Operation and maintenance knowledge graph acquisition and application method, device, equipment and storage medium

Abstract

This disclosure provides a method, apparatus, device, and storage medium for acquiring and applying an operations and maintenance (O&M) knowledge graph, which can be applied to the field of O&M service technology. The method includes: acquiring O&M resource data from a device information resource database; performing structured identification on the O&M resource data to obtain static O&M knowledge, which includes multiple first entities; acquiring real-time O&M service work order data; using a pre-trained knowledge extraction model to perform unstructured identification on the O&M service work order data to obtain dynamic O&M knowledge, which includes multiple second entities; acquiring historical relationships between the multiple first entities and the multiple second entities; inferring new relationships corresponding to the first entities in the O&M service work order data based on the first similarity between the multiple first entities, the second similarity between the multiple second entities, and the historical relationships; and obtaining an O&M knowledge graph based on the new relationships.

Description

Technical Field

[0001] This disclosure relates to the field of operation and maintenance services, and in particular to a method, apparatus, equipment, medium, and program product for acquiring and applying operation and maintenance knowledge graphs. Background Technology

[0002] Operation and maintenance service tickets record various operation and maintenance tasks (including but not limited to changes, troubleshooting, and inspections) during the operation and maintenance process, accumulating a wealth of valuable operation and maintenance service experience. If this operation and maintenance service experience can be automatically extracted from operation and maintenance service tickets and automatically updated into the operation and maintenance knowledge graph, then the rapid replication and low-cost reuse of operation and maintenance service experience will become possible.

[0003] The large number of maintenance service work orders, the interrelationships between them, and the inconsistent personalities in filling out text-based fields (such as "fault description," "business impact," "cause analysis," and "handling plan") all contribute to the problem. Currently, manual knowledge extraction is labor-intensive and time-consuming, making it difficult to meet the need for automated, scheduled updates to the maintenance knowledge graph. Furthermore, the varying understanding of text-based fields makes it difficult to guarantee the quality of extracted maintenance knowledge. Alternatively, relying entirely on semantic analysis and machine learning places high demands on computing resources and incurs high training costs. Summary of the Invention

[0004] In view of the above problems, this disclosure provides a method, apparatus, equipment, medium and program product for acquiring and applying operation and maintenance knowledge graphs.

[0005] According to the first aspect of this disclosure, a method for obtaining an operation and maintenance knowledge graph is provided, comprising: obtaining operation and maintenance resource data from an equipment information resource database; performing structured identification on the operation and maintenance resource data to obtain static operation and maintenance knowledge, wherein the static operation and maintenance knowledge includes multiple first entities; obtaining real-time operation and maintenance service work order data; using a pre-trained knowledge extraction model to perform unstructured identification on the operation and maintenance service work order data to obtain dynamic operation and maintenance knowledge, wherein the dynamic operation and maintenance knowledge includes multiple second entities; obtaining historical association relationships between the multiple first entities and the multiple second entities; inferring new association relationships corresponding to the first entities in the operation and maintenance service work order data based on the first similarity between the multiple first entities, the second similarity between the multiple second entities, and the historical association relationships; and obtaining an operation and maintenance knowledge graph based on the new association relationships.

[0006] According to embodiments of this disclosure, the structured identification of operation and maintenance resource data to obtain static operation and maintenance knowledge includes: periodically pulling operation and maintenance resource data from the equipment information resource database using a preset tool; performing entity identification, entity relationship identification, and entity attribute identification on the operation and maintenance resource data to obtain relational data composed of entities, entity relationships, and entity attributes; and converting the relational data into RDF triples to obtain static operation and maintenance knowledge.

[0007] According to embodiments of this disclosure, the method for obtaining an operation and maintenance knowledge graph further includes: constructing an operation and maintenance domain corpus based on static operation and maintenance knowledge and operation and maintenance service work order data; calculating the third similarity between the operation and maintenance service work order data and the words in the operation and maintenance domain corpus; filtering out entity pairs corresponding to words with a third similarity greater than a preset value; and using sentences containing entity pairs as training samples to train a knowledge extraction model.

[0008] According to embodiments of this disclosure, constructing an operations and maintenance domain corpus based on static operations and maintenance knowledge and operations and maintenance service work order data includes: constructing an initial operations and maintenance domain corpus; outputting relational data into the initial operations and maintenance domain corpus; segmenting the operations and maintenance service work order data, removing stop words, and calculating the weighted weights of the remaining words; and outputting words that meet the preset weights into the initial operations and maintenance domain corpus to obtain the operations and maintenance domain corpus.

[0009] According to embodiments of this disclosure, obtaining historical associations between multiple first entities and multiple second entities includes: obtaining historical operation and maintenance service work orders, which include first change request work orders and event management work orders; establishing multiple sub-work orders of the first change request work order; obtaining first work order associations between the first change request work order and the multiple sub-work orders, and first work order resource associations between the first change request work order and the multiple first entities and multiple second entities contained therein; obtaining historical associations based on the first work order associations and the first work order resource associations; and / or, establishing a second change request work order corresponding to the event management work order; obtaining second work order associations between the event management work order and the second change request work order, and second work order resource associations between the event management work order and the multiple first entities and multiple second entities contained therein; obtaining historical associations based on the second work order associations and the second work order resource associations.

[0010] The second aspect of this disclosure provides a method for applying an operations and maintenance (O&M) knowledge graph. The O&M knowledge graph obtained using an O&M knowledge graph acquisition method based on any embodiment of this application includes: acquiring an O&M service work order to be processed; retrieving O&M knowledge associated with the O&M service work order from the O&M knowledge graph and calculating a fourth similarity; sorting the O&M knowledge by weight based on business relevance and the fourth similarity; and processing the O&M service work order to be processed according to the O&M knowledge with the highest weight.

[0011] According to embodiments of this disclosure, the method for applying an operations and maintenance knowledge graph further includes: updating the weight of the corresponding operations and maintenance knowledge in the operations and maintenance knowledge graph based on the operations and maintenance knowledge with the highest weight.

[0012] A third aspect of this disclosure provides an operation and maintenance knowledge graph acquisition device, comprising: a first acquisition module for acquiring operation and maintenance resource data from a device information resource database; a first identification module for performing structured identification on the operation and maintenance resource data to obtain static operation and maintenance knowledge, wherein the static operation and maintenance knowledge includes multiple first entities; a second acquisition module for acquiring real-time operation and maintenance service work order data; a second identification module for performing unstructured identification on the operation and maintenance service work order data using a pre-trained knowledge extraction model to obtain dynamic operation and maintenance knowledge, wherein the dynamic operation and maintenance knowledge includes multiple second entities; a third acquisition module for acquiring historical association relationships between the multiple first entities and the multiple second entities; a reasoning module for reasoning new association relationships corresponding to the first entities in the operation and maintenance service work order data based on the first similarity between the multiple first entities, the second similarity between the multiple second entities, and the historical association relationships; and a graph acquisition module for obtaining an operation and maintenance knowledge graph based on the new association relationships.

[0013] A fourth aspect of this disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors perform the above-described operation and maintenance knowledge graph acquisition method.

[0014] The fifth aspect of this disclosure also provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the above-described operation and maintenance knowledge graph acquisition method.

[0015] The sixth aspect of this disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for acquiring operational knowledge graphs.

[0016] According to the operation and maintenance knowledge graph acquisition method, device, equipment, medium and program products provided in this disclosure, static operation and maintenance knowledge is extracted from the IT equipment resource library and dynamic operation and maintenance knowledge is extracted from operation and maintenance service work orders. Then, the entity similarity in the static operation and maintenance knowledge and dynamic operation and maintenance knowledge are calculated respectively. Based on the historical association data between static entities and dynamic entities in historical service work orders, dynamic operation and maintenance knowledge and static operation and maintenance knowledge can be merged and inferred to obtain new association relationships between static entities in real-time work orders and historical dynamic entities, thereby obtaining an operation and maintenance knowledge graph containing the new association relationship and related entities.

[0017] Since static entities change at a low frequency while dynamic entities change at a high frequency, extracting static and dynamic operation and maintenance knowledge separately allows for the use of simple extraction methods to extract static operation and maintenance knowledge, greatly saving system resources and reducing knowledge extraction costs. Furthermore, by fusing and reasoning through dynamic and static knowledge, the knowledge extraction model gains knowledge expansion capabilities, enabling the operation and maintenance knowledge graph to automatically handle issues related to similar machines (i.e., static entities). Therefore, it at least partially solves the problems of high workload, long time consumption, and low quality in manual extraction of operation and maintenance knowledge, achieving the technical effect of low-cost and automated acquisition of operation and maintenance knowledge graphs. Attached Figure Description

[0018] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0019] Figure 1 The illustration schematically depicts application scenarios of the operation and maintenance knowledge graph acquisition method, apparatus, device, medium, and program product according to embodiments of this disclosure;

[0020] Figure 2 A flowchart illustrating a method for obtaining an operation and maintenance knowledge graph according to an embodiment of this disclosure is shown schematically.

[0021] Figure 3 A flowchart illustrating a method for extracting static operation and maintenance knowledge according to an embodiment of the present disclosure is shown schematically.

[0022] Figure 4 A schematic diagram illustrating the composition of static operation and maintenance knowledge according to an embodiment of this disclosure is shown.

[0023] Figure 5 A flowchart illustrating a method for training a knowledge extraction model according to an embodiment of the present disclosure is shown.

[0024] Figure 6 A flowchart illustrating a method for constructing an operations and maintenance domain corpus according to an embodiment of this disclosure is shown schematically.

[0025] Figure 7 A flowchart illustrating a method for obtaining entity association relationships according to an embodiment of this disclosure is shown schematically.

[0026] Figure 8 A flowchart illustrating an operation and maintenance knowledge graph application method according to an embodiment of the present disclosure is shown schematically.

[0027] Figure 9 This schematically illustrates a structural block diagram of an operation and maintenance knowledge graph acquisition device according to an embodiment of the present disclosure; and

[0028] Figure 10A block diagram schematically illustrates an electronic device suitable for implementing an operation and maintenance knowledge graph acquisition method according to an embodiment of the present disclosure. Detailed Implementation

[0029] The embodiments of the present disclosure will now 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 disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0030] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0031] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0032] When using expressions such as "at least one of A, B, and C", the expression should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (e.g., "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B, and C, etc.).

[0033] In the technical solution disclosed herein, the collection, storage, use, processing, transmission, provision, disclosure, and application of user personal information comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and there is no violation of public order and good morals.

[0034] In the technical solution disclosed herein, the acquisition, collection, storage, use, processing, transmission, provision, disclosure, and application of data all comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and they do not violate public order and good morals.

[0035] CMDB: A resource database that stores information about IT equipment, including information such as IP addresses, ports, and topology relationships of resources such as physical subsystems, deployment units, servers, operating systems, and middleware.

[0036] ITSM: Operation and Maintenance Service Tickets. In daily operations and maintenance work, work tickets are usually used to record, review, and follow up on operations and maintenance services, including daily inspections, troubleshooting, upgrades, and changes. A wealth of valuable operations and maintenance experience has been accumulated in these work tickets.

[0037] Operations and Maintenance Knowledge Graph: A knowledge graph in the IT operations and maintenance field typically stores IT entities (servers, hosts, databases, racks, service modules, service processes, etc.) and the pairwise relationships between entities using triples. A well-designed operations and maintenance knowledge graph can make it possible to quickly and cost-effectively reuse operations and maintenance service experience.

[0038] Figure 1 The diagram illustrates an application scenario of the operation and maintenance knowledge graph acquisition method according to an embodiment of the present disclosure.

[0039] like Figure 1 As shown, application scenario 100 according to this embodiment may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as a medium for providing a communication link between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0040] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0041] Terminal devices 101, 102, and 103 can be various electronic devices with displays and web browsing capabilities, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0042] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using terminal devices 101, 102, and 103 (this is just an example). The backend management server, such as an operations and maintenance management server, can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated based on user requests) to the terminal devices.

[0043] It should be noted that the operation and maintenance knowledge graph acquisition method provided in this embodiment can generally be executed by server 105. Correspondingly, the operation and maintenance knowledge graph acquisition device provided in this embodiment can generally be located in server 105. The operation and maintenance knowledge graph acquisition method provided in this embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with terminal devices 101, 102, 103 and / or server 105. Correspondingly, the operation and maintenance knowledge graph acquisition device provided in this embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with terminal devices 101, 102, 103 and / or server 105.

[0044] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0045] The following will be based on Figure 1 The described scene, through Figures 2-7 The method for obtaining the operation and maintenance knowledge graph according to the disclosed embodiments is described in detail.

[0046] Figure 2 A flowchart illustrating a method for obtaining an operation and maintenance knowledge graph according to an embodiment of this disclosure is shown schematically.

[0047] like Figure 2 As shown, the operation and maintenance knowledge graph acquisition method in this embodiment includes operations S210 to S270.

[0048] When operating S210, retrieve operation and maintenance resource data from the equipment information resource database.

[0049] According to embodiments of this disclosure, the device information resource database is, for example, a CMDB (a database for storing IT device information). For example, maintenance resource data in the CMDB can be periodically retrieved via Pull ETL or by calling an open API.

[0050] Specifically, the Pull ETL program, for example, calls the CMDB's openAPI interface via Python to pull and parse operation and maintenance resource data according to ETL rules. Compared to service ticket data, the operation and maintenance resources in the device information resource library, such as physical subsystems, deployment units, hosts, storage, databases, and middleware, change less frequently, so periodic, scheduled retrieval is sufficient.

[0051] During the operation of S220, structured identification is performed on the operation and maintenance resource data to obtain static operation and maintenance knowledge, which includes multiple first entities.

[0052] According to embodiments of this disclosure, since the change rate of operation and maintenance resource data is relatively low compared to service ticket data, this part of the operation and maintenance knowledge belongs to static knowledge. Because this data is structured, a D2RQ Mapping file can be written, and using D2R tools, the operation and maintenance resource data can be abstracted into static operation and maintenance knowledge according to entities, attributes, and relationships. Simultaneously, it can be saved in RDF format, in the form of triples, into a knowledge base. Multiple first entities are multiple entities representing hardware information such as devices extracted from the operation and maintenance resource data.

[0053] When operating S230, obtain real-time operation and maintenance service work order data.

[0054] According to embodiments of this disclosure, for example, ITSM (Installation and Maintenance Service Request) push data can be obtained in real time by Push ETL or by calling the open API.

[0055] Specifically, the Push ETL program can be used to push maintenance service ticket data from ITSM to a message queue, and the message queue can be subscribed to to receive maintenance service ticket data in real time. Since maintenance service tickets carry event management tickets such as fault handling and emergency handling in the production environment maintenance process, as well as maintenance service tickets such as data backup and disk expansion, the relationships between tickets and the content of tickets will change as the maintenance service process progresses, so it is necessary to obtain them in real time.

[0056] When operating S240, a pre-trained knowledge extraction model is used to perform unstructured identification on the operation and maintenance service work order data to obtain dynamic operation and maintenance knowledge, which includes multiple second entities.

[0057] According to embodiments of this disclosure, since the operation and maintenance service ticket data is semi-structured data, such as fault descriptions, causes, and solutions in event management tickets, and unstructured text information such as change content, reasons for change, change risks, and change control tables in change service tickets, knowledge extraction can be performed on this part of the text, for example, using remote monitoring methods. Similarly, the extracted dynamic operation and maintenance knowledge also contains multiple second entities, such as fault descriptions, causes, solutions, follow-up measures, and emergency response scripts.

[0058] In operation S250, the historical relationships between multiple first entities and multiple second entities are obtained.

[0059] According to embodiments of this disclosure, the maintenance service ticket includes entities from static maintenance knowledge such as equipment and other hardware types, as well as entities from dynamic maintenance knowledge such as fault descriptions and solutions. Historical relationships can be, for example, the correspondence between a known machine entity in a historical maintenance service ticket and the related problem solution it resolved.

[0060] In operation S260, a new association relationship corresponding to the first entity in the operation and maintenance service work order data is inferred based on the first similarity between multiple first entities, the second similarity between multiple second entities, and the historical association relationship.

[0061] According to embodiments of this disclosure, for example, the fusion of static and dynamic operation and maintenance knowledge can be performed based on the unique identifier of the operation and maintenance entity (e.g., physical address, work order ID, etc.). Static operation and maintenance knowledge may include, for example, static operation and maintenance entities for multiple devices, while dynamic operation and maintenance knowledge may include, for example, dynamic operation and maintenance entities describing multiple faults. From known operation and maintenance service work orders, a fault C occurring on device A and its corresponding fault handling solution can be obtained; that is, there is an association between the device, the fault, and the handling solution. If another device B and a fault D occurring on this device are extracted from the acquired real-time operation and maintenance service work order, then the similarity between device B and device A, and the similarity between fault D and fault C, can be calculated. Assuming that the similarity calculation shows device B is similar to device A, and fault D is similar to fault C, then based on the association between device A and fault C, a new equivalent association between device B and fault C can be inferred, thereby obtaining an operation and maintenance knowledge graph containing this new association and related entities. The handling solution for fault D can then be automatically generated using this operation and maintenance knowledge graph.

[0062] It is understandable that the device A, device B, fault C, and fault D mentioned above are merely illustrative designations, and the correspondence between devices and faults is not clear before the similarity is calculated. There are multiple historical maintenance service tickets, containing multiple static and dynamic maintenance entities. By calculating the similarity between each entity in the real-time maintenance service ticket and each entity in the historical maintenance service tickets, we can find device A, which is similar to device B, and fault C, which is similar to fault D. Dynamic maintenance entities can include not only faults but also changes, such as replacing server disks or relocating data center addresses.

[0063] Specifically, for example, there is a relationship R1 between Linux server ED1 and entity ST1 in the implemented expansion service work order. The head entity is static operation and maintenance entity ED1, and the tail entity is dynamic operation and maintenance entity ST1. Entity ST2 in the unimplemented expansion service work order is implemented on another Linux server ED2. Based on the high similarity between ST2 and ST1, and the high similarity between ED2 and ED1, it can be inferred that there is a similar relationship between static operation and maintenance entity ED2 and dynamic operation and maintenance entity ST1. That is, the change risk, change control table, change script, and verification script in ST1 are applicable to the expansion service work order of ED2.

[0064] When operating S270, an operations and maintenance knowledge graph is obtained based on the new relationships.

[0065] According to embodiments of this disclosure, after obtaining the operation and maintenance knowledge graph, the operation and maintenance knowledge graph obtained by fusing and reasoning static and dynamic operation and maintenance entities can be updated to the graph database and represented in a machine-processable mode to facilitate subsequent applications of the operation and maintenance knowledge graph. This disclosure, based on semi-structured data in JSON format obtained from the operation and maintenance service work order interface, reduces the cost of acquiring operation and maintenance knowledge while achieving automatic extraction of operation and maintenance knowledge and automatic updating of the operation and maintenance knowledge graph.

[0066] Figure 3 A flowchart illustrating a method for extracting static operation and maintenance knowledge according to an embodiment of the present disclosure is shown.

[0067] According to embodiments of this disclosure, such as Figure 3 As shown, static operation and maintenance knowledge is extracted through steps S221 to S223.

[0068] Step S221: Use a preset tool to periodically retrieve operation and maintenance resource data from the equipment information resource database.

[0069] According to embodiments of this disclosure, the preset tool is, for example, the Pull ETL program. Pull ETL periodically and actively pulls operation and maintenance resource data from the CMDB. Before pulling the data, for example, a Python editor such as PyCharm (Python version 3.8.8) can be installed, along with modules such as requests, json, jieba, and word2vec. A network environment that can call the operation and maintenance service ticket query interface is also required. Specifically, the requests module is used to call the openAPI interface to obtain operation and maintenance ticket data; the json module mainly formats the obtained JSON data; the jieba module is mainly used for text segmentation and stop word removal; and the Word2Vec module performs named entity recognition and relation extraction.

[0070] Step S222: Perform entity identification, entity relationship identification, and entity attribute identification on the operation and maintenance resource data to obtain relational data composed of entities, entity relationships, and entity attributes.

[0071] According to embodiments of this disclosure, by identifying structured operation and maintenance resource data, static knowledge entities (such as physical subsystems, deployment units, storage, databases, processes, middleware, etc.), entity attributes (such as IP addresses, ports, etc.), and entity relationships (such as inclusion, composition, connection, etc.) can be obtained. These interconnected static knowledge entities, entity attributes, and entity relationships together constitute relational data.

[0072] Step S223: Convert relational data into RDF triples to obtain static operation and maintenance knowledge.

[0073] Figure 4 A schematic diagram illustrating the composition of static operation and maintenance knowledge according to an embodiment of this disclosure is shown.

[0074] According to embodiments of this disclosure, such as Figure 4 As shown, by converting the aforementioned relational data into RDF triples, abstract static operational knowledge can be obtained. For example, physical subsystem 1 contains n deployment units, each containing, for example, two Linux servers connected to an NFS shared storage. For instance, static operational knowledge related to the machine itself, such as machine expansion, has a low update frequency. Separating and extracting low-frequency static operational knowledge from business-related dynamic operational knowledge, and using simple extraction methods such as structured recognition and format conversion, significantly reduces the workload of high-frequency operational management tasks such as business changes and fault handling, thereby lowering operational management costs.

[0075] Figure 5 A flowchart illustrating a method for training a knowledge extraction model according to an embodiment of the present disclosure is shown.

[0076] According to embodiments of this disclosure, such as Figure 5 As shown, for example, the knowledge extraction model is trained through steps S241 to S244.

[0077] Step S241: Construct an operations and maintenance domain corpus based on static operations and maintenance knowledge and operations and maintenance service work order data.

[0078] According to embodiments of this disclosure, the operations and maintenance (O&M) domain corpus is a collection of high-quality static O&M knowledge and dynamic O&M knowledge with high similarity. By constructing an O&M domain corpus and training a dynamic O&M knowledge extraction model based on it, the accuracy of dynamic O&M knowledge extraction is improved. For example, O&M service work orders are manually processed work orders that are continuously accumulated. Because a high-quality O&M domain corpus has been constructed, only a portion of these work orders can be extracted as training samples. This improves training efficiency, reduces training costs, and, based on their semi-structured data type and combined with the high-quality O&M domain corpus, also ensures the accuracy of dynamic O&M knowledge extraction.

[0079] Step S242: Calculate the third similarity between the operation and maintenance service work order data and the words in the operation and maintenance domain corpus.

[0080] Step S243: Filter out entity pairs corresponding to words with a third similarity greater than a preset value.

[0081] According to embodiments of this disclosure, entity pairs are obtained by filtering out words in the operation and maintenance service work order data that are similar to words in the operation and maintenance domain corpus. The higher the similarity, the closer the entities are. If the cosine similarity of word vectors is calculated, the range is [0, 1], and generally, a similarity greater than 0.8 is considered similar. If high quality is required, the threshold can be increased to 0.85. Entity pairs are, for example, resource entity pairs, such as disk and server in "disk mounted to server", and data center and address in "data center migrated from address A to address B".

[0082] Step S244: Use sentences containing entity pairs as training samples to train the knowledge extraction model.

[0083] According to embodiments of this disclosure, the knowledge extraction model is, for example, a supervised learning model. After word segmentation, sentences containing the aforementioned entity pairs are extracted from unstructured text in the operation and maintenance service work order data, for example, based on a constructed high-quality operation and maintenance domain corpus, as training samples. The supervised learning model is then trained to perform dynamic entity recognition, attribute recognition, and relation recognition. The extracted dynamic operation and maintenance entities, entity attributes, and entity relations are then added back to the operation and maintenance domain corpus as the next batch of training sample data, thus iteratively optimizing the supervised learning model. Both the operation and maintenance domain corpus and the knowledge extraction model are continuously and dynamically optimized to further ensure the accuracy of dynamic knowledge extraction.

[0084] Figure 6 A flowchart illustrating a method for constructing an operations and maintenance domain corpus according to an embodiment of this disclosure is shown.

[0085] According to embodiments of this disclosure, such as Figure 6 As shown, for example, the knowledge extraction model is trained through steps S2411 to S2414.

[0086] Step S2411: Construct an initial corpus of operational and maintenance domains.

[0087] Step S2412: Output the relational data to the initial operation and maintenance domain corpus.

[0088] According to embodiments of this disclosure, relational data includes, for example, the aforementioned static knowledge entities (e.g., physical subsystems, deployment units, storage, databases, processes, middleware, etc.), entity attributes (e.g., IP addresses, ports, etc.), and entity relationships (e.g., inclusion, composition, connection, etc.).

[0089] Step S2413: Segment the operation and maintenance service work order data, remove stop words, and calculate the weighted weight of the remaining words.

[0090] According to embodiments of this disclosure, for example, the weighted weights of the remaining words in the maintenance service work order data after removing stop words are calculated using the TF-IDF algorithm.

[0091] Step S2414: Output words that meet the preset weights to the initial operation and maintenance domain corpus to obtain the operation and maintenance domain corpus.

[0092] According to embodiments of this disclosure, the theoretical threshold for TF-IDF is [0, positive infinity), with larger values ​​indicating greater importance. Therefore, the top N terms are typically added to the vocabulary. In this scenario, for example, the top 5 terms can be added to the operations and maintenance (O&M) domain corpus. Word mining in O&M service tickets is, for example, based on the O&M domain corpus, and the mined words are then input into the O&M domain corpus. This means the vocabulary input from O&M service tickets to the O&M domain corpus is a dynamic, cyclical process. Combining this with static O&M knowledge as the vocabulary input source ensures the breadth of the O&M domain corpus, and the selection of word weights also improves the accuracy of the vocabulary.

[0093] Figure 7 A flowchart illustrating a method for obtaining entity association relationships according to an embodiment of this disclosure is shown.

[0094] According to embodiments of this disclosure, such as Figure 7 As shown, entity association relationships are obtained, for example, through steps S251 to S257.

[0095] Step S251: Obtain historical operation and maintenance service work orders, which include the first change request work order and the event management work order.

[0096] According to embodiments of this disclosure, before fusing and reasoning about static and dynamic operation and maintenance knowledge, it is necessary to first obtain the relationships between work orders and between work orders and the various entities contained within them. Then, by establishing work order relationships between static operation and maintenance entities such as physical subsystems, deployment units, and hosts, and dynamic operation and maintenance entities such as change types, change content, change risks, change control tables, fault types, and solutions, knowledge reasoning is performed based on the static operation and maintenance entity types and work order relationships. Simultaneously, relation vectors are used to transform the entity vectors in the static knowledge graph (obtained by vectorizing static operation and maintenance knowledge). Each triple is defined based on the similarity between the transformed head and tail entities, thereby obtaining the required operation and maintenance knowledge graph.

[0097] Step S252: Create multiple sub-work orders for the first change request work order.

[0098] According to embodiments of this disclosure, the first change request work order is, for example, an independent change request work order resulting from non-fault-related changes such as business changes. For a change request work order, if a change involves several implementation stages or steps, corresponding sub-work orders need to be created to track the implementation progress in detail. For example, a machine resource deployment application will create corresponding machine power-on service work orders, operating system installation work orders, network debugging work orders, and service component deployment work orders according to the implementation stage. These change work orders are related to the resource entity, and there are also correlations between the work orders themselves.

[0099] Step S253: Obtain the first work order association relationship between the first change application work order and multiple sub-work orders, and the first work order resource association relationship between the first change application work order and multiple first entities and multiple second entities contained therein.

[0100] According to embodiments of this disclosure, the first work order association relationship is, for example, the association relationship between a general change request work order and each sub-work order. The first work order resource association relationship is, for example, the association relationship between the general change request work order and each static operation and maintenance entity and dynamic operation and maintenance entity contained in the general change request work order.

[0101] Step S254: Based on the first work order association and the first work order resource association, obtain the historical association. And / or,

[0102] Step S255: Create a second change request work order corresponding to the event management work order.

[0103] According to embodiments of this disclosure, for event management, in addition to filling in information such as fault description and severity level in the event management work order, there will also be a corresponding change request work order, i.e., a second change request work order. For example, for an Oracle fault caused by insufficient tablespace (corresponding to event management work order ST3), after completing the event analysis, it is necessary to add the corresponding tablespace, so there will be a "tablespace expansion" change request work order (corresponding to change request work order ST4). Therefore, there is a correlation between event management work order ST3 and change request work order ST4.

[0104] Step S256: Obtain the second work order association relationship between the event management work order and the second change application work order, as well as the second work order resource association relationship between the event management work order and the multiple first entities and multiple second entities contained therein.

[0105] According to embodiments of this disclosure, the second work order resource association relationship is, for example, the association relationship between the event management work order and the various static operation and maintenance entities and dynamic operation and maintenance entities contained in the event management work order.

[0106] Step S257: Based on the second work order association relationship and the second work order resource association relationship, obtain the historical association relationship.

[0107] According to embodiments of this disclosure, historical relationships include relationships between work orders and relationships between work orders and entities within work orders. By examining the work order relationships and entity relationships in the change request work order data, the knowledge graph related to change request processing can be improved. Similarly, by examining the work order relationships and entity relationships in the event management work orders, the knowledge graph related to event attribution and fault handling can be improved.

[0108] According to embodiments of this disclosure, the association between real-time acquired operation and maintenance service work orders and static and dynamic operation and maintenance entities in historical operation and maintenance service work orders can be realized based on the above-mentioned historical association relationships.

[0109] The change request work order may include, but is not limited to, the following information, as shown in Table 1:

[0110] Table 1 Change Request Work Order

[0111]

[0112]

[0113] This allows for the establishment of connections between dynamic operational entities such as change content, change reasons, change plans, change schedules, and change control tables, and static operational entities such as physical subsystems, deployment units, and hosts. Event management work orders may include, but are not limited to, the following information, as shown in Table 2:

[0114] Table 2 Event Management Work Orders

[0115]

[0116] This allows for the establishment of connections between dynamic O&M entities such as fault descriptions, causes, solutions, follow-up measures, and emergency response scripts, and static O&M entities such as physical subsystems, deployment units, and hosts. By extracting fault entities and causal relationships, a fault attribution chain is established, facilitating the later application of the O&M knowledge graph for fault prediction. Furthermore, establishing connections between static O&M entities and dynamic O&M entities such as fault descriptions, solutions, and emergency response scripts facilitates the later application of the O&M knowledge graph for fault emergency response.

[0117] Figure 8 A flowchart illustrating an operation and maintenance knowledge graph application method according to an embodiment of this disclosure is shown.

[0118] Based on the operation and maintenance knowledge graph obtained by the above-mentioned method, this disclosure also provides an operation and maintenance knowledge graph application method, including, for example:

[0119] S810, retrieve pending maintenance service work orders.

[0120] According to embodiments of this disclosure, for example, dynamic operation and maintenance entities such as operation and maintenance service work order type and work order title, as well as static operation and maintenance entities such as physical subsystems, deployment units, and hosts in the operation and maintenance service work order, can be obtained.

[0121] S820: Retrieve operation and maintenance knowledge associated with the operation and maintenance service work order to be processed from the operation and maintenance knowledge graph, and calculate the fourth similarity.

[0122] According to embodiments of this disclosure, for example, based on dynamic and static operation and maintenance entities such as operation and maintenance service work order type and work order title, related operation and maintenance knowledge can be retrieved in the operation and maintenance knowledge graph.

[0123] S830 sorts operational knowledge by weight based on business relevance and fourth similarity.

[0124] According to embodiments of this disclosure, for example, the retrieved operation and maintenance knowledge is sorted in reverse order based on similarity weight, and the recommendation results are transmitted to the operation and maintenance service ticket system through an interface. The operation and maintenance engineer selects the most relevant handling process according to the business.

[0125] S840 processes pending maintenance service work orders based on the maintenance knowledge with the highest weight.

[0126] According to embodiments of this disclosure, after obtaining the highest-weighted operation and maintenance knowledge, the weights of the corresponding operation and maintenance knowledge in the operation and maintenance knowledge graph can be updated to further improve the accuracy of the operation and maintenance knowledge graph and the recommendation accuracy of the recommendation system that recommends the operation and maintenance knowledge graph. Applying the operation and maintenance knowledge graph to operation and maintenance services, such as to operation and maintenance tools including but not limited to intelligent event handling and intelligent handling of operation and maintenance service work orders, improves operation and maintenance efficiency and quality.

[0127] Specifically, once the accuracy rate reaches a certain level, low-risk operation and maintenance service orders can be implemented automatically. For example, upon receiving an operation and maintenance service order ST5 for server expansion, the system retrieves the server for which the change is being implemented in ST5, whose operating system attribute is Linux. Based on entity attributes and entity relationships, the system retrieves the change risk (low), change control table, change script, and verification script for operation and maintenance service order ST5 from the operation and maintenance knowledge graph. This allows for the automatic expansion of the server that initiated the request, while simultaneously returning the verification results to the console and providing feedback to the operation and maintenance engineer.

[0128] Based on the above-described method for obtaining an operations and maintenance knowledge graph, this disclosure also provides an apparatus for obtaining such a graph. The following will combine... Figure 9 The device is described in detail.

[0129] Figure 9 A schematic block diagram of an operation and maintenance knowledge graph acquisition device according to an embodiment of the present disclosure is shown.

[0130] like Figure 8 As shown, the operation and maintenance knowledge graph acquisition device 900 in this embodiment includes, for example, a first acquisition module 910, a first identification module 920, a second acquisition module 930, a second identification module 940, a third acquisition module 950, a reasoning module 960, and a graph acquisition module 970.

[0131] The first acquisition module 910 is used to acquire operation and maintenance resource data from the device information resource database. In one embodiment, the first acquisition module 910 can be used to perform the operation S210 described above, which will not be repeated here.

[0132] The first identification module 920 is used to perform structured identification on the operation and maintenance resource data to obtain static operation and maintenance knowledge, which includes multiple first entities. In one embodiment, the first identification module 920 can be used to perform the operation S220 described above, which will not be repeated here.

[0133] The second acquisition module 930 is used to acquire real-time operation and maintenance service work order data. In one embodiment, the second acquisition module 930 can be used to perform the operation S230 described above, which will not be repeated here.

[0134] The second identification module 940 is used to perform unstructured identification on the operation and maintenance service work order data using a pre-trained knowledge extraction model to obtain dynamic operation and maintenance knowledge, which includes multiple second entities. In one embodiment, the second identification module 940 can be used to perform the operation S240 described above, which will not be repeated here.

[0135] The third acquisition module 950 is used to acquire the historical association relationships between multiple first entities and multiple second entities. In one embodiment, the third acquisition module 950 can be used to perform the operation S250 described above, which will not be repeated here.

[0136] The reasoning module 960 is used to infer a new association relationship corresponding to the first entity in the operation and maintenance service work order data based on the first similarity between multiple first entities, the second similarity between multiple second entities, and historical association relationships. In one embodiment, the reasoning module 960 can be used to perform the operation S260 described above, which will not be repeated here.

[0137] The graph acquisition module 970 is used to obtain an operation and maintenance knowledge graph based on the new relationships. In one embodiment, the graph acquisition module 970 can be used to perform the operation S270 described above, which will not be repeated here.

[0138] According to embodiments of this disclosure, any plurality of modules among the first acquisition module 910, the first identification module 920, the second acquisition module 930, the second identification module 940, the third acquisition module 950, the inference module 960, and the map acquisition module 970 can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least a portion of the functionality of one or more of these modules can be combined with at least a portion of the functionality of other modules and implemented in one module. According to embodiments of this disclosure, at least one of the first acquisition module 910, the first identification module 920, the second acquisition module 930, the second identification module 940, the third acquisition module 950, the inference module 960, and the map acquisition module 970 can be at least partially implemented as hardware circuitry, 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-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the first acquisition module 910, the first identification module 920, the second acquisition module 930, the second identification module 940, the third acquisition module 950, the reasoning module 960, and the map acquisition module 970 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0139] Figure 10 A block diagram schematically illustrates an electronic device suitable for implementing an operation and maintenance knowledge graph acquisition method according to an embodiment of the present disclosure.

[0140] like Figure 10 As shown, an electronic device 1000 according to an embodiment of the present disclosure includes a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage portion 1008 into a random access memory (RAM) 1003. The processor 1001 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1001 may also include onboard memory for caching purposes. The processor 1001 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.

[0141] RAM 1003 stores various programs and data required for the operation of electronic device 1000. Processor 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. Processor 1001 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 1002 and / or RAM 1003. It should be noted that the programs may also be stored in one or more memories other than ROM 1002 and RAM 1003. Processor 1001 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0142] According to embodiments of this disclosure, electronic device 1000 may further include an input / output (I / O) interface 1005, which is also connected to bus 1004. Electronic device 900 may further include one or more of the following components connected to I / O interface 1005: an input section 1006 including a keyboard, mouse, etc.; an output section 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to I / O interface 1005 as needed. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 1010 as needed so that computer programs read from it can be installed into storage section 1008 as needed.

[0143] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the operation and maintenance knowledge graph acquisition method according to the embodiments of this disclosure.

[0144] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include ROM 1002 and / or RAM 1003 and / or one or more memories other than ROM 1002 and RAM 1003 described above.

[0145] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code enables the computer system to implement the operation and maintenance knowledge graph acquisition method provided in the embodiments of this disclosure.

[0146] When the computer program is executed by the processor 1001, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0147] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 1009, and / or installed from a removable medium 1011. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0148] In such an embodiment, the computer program can be downloaded and installed from a network via communication section 1009, and / or installed from removable medium 1011. When the computer program is executed by processor 1001, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0149] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can execute entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0150] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0151] Those skilled in the art will understand that the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0152] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A method for obtaining an operations and maintenance knowledge graph, comprising: Obtain operation and maintenance resource data from the equipment information resource database; The operation and maintenance resource data is subjected to structured identification to obtain static operation and maintenance knowledge, which includes multiple first entities; Obtain real-time operation and maintenance service work order data; A pre-trained knowledge extraction model is used to perform unstructured identification on the operation and maintenance service work order data to obtain dynamic operation and maintenance knowledge, which includes multiple second entities. Obtaining the historical association relationships between the plurality of first entities and the plurality of second entities includes: obtaining historical operation and maintenance service work orders, the historical operation and maintenance service work orders including first change request work orders and event management work orders; establishing a plurality of sub-work orders of the first change request work order; obtaining the first work order association relationship between the first change request work order and the plurality of sub-work orders, and the first work order resource association relationship between the first change request work order and the plurality of first entities and the plurality of second entities contained therein; obtaining the historical association relationships based on the first work order association relationships and the first work order resource association relationships; and / or, establishing a second change request work order corresponding to the event management work order; obtaining the second work order association relationship between the event management work order and the second change request work order, and the second work order resource association relationship between the event management work order and the plurality of first entities and the plurality of second entities contained therein; obtaining the historical association relationships based on the second work order association relationships and the second work order resource association relationships. Based on the first similarity between the plurality of first entities, the second similarity between the plurality of second entities, and the historical association, a new association relationship corresponding to the first entity in the operation and maintenance service work order data is inferred. Based on the new relationships, the operation and maintenance knowledge graph is obtained.

2. The method for obtaining an operation and maintenance knowledge graph according to claim 1, wherein the step of performing structured identification on the operation and maintenance resource data to obtain static operation and maintenance knowledge includes: The operation and maintenance resource data is periodically retrieved from the equipment information resource database using a preset tool; Entity identification, entity relationship identification, and entity attribute identification are performed on the operation and maintenance resource data to obtain relational data composed of entities, entity relationships, and entity attributes; The relational data is converted into RDF triples to obtain the static operation and maintenance knowledge.

3. The method for obtaining an operation and maintenance knowledge graph according to claim 2 further includes: A corpus of operational and maintenance terminology is constructed based on the static operational and maintenance knowledge and the operational and maintenance service work order data. Calculate the third similarity between the operation and maintenance service work order data and the words in the operation and maintenance domain corpus; Filter out entity pairs corresponding to words whose third similarity is greater than a preset value; The knowledge extraction model is trained by using sentences containing the entity pairs as training samples.

4. The method for obtaining an operation and maintenance knowledge graph according to claim 3, wherein constructing an operation and maintenance domain corpus based on the static operation and maintenance knowledge and the operation and maintenance service work order data includes: Build an initial corpus of operational and maintenance terminology; The relational data is output to the initial operation and maintenance domain corpus; as well as, The operation and maintenance service work order data is segmented into words, stop words are removed, and the weighted weight of the remaining words is calculated. Words that meet the preset weights are output to the initial operation and maintenance domain corpus to obtain the operation and maintenance domain corpus.

5. A method for applying an operations and maintenance knowledge graph, comprising using an operations and maintenance knowledge graph obtained based on the operations and maintenance knowledge graph acquisition method according to any one of claims 1 to 4, including: Get pending maintenance service work orders; Retrieve maintenance knowledge associated with the maintenance service work order to be processed from the maintenance knowledge graph, and calculate the fourth similarity. Based on business relevance and the fourth similarity, the operation and maintenance knowledge is sorted by weight; The pending maintenance service work order is processed based on the maintenance knowledge with the highest weight.

6. The operation and maintenance knowledge graph application method according to claim 5 further includes: Based on the operation and maintenance knowledge with the highest weight, update the weight of the corresponding operation and maintenance knowledge in the operation and maintenance knowledge graph.

7. A device for acquiring an operation and maintenance knowledge graph, comprising: The first acquisition module is used to acquire operation and maintenance resource data from the equipment information resource database; The first identification module is used to perform structured identification on the operation and maintenance resource data to obtain static operation and maintenance knowledge, wherein the static operation and maintenance knowledge includes multiple first entities; The second acquisition module is used to acquire real-time operation and maintenance service work order data; The second identification module is used to perform unstructured identification on the operation and maintenance service work order data using a pre-trained knowledge extraction model to obtain dynamic operation and maintenance knowledge, which includes multiple second entities. The third acquisition module is used to acquire the historical association relationships between the plurality of first entities and the plurality of second entities, including: acquiring historical operation and maintenance service work orders, the historical operation and maintenance service work orders including first change request work orders and event management work orders; establishing a plurality of sub-work orders of the first change request work order; acquiring the first work order association relationship between the first change request work order and the plurality of sub-work orders, and the first work order resource association relationship between the first change request work order and the plurality of first entities and the plurality of second entities contained therein; obtaining the historical association relationship based on the first work order association relationship and the first work order resource association relationship; and / or, establishing a second change request work order corresponding to the event management work order; acquiring the second work order association relationship between the event management work order and the second change request work order, and the second work order resource association relationship between the event management work order and the plurality of first entities and the plurality of second entities contained therein; obtaining the historical association relationship based on the second work order association relationship and the second work order resource association relationship. The reasoning module is used to infer a new association relationship corresponding to the first entity in the operation and maintenance service work order data based on the first similarity between the plurality of first entities, the second similarity between the plurality of second entities, and the historical association relationship; and The graph acquisition module is used to obtain the operation and maintenance knowledge graph based on the new association.

8. An electronic device, comprising: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors execute the operation and maintenance knowledge graph acquisition method according to any one of claims 1 to 4.

9. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the operation and maintenance knowledge graph acquisition method according to any one of claims 1 to 4.

Images