved in response to the operation and maintenance knowledge retrieval operation before obtaining the correction information of the target operation and maintenance knowledge in response to the correction event of the target operation and maintenance knowledge; In the vector knowledge base, at least one piece of operational knowledge that matches the operational entity to be retrieved is identified; The search results are determined from at least one of the aforementioned operational knowledge.

[0118] Optionally, the vector knowledge base construction module 401 is specifically used to collect multiple operation and maintenance documents of the application system through multiple data collection interfaces; Each of the operation and maintenance documents is broken down to obtain the operation and maintenance knowledge of each of the operation and maintenance documents and the knowledge tags of each of the operation and maintenance knowledge. The various operational and maintenance knowledge items and their corresponding knowledge tags are stored in a vector knowledge base.

[0119] Optionally, the device further includes: The correction triggering module is used to respond to a correction event of the target operation and maintenance knowledge, and before obtaining the correction information of the target operation and maintenance knowledge, respond to the correction feedback of any operation and maintenance knowledge and accumulate the correction feedback number of any operation and maintenance knowledge. When the number of correction feedbacks for any of the operation and maintenance knowledge exceeds a preset threshold, the any of the operation and maintenance knowledge is taken as the target operation and maintenance knowledge, and a correction event for the target operation and maintenance knowledge is generated.

[0120] In this embodiment, a large model is used to obtain correction information for automated updating of a vector knowledge base containing multiple operation and maintenance documents. This eliminates the need for full user involvement in the update process, enabling automated updates of operation and maintenance knowledge. It can also update other knowledge related to erroneous operation and maintenance knowledge, preventing content conflicts or gaps and improving the confidence level of the knowledge. Furthermore, the accuracy of the updated operation and maintenance knowledge is further improved through correction information review.

[0121] Based on the automatic update method for operation and maintenance knowledge provided in the above embodiments, this application also provides specific implementation methods for electronic devices. Figure 5 A schematic diagram of an electronic device 500 provided in an embodiment of this application is shown.

[0122] Electronic device 500 may include processor 510 and memory 520 storing computer program instructions.

[0123] Specifically, the processor 510 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0124] Memory 520 may include mass storage for data or instructions. For example, and not limitingly, memory 520 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 520 may include removable or non-removable (or fixed) media. Where appropriate, memory 520 may be internal or external to electronic device 500. In a particular embodiment, memory 520 is a non-volatile solid-state memory.

[0125] In a specific embodiment, the memory 520 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 520 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 520 and is called and executed by the processor 510. The processor 510 reads and executes the computer program instructions stored in the memory 520 to implement any of the automatic update methods for operation and maintenance knowledge in the above embodiments.

[0126] The processor 510 reads and executes computer program instructions stored in the memory 520 to implement any of the automatic update methods for operation and maintenance knowledge in the above embodiments.

[0127] In one example, electronic device 500 may also include communication interface 530 and bus 540. Wherein, as... Figure 5 As shown, the processor 510, memory 520, and communication interface 530 are connected through bus 540 and complete communication with each other.

[0128] The communication interface 530 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0129] Bus 540 includes hardware, software, or both, that couples components of an electronic device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 540 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0130] For example, the electronic device 500 can be a mobile phone, tablet computer, laptop computer, handheld computer, in-vehicle electronic device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc.

[0131] The electronic device can execute the automatic update method of operation and maintenance knowledge in the embodiments of this application, thereby achieving the combination of Figures 1 to 3 The method for automatically updating operational and maintenance knowledge described herein, and the beneficial effects of the corresponding method implementations, will not be elaborated further here.

[0132] Furthermore, in conjunction with the automatic update method for operation and maintenance knowledge in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the automatic update methods for operation and maintenance knowledge in the above embodiments.

[0133] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0134] The computer program instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the automatic update method of operation and maintenance knowledge as shown in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0135] Based on the automatic update method for operation and maintenance knowledge in the above embodiments, this application embodiment can provide a computer program product to implement it. When the instructions in this computer program product are executed by the processor of an electronic device, they implement any of the automatic update methods for operation and maintenance knowledge in the above embodiments.

[0136] The computer program products of the above embodiments are used to implement the automatic updating method of operation and maintenance knowledge as shown in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0137] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0138] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0139] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0140] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0141] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for automatically updating operation and maintenance knowledge, characterized in that, The method includes: A vector knowledge base is pre-built to store operation and maintenance knowledge. Each piece of operation and maintenance knowledge has a corresponding knowledge tag, which includes source document information, knowledge location information within the document, and operation and maintenance entity information. In response to a correction event of the target operation and maintenance knowledge, obtain the correction information of the target operation and maintenance knowledge; The target operation and maintenance knowledge and the correction information are input into the large model, so that the large model generates associated knowledge related to the operation and maintenance knowledge and correction information of the associated knowledge; In response to receiving the approval information for the correction information, determine the knowledge tags corresponding to the target operation and maintenance knowledge and the associated knowledge; Based on the target operation and maintenance knowledge, the knowledge tags corresponding to the associated knowledge, and the correction information, the target operation and maintenance knowledge and the associated knowledge are updated in the vector knowledge base.

2. The method according to claim 1, characterized in that, Based on the target operation and maintenance knowledge, the knowledge tags corresponding to the associated knowledge, and the correction information, updating the target operation and maintenance knowledge and the associated knowledge in the vector knowledge base includes: Based on the knowledge tags corresponding to the target operation and maintenance knowledge and the associated knowledge, determine the knowledge that already exists in the vector knowledge base and needs to be updated and / or the knowledge that does not exist in the vector knowledge base and needs to be added. For each piece of knowledge to be updated, the knowledge in the vector knowledge base is updated based on the correction information of the knowledge to be updated; For each piece of knowledge to be added, based on the correction information of the knowledge to be added, the knowledge to be added and the corresponding knowledge tag are added to the vector knowledge base.

3. The method according to claim 1, characterized in that, The method further includes: In response to the operation and maintenance document retrieval operation, determine the document information of the operation and maintenance document to be retrieved; The document information is matched with the knowledge tags in the vector knowledge base to find multiple operation and maintenance knowledge related to the operation and maintenance document to be obtained; Based on the multiple operational and maintenance knowledge points found, the operational and maintenance documents to be obtained are reconstructed.

4. The method according to claim 2, characterized in that, Based on the knowledge tags corresponding to the target operation and maintenance knowledge and the associated knowledge, and the correction information, after updating the target operation and maintenance knowledge and the associated knowledge in the vector knowledge base, the method further includes: In the event repository, add a document correction event, which includes the source document information of the updated knowledge, the review object of the updated knowledge, the knowledge update time, and the updated version; After reconstructing the maintenance document to be obtained, the method further includes: Based on the document correction event, revision information is generated in the maintenance document to be obtained.

5. The method according to claim 1, characterized in that, The target operation and maintenance knowledge and the correction information are input into the large model, causing the large model to generate associated knowledge related to the operation and maintenance knowledge and correction information for the associated knowledge, including: Get the related query prompt template; Based on the target operation and maintenance knowledge and the associated query prompt template, generate associated query prompt information; The associated query prompt information and the correction information of the target operation and maintenance knowledge are input into the large model, so that the large model can find the associated operation and maintenance document in the database associated with the operation and maintenance entity of the target operation and maintenance knowledge, and find the associated knowledge associated with the target operation and maintenance knowledge in the operation and maintenance knowledge corresponding to the associated operation and maintenance document included in the vector knowledge base, and generate the correction information of the associated knowledge.

6. The method according to claim 1, characterized in that, Before obtaining the correction information of the target operation and maintenance knowledge in response to a correction event of the target operation and maintenance knowledge, the method further includes: In response to the operation and maintenance knowledge retrieval operation, determine the operation and maintenance entity to be retrieved; In the vector knowledge base, at least one piece of operational knowledge that matches the operational entity to be retrieved is identified; The search results are determined from at least one of the aforementioned operational knowledge.

7. The method according to claim 1, characterized in that, A vector knowledge base is pre-built, including: Multiple operation and maintenance documents of the application system are collected through multiple data collection interfaces; Each of the operation and maintenance documents is broken down to obtain the operation and maintenance knowledge of each of the operation and maintenance documents and the knowledge tags of each of the operation and maintenance knowledge. The various operational and maintenance knowledge items and their corresponding knowledge tags are stored in a vector knowledge base.

8. The method according to claim 1, characterized in that, Before obtaining the correction information of the target operation and maintenance knowledge in response to a correction event of the target operation and maintenance knowledge, the method further includes: In response to any feedback on the correction of operational knowledge, the number of feedback on the correction of any operational knowledge is accumulated; When the number of correction feedbacks for any of the operation and maintenance knowledge exceeds a preset threshold, the any of the operation and maintenance knowledge is taken as the target operation and maintenance knowledge, and a correction event for the target operation and maintenance knowledge is generated.

9. An automatic update device for operation and maintenance knowledge, characterized in that, include: Processor and memory storing computer program instructions; When the processor executes computer program instructions, it implements the automatic update method for operation and maintenance knowledge as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the automatic update method for operation and maintenance knowledge as described in any one of claims 1-8.

11. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the automatic update method for operation and maintenance knowledge as described in any one of claims 1-8.

Images