Network operation and maintenance metrics question and answer method and system based on knowledge graph and large model

Abstract

A network operation and maintenance metrics question and answer method and system based on a knowledge graph and a large model, relating to the technical field of computers. According to the method, a large network model may be triggered to extract at least one type of target operation and maintenance data from entity objects, entity relationships, entity attributes, and operation and maintenance events in network operation and maintenance data, so as to construct a knowledge graph on the basis of the target operation and maintenance data; the knowledge graph can assist the large network model in responding to knowledge question and answer, and the knowledge question and answer result can further assist in enhancing a modeling process of the knowledge graph, thereby improving the response precision of the large network model to knowledge question and answer in the field of operation and maintenance; and the knowledge graph can model and represent operation and maintenance actions, thereby achieving the standardization and automation of the operation and maintenance actions. The large network model can further optimize these operation procedures by learning a large amount of operation and maintenance action data, thereby improving the efficiency of operation and maintenance.

Description

Network operation and maintenance indicator question-answering method and system based on knowledge graph and large model

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese patent application number 202410339664.4, filed on March 22, 2024, entitled “Network operation and maintenance indicator question and answer method and system based on knowledge graph and large model”. The entire contents of this Chinese patent application are incorporated by reference into this application. Technical Field

[0003] The present application relates to the field of network operation and management technology, and specifically to a network operation and maintenance indicator question and answer method based on knowledge graphs and large models, a network operation and maintenance indicator question and answer device based on knowledge graphs and large models, a network operation and maintenance indicator question and answer system based on knowledge graphs and large models, a computer-readable storage medium, and an electronic device. Background Art

[0004] Large network models possess powerful knowledge processing capabilities and possess comprehensive multimodal, multilingual, and multi-domain application capabilities. Specifically, large network models are machine learning models with large-scale parameters and complex computational structures. They are designed to improve the model's expressiveness and predictive performance, enabling it to handle more complex tasks and data. They can be applied in scenarios such as natural language processing, computer vision, speech recognition, and recommendation systems.

[0005] However, due to the limitations of corpus data such as the lack of vertical professional fields and data type diversity, large network models find it difficult to accurately respond to user requests in terms of operation and maintenance.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute an existing solution known to ordinary technicians in this field.

[0007] Summary of the Invention

[0008] The purpose of this application is to provide a network operation and maintenance indicator question and answer method based on knowledge graph and large model, a network operation and maintenance indicator question and answer device based on knowledge graph and large model, a network operation and maintenance indicator question and answer system based on knowledge graph and large model, a computer-readable storage medium and an electronic device, which can trigger the network large model to extract at least one target operation and maintenance data from the entity objects, entity relationships, entity attributes and operation and maintenance events in the network operation and maintenance data, so as to build a knowledge graph based on the target operation and maintenance data. The knowledge graph can assist the network large model in responding to knowledge questions and answers, and the knowledge question and answer results can also assist in enhancing the knowledge graph to improve the response accuracy of the network large model to knowledge questions and answers in the operation and maintenance field. The combination of knowledge graph and network large model is realized, which helps to improve the operation and maintenance question and answer capabilities of the network large model with semantic understanding and reasoning capabilities, improve the efficiency of network operation analysis, and thus help improve the efficiency of communication infrastructure construction and operation. In addition, since the knowledge graph can provide structured knowledge representation, it can also improve the indicator analysis accuracy of the network large model for knowledge questions and answers in operation and maintenance.

[0009] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by practice of the present application.

[0010] According to one aspect of the present application, a network operation and maintenance indicator question-answering method based on a knowledge graph and a large model is provided, the method comprising:

[0011] Acquire network operation and maintenance data, and trigger the network big model to extract at least one target operation and maintenance data from the entity objects, entity relationships, entity attributes, and operation and maintenance events in the network operation and maintenance data;

[0012] Trigger the network big model to build a knowledge graph based on target operation and maintenance data;

[0013] Trigger the network model to respond to the knowledge question and answer request based on the knowledge graph to generate the knowledge question and answer results;

[0014] The network big model is triggered to enhance the knowledge graph based on the knowledge question and answer results. The enhanced knowledge graph is used to assist the network big model in responding to new knowledge question and answer requests.

[0015] In an exemplary embodiment of the present application, the trigger network big model constructs a knowledge graph based on target operation and maintenance data, including:

[0016] Label the data in the target operation and maintenance data that meets the preset labeling rules to obtain the labeling results;

[0017] Trigger the network model to convert the annotation results into feature vectors;

[0018] The trigger network model builds a knowledge graph based on feature vectors.

[0019] In an exemplary embodiment of the present application, it further includes:

[0020] Clean homogeneous and heterogeneous data in network operation and maintenance data.

[0021] In an exemplary embodiment of the present application, the network big model is triggered to respond to the knowledge question and answer request according to the knowledge graph to generate a knowledge question and answer result, including:

[0022] Triggering the network large model to respond to the knowledge question and answer request, performing intent recognition on the knowledge question and answer request, and obtaining the intent recognition result;

[0023] The trigger network model converts the intent recognition results into a text matrix and extracts the text summary and text classification of the intent recognition results;

[0024] Trigger the network model to perform knowledge graph search operations based on text matrix, text summary and text classification to obtain search results;

[0025] The network model is triggered to encapsulate the search results into a matrix and convert the matrix into knowledge question answering results.

[0026] In an exemplary embodiment of the present application, triggering the network macro model to enhance the knowledge graph based on the knowledge question answering results includes:

[0027] Trigger the network model to generate relevant recommended text based on the knowledge question and answer results;

[0028] Enhance knowledge graph based on recommended text.

[0029] In an exemplary embodiment of the present application, it further includes:

[0030] The trigger network model enhances the knowledge graph based on the question-answering knowledge obtained during the knowledge question-answering process.

[0031] According to one aspect of the present application, a network operation and maintenance indicator question-answering device based on a knowledge graph and a large model is provided, the device comprising:

[0032] An operation and maintenance data acquisition unit is used to acquire network operation and maintenance data and trigger the network big model to extract at least one target operation and maintenance data from the network operation and maintenance data, including entity objects, entity relationships, entity attributes, and operation and maintenance events;

[0033] The knowledge graph construction unit is used to trigger the network model to build a knowledge graph based on the target operation and maintenance data;

[0034] The knowledge question answering unit is used to trigger the network model to respond to the knowledge question answering request based on the knowledge graph to generate the knowledge question answering results;

[0035] The knowledge graph enhancement unit is used to trigger the network large model to enhance the knowledge graph based on the knowledge question and answer results. The enhanced knowledge graph is used to assist the network large model in responding to new knowledge question and answer requests.

[0036] In an exemplary embodiment of the present application, the knowledge graph construction unit triggers the network macro model to construct a knowledge graph based on target operation and maintenance data, including:

[0037] Label the data in the target operation and maintenance data that meets the preset labeling rules to obtain the labeling results;

[0038] Trigger the network model to convert the annotation results into feature vectors;

[0039] The trigger network model builds a knowledge graph based on feature vectors.

[0040] In an exemplary embodiment of the present application, it further includes:

[0041] The data cleaning unit is used to clean homogeneous and heterogeneous data in network operation and maintenance data.

[0042] In an exemplary embodiment of the present application, the knowledge question answering unit triggers the network large model to respond to the knowledge question answering request according to the knowledge graph to generate a knowledge question answering result, including:

[0043] Triggering the network large model to respond to the knowledge question and answer request, performing intent recognition on the knowledge question and answer request, and obtaining the intent recognition result;

[0044] The trigger network model converts the intent recognition results into a text matrix and extracts the text summary and text classification of the intent recognition results;

[0045] Trigger the network model to perform knowledge graph search operations based on text matrix, text summary and text classification to obtain search results;

[0046] The network model is triggered to encapsulate the search results into a matrix and convert the matrix into knowledge question answering results.

[0047] In an exemplary embodiment of the present application, the knowledge graph enhancement unit triggers the network macro model to enhance the knowledge graph based on the knowledge question and answer results, including:

[0048] Trigger the network model to generate relevant recommended text based on the knowledge question and answer results;

[0049] Enhance knowledge graph based on recommended text.

[0050] In an exemplary embodiment of the present application, the knowledge graph enhancement unit is also used to trigger the network large model to enhance the knowledge graph based on the question and answer knowledge obtained in the knowledge question and answer process.

[0051] According to one aspect of the present application, a network operation and maintenance indicator question-answering system based on a knowledge graph and a large model is provided, the system comprising:

[0052] The offline module is used to obtain network operation and maintenance data and trigger the network large model to extract at least one target operation and maintenance data from the network operation and maintenance data, including entity objects, entity relationships, entity attributes, and operation and maintenance events;

[0053] Offline module, used to trigger the network model to build a knowledge graph based on target operation and maintenance data;

[0054] The online question-answering module is used to trigger the network model to respond to the knowledge question-answering request based on the knowledge graph to generate the knowledge question-answering results;

[0055] The online question-and-answer module is used to trigger the network model to enhance the knowledge graph based on the knowledge question-and-answer results. The enhanced knowledge graph is used to assist the network model in responding to new knowledge question-and-answer requests.

[0056] According to one aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any one of the above methods is implemented.

[0057] According to one aspect of the present application, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform any one of the above methods by executing the executable instructions.

[0058] The exemplary embodiments of the present application may have some or all of the following beneficial effects:

[0059] In the network operation and maintenance indicator question and answer method based on the knowledge graph and the big model provided in an example embodiment of the present application, the network big model can be triggered to extract at least one target operation and maintenance data among the entity objects, entity relationships, entity attributes and operation and maintenance events in the network operation and maintenance data to build a knowledge graph based on the target operation and maintenance data. The knowledge graph can assist the network big model in responding to knowledge questions and answers, and the knowledge question and answer results can also assist in enhancing the knowledge graph to improve the accuracy of the network big model's response to knowledge questions and answers in the operation and maintenance field. The combination of the knowledge graph and the network big model is achieved, which helps to improve the operation and maintenance question and answer capabilities of the network big model with semantic understanding and reasoning capabilities, improve the efficiency of network operation analysis, and thus help improve the efficiency of communication infrastructure construction and operation. In addition, because the knowledge graph can provide structured knowledge representation, it can also improve the indicator analysis accuracy of the network big model for knowledge questions and answers in operation and maintenance.

[0060] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The accompanying drawings are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0062] FIG1 schematically shows a flow chart of a network operation and maintenance indicator question-answering method based on a knowledge graph and a large model according to an embodiment of the present application;

[0063] FIG2 schematically shows a schematic diagram of an indicator question-and-answer interface according to an embodiment of the present application;

[0064] FIG3 schematically shows a diagram of visualization of network large model capabilities according to an embodiment of the present application;

[0065] FIG4 schematically shows a schematic diagram of a network large model capability structure according to an embodiment of the present application;

[0066] FIG5 schematically shows a flow chart of a network operation and maintenance indicator question-answering method based on a knowledge graph and a large model according to another embodiment of the present application;

[0067] FIG6 schematically shows a structural block diagram of a network operation and maintenance indicator question-answering system based on a knowledge graph and a large model according to an embodiment of the present application;

[0068] FIG7 schematically shows a flow diagram of module execution capability according to an embodiment of the present application;

[0069] FIG8 schematically shows an application architecture diagram according to an embodiment of the present application;

[0070] FIG9 schematically shows a structural block diagram of a network operation and maintenance indicator question-answering device based on a knowledge graph and a large model according to an embodiment of the present application;

[0071] FIG10 schematically shows a structural diagram of a computer system suitable for implementing an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0072] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; on the contrary, these embodiments are provided so that this application will be more comprehensive and complete and the concepts of the example embodiments will be fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical solutions of the present application may be practiced while omitting one or more of the specific details, or that other methods, components, devices, steps, etc. may be employed. In other cases, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present application.

[0073] In addition, the accompanying drawings are merely schematic illustrations of the present application and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the blocks shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0074] Please refer to Figure 1, which schematically shows a flow chart of a network operation and maintenance indicator question-answering method based on a knowledge graph and a large model according to an embodiment of the present application. As shown in Figure 1, the network operation and maintenance indicator question-answering method based on a knowledge graph and a large model may include: steps S110 to S140.

[0075] Step S110: Acquire network operation and maintenance data, and trigger the network big model to extract at least one target operation and maintenance data from entity objects, entity relationships, entity attributes, and operation and maintenance events in the network operation and maintenance data.

[0076] Step S120: Trigger the network big model to build a knowledge graph based on the target operation and maintenance data.

[0077] Step S130: triggering the network big model to respond to the knowledge question and answer request according to the knowledge graph to generate a knowledge question and answer result.

[0078] Step S140: triggering the network big model to enhance the knowledge graph based on the knowledge question and answer results. The enhanced knowledge graph is used to assist the network big model in responding to new knowledge question and answer requests.

[0079] Implementing the method shown in Figure 1 can trigger the network big model to extract at least one target operation and maintenance data from the entity objects, entity relationships, entity attributes, and operation and maintenance events in the network operation and maintenance data, so as to construct a knowledge graph based on the target operation and maintenance data. The knowledge graph can assist the network big model in responding to knowledge questions and answers, and the knowledge question and answer results can also assist in enhancing the knowledge graph to improve the accuracy of the network big model's response to knowledge questions and answers in the operation and maintenance field. The combination of the knowledge graph and the network big model helps to improve the operation and maintenance question and answer capabilities of the network big model with semantic understanding and reasoning capabilities, improve the efficiency of network operation analysis, and thus help improve the efficiency of communication infrastructure construction and operation. In addition, because the knowledge graph can provide structured knowledge representation, it can also improve the accuracy of the network big model's indicator analysis of knowledge questions and answers in the operation and maintenance field.

[0080] The above steps of this exemplary embodiment are described in more detail below.

[0081] In step S110, network operation and maintenance data is acquired, and the network big model is triggered to extract at least one target operation and maintenance data among entity objects, entity relationships, entity attributes and operation and maintenance events in the network operation and maintenance data.

[0082] Network operation and maintenance data refers to various types of data related to network operation and maintenance. Specifically, network operation and maintenance data includes at least business domain (B domain) data, operation domain (O domain) data, and management domain (M domain) data.

[0083] Domain B contains user and service data, such as consumer spending habits, terminal information, ARPU groups, service content, and target audiences. The Business Support System (BSS) manages telecom services, rates, and marketing, as well as customer management and service. The BSS's main subsystems include billing, customer service, accounting, settlement, and business analysis systems.

[0084] The O domain contains network data, such as signaling, alarms, faults, and network resources. The Operations Support System (OSS) is a backend support system for resources (such as networks, equipment, and computing systems). OSS includes specialized network management systems, integrated network management systems, resource management systems, service provisioning systems, and service assurance systems, providing support for reliable, secure, and stable network operation.

[0085] The M domain contains location information, such as crowd flow trajectories and map information. The Management Support System (MSS) includes all non-core business processes required by the enterprise, including the formulation of corporate strategy and development direction, enterprise risk management, audit management, public relations and image management, financial and asset management, human resources management, knowledge and R&D management, shareholder and external relationship management, procurement management, enterprise performance evaluation, government policies and laws, etc.

[0086] Because the network big model has data understanding capabilities, it can be triggered to extract at least one target operation and maintenance data from the network operation and maintenance data, including entity objects, entity relationships, entity attributes, and operation and maintenance events. The target operation and maintenance data can be used as a node in the knowledge graph.

[0087] Among them, entity objects refer to entities of network operation and maintenance data, and entity objects include at least one of equipment, interface, and traffic; entity relationships refer to relationships between entities, and entity relationships include at least one of connection relationships and interface communication relationships; entity attributes refer to attributes possessed by entities, and entity attributes include at least one of device model and interface speed; operation and maintenance events refer to various events that may be triggered during the network operation and maintenance process, and operation and maintenance events include at least one of fault events and upgrade events.

[0088] As an optional embodiment, the method further includes:

[0089] Clean homogeneous and heterogeneous data in network operation and maintenance data.

[0090] It can be seen that implementing this optional embodiment can improve data availability through data cleaning.

[0091] Specifically, homogeneous data refers to data from the same data source but in different data formats. The above-mentioned cleaning refers to data standardization operations and / or data clearing operations.

[0092] In step S120, the network big model is triggered to build a knowledge graph based on the target operation and maintenance data.

[0093] Specifically, the knowledge graph, a key branch of artificial intelligence technology, is a structured semantic knowledge base used to symbolically describe concepts and their relationships in the physical world. Its basic building blocks are "entity-relationship-entity" triples, along with entity and associated attribute-value pairs. Entities are interconnected through relationships, forming a network-like knowledge structure. Knowledge graphs help companies automatically build industry maps, eliminating the need for manual input. They can be applied to scenarios such as intelligent search, text analysis, machine reading comprehension, anomaly monitoring, and risk control.

[0094] Therefore, the knowledge graph constructed based on the target operation and maintenance data can help achieve more accurate knowledge questions and answers. Specifically, based on the knowledge graph, structural query and analysis can be performed, which has explanatory characteristics. The knowledge graph is usually used in specific vertical fields. This application takes into account that the knowledge graph and the network model can support each other as different knowledge platforms. Therefore, the entities and relationships in the knowledge graph can be represented as vectors, and trained using a neural network model to represent the entities and relationships as low-dimensional vectors in order to improve processing efficiency.

[0095] Specifically, the method of triggering the network big model to construct a knowledge graph based on the target operation and maintenance data can be: triggering the network big model to calculate operation and maintenance indicators (such as failure rate, average recovery time, etc.) based on entity objects, entity relationships, entity attributes and operation and maintenance events, and then constructing a knowledge graph based on operation and maintenance indicators, entity objects, entity relationships, entity attributes and operation and maintenance events; among which, operation and maintenance indicators can be used to evaluate the status and performance of network operation and maintenance.

[0096] As an optional embodiment of step S120, triggering the network big model to build a knowledge graph based on the target operation and maintenance data includes:

[0097] Step S1201: labeling the data in the target operation and maintenance data that conforms to the preset labeling rules to obtain labeling results;

[0098] Step S1202: triggering the network large model to convert the annotation results into feature vectors;

[0099] Step S1203: Trigger the network model to build a knowledge graph based on the feature vector.

[0100] It can be seen that by implementing this optional embodiment, operation and maintenance data can be converted into a structured knowledge graph, which can facilitate subsequent knowledge question and answer based on the knowledge graph and help improve the accuracy of knowledge question and answer.

[0101] Specifically, in an optional embodiment, since not all data in the target operation and maintenance data need to be applied to the indication map, the data in the target operation and maintenance data that meets the preset annotation rules can be annotated to obtain specific operation and maintenance data containing the mark, that is, the annotation result. Among them, if all target operation and maintenance data meet the preset annotation rules, the specific operation and maintenance data is consistent with the target operation and maintenance data. In addition, the above-mentioned preset annotation rules are used to indicate the annotation conditions and limit the characteristics of the data that can be annotated (such as data type, data format, etc.).

[0102] Furthermore, the feature vector can be used as a digital representation of the annotation results, and the feature vector can be used as a node to construct a knowledge graph.

[0103] In step S130, the network big model is triggered to respond to the knowledge question and answer request according to the knowledge graph to generate a knowledge question and answer result.

[0104] Specifically, the network big model, combined with the knowledge graph, can more efficiently and accurately respond to knowledge question and answer requests. The specific question and answer process can be seen in the knowledge question and answer visualization interface shown in Figure 2. During this process, the network big model can achieve the effect shown in Figure 3, namely, the three forces positioning of "understanding the network," "understanding operations," and "leading the development of the information and communications industry."

[0105] As an optional embodiment of step S130, triggering the network model to respond to the knowledge question and answer request according to the knowledge graph to generate a knowledge question and answer result includes:

[0106] Step S1301: triggering the network big model to respond to the knowledge question and answer request, performing intent recognition on the knowledge question and answer request, and obtaining an intent recognition result;

[0107] Step S1302: triggering the network large model to convert the intent recognition result into a text matrix, and extracting the text summary and text classification of the intent recognition result;

[0108] Step S1303: triggering the network model to perform a knowledge graph search operation based on the text matrix, text summary, and text classification to obtain search results;

[0109] Step S1304: triggering the network macro model to encapsulate the search results into a matrix, and converting the matrix into knowledge question and answer results.

[0110] It can be seen that the implementation of this optional embodiment can realize operational knowledge question and answer based on the network big model and knowledge graph.

[0111] Specifically, because the network model possesses language understanding capabilities, it can perform intent recognition on the textual content of knowledge Q&A requests, generating an intent recognition result that can be understood by the network model. The intent recognition result can be represented in any form, such as text, characters, or strings. Furthermore, the network model can represent the intent recognition result as a text matrix and traverse the knowledge graph based on the text summary and text classification of the intent recognition result, thereby obtaining search results related to the knowledge Q&A request. The search results can be encapsulated as a matrix and then converted into knowledge Q&A results for user reference.

[0112] In step S140, the network big model is triggered to enhance the knowledge graph based on the knowledge question and answer results, and the enhanced knowledge graph is used to assist the network big model in responding to new knowledge question and answer requests.

[0113] Specifically, in order to enable the knowledge graph to further assist the network large model to achieve more accurate and effective knowledge questions and answers, the enhanced knowledge graph can be enhanced based on the knowledge question and answer results. The knowledge question and answer results may include answers and / or answer analysis processes generated for the questions in the knowledge question and answer request. The capabilities of the network large model can be found in Figure 4. As shown in Figure 4, the network large model can at least achieve: intelligent task scheduling 410, domain knowledge analysis 420, rule understanding and decision-making 430, and precise intent recognition 440. The algorithm flow that the network large model can execute is: text and position encoding, masked multi-head attention mechanism, layer normalization, feedforward application network, layer normalization, and text prediction.

[0114] As an optional embodiment of step S140, triggering the network macro model to enhance the knowledge graph based on the knowledge question answering results includes:

[0115] Step S1401: triggering the network macro model to generate related recommended texts based on the knowledge question and answer results;

[0116] Step S1402: Enhance the knowledge graph based on the recommended text.

[0117] It can be seen that by implementing this optional embodiment, other unasked questions related to the knowledge question-answering results (i.e., recommended texts) can be generated for user reference, and the knowledge graph can be enhanced based on the recommended texts, thereby helping to improve the accuracy of the large network model.

[0118] Specifically, the recommended text associated with the knowledge quiz result may include preset indicators or random indicators. The recommended text may be understood as an extended reply to the knowledge quiz result for user reference.

[0119] As an optional embodiment, the method further includes:

[0120] Step S150: Trigger the network big model to enhance the knowledge graph based on the question and answer knowledge obtained during the knowledge question and answer process.

[0121] It can be seen that the implementation of this optional embodiment can achieve knowledge graph enhancement based on external question-answering knowledge, which is beneficial to improving the accuracy of large network models.

[0122] Specifically, question-and-answer knowledge can include collected user responses, which may come from different network operation indicators. The semantic alignment and mapping problems between different knowledge sources can be solved based on semantic alignment algorithms and large model technologies, and the content of user responses can be integrated into the knowledge graph to enhance the knowledge graph.

[0123] Please refer to Figure 5, which schematically shows a flow chart of a network operation and maintenance indicator question-answering method based on a knowledge graph and a large model according to another embodiment of the present application. As shown in Figure 5, the network operation and maintenance indicator question-answering method based on a knowledge graph and a large model includes: steps S510 to S590.

[0124] Step S510: Acquire network operation and maintenance data, and trigger the network big model to extract at least one target operation and maintenance data from entity objects, entity relationships, entity attributes, and operation and maintenance events in the network operation and maintenance data.

[0125] Step S520: Cleaning homogeneous heterogeneous data in the network operation and maintenance data.

[0126] Step S530: Label the data in the target operation and maintenance data that meets the preset labeling rules to obtain the labeling results; trigger the network macro model to convert the labeling results into feature vectors; trigger the network macro model to construct a knowledge graph based on the feature vectors.

[0127] Step S540: triggering the network big model to respond to the knowledge question and answer request, performing intent recognition on the knowledge question and answer request, and obtaining an intent recognition result.

[0128] Step S550: trigger the network big model to convert the intent recognition result into a text matrix, and extract the text summary and text classification of the intent recognition result.

[0129] Step S560: Trigger the network model to perform a knowledge graph search operation based on the text matrix, text summary, and text classification to obtain search results.

[0130] Step S570: triggering the network big model to encapsulate the search results into a matrix, and converting the matrix into knowledge question and answer results.

[0131] Step S580: trigger the network big model to generate related recommended texts based on the knowledge question and answer results, and enhance the knowledge graph based on the recommended texts.

[0132] Step S590: Trigger the network big model to enhance the knowledge graph based on the question and answer knowledge obtained during the knowledge question and answer process.

[0133] It should be noted that steps S510 to S590 correspond to the steps and their embodiments shown in FIG1 . For the specific implementation of steps S510 to S590 , please refer to the steps and their embodiments shown in FIG1 , which will not be repeated here.

[0134] It can be seen that the implementation of the method shown in Figure 5 can trigger the network big model to extract at least one target operation and maintenance data from the entity objects, entity relationships, entity attributes and operation and maintenance events in the network operation and maintenance data, so as to build a knowledge graph based on the target operation and maintenance data. The knowledge graph can assist the network big model in responding to knowledge questions and answers, and the knowledge question and answer results can also assist in enhancing the knowledge graph to improve the accuracy of the network big model's response to knowledge questions and answers in the operation and maintenance field. The combination of knowledge graph and network big model helps to improve the operation and maintenance question and answer capabilities of the network big model with semantic understanding and reasoning capabilities, improve the efficiency of network operation analysis, and thus help improve the efficiency of communication infrastructure construction and operation. In addition, because the knowledge graph can provide structured knowledge representation, it can also improve the accuracy of the network big model's indicator analysis of knowledge questions and answers in operation and maintenance.

[0135] Please refer to Figure 6, which schematically shows a block diagram of a network operation and maintenance indicator question-answering system based on a knowledge graph and a large model according to an embodiment of the present application. As shown in Figure 6, the network operation and maintenance indicator question-answering system based on a knowledge graph and a large model includes:

[0136] The offline module 610 is used to obtain network operation and maintenance data and trigger the network large model to extract at least one target operation and maintenance data from the network operation and maintenance data, including entity objects, entity relationships, entity attributes, and operation and maintenance events;

[0137] Offline module 610, used to trigger the network big model to build a knowledge graph based on the target operation and maintenance data;

[0138] Online question-answering module 620, used to trigger the network model to respond to the knowledge question-answering request according to the knowledge graph to generate the knowledge question-answering result;

[0139] The online question and answer module 620 is used to trigger the network large model to enhance the knowledge graph based on the knowledge question and answer results. The enhanced knowledge graph is used to assist the network large model in responding to new knowledge question and answer requests.

[0140] It can be seen that the implementation of the system shown in Figure 6 can trigger the network big model to extract at least one target operation and maintenance data from the entity objects, entity relationships, entity attributes and operation and maintenance events in the network operation and maintenance data, so as to build a knowledge graph based on the target operation and maintenance data. The knowledge graph can assist the network big model in responding to knowledge questions and answers, and the knowledge question and answer results can also assist in enhancing the knowledge graph to improve the accuracy of the network big model's response to knowledge questions and answers in the operation and maintenance field. The combination of knowledge graph and network big model helps to improve the operation and maintenance question and answer capabilities of the network big model with semantic understanding and reasoning capabilities, improve the efficiency of network operation analysis, and thus help improve the efficiency of communication infrastructure construction and operation. In addition, because the knowledge graph can provide structured knowledge representation, it can also improve the accuracy of the network big model's indicator analysis of knowledge questions and answers in operation and maintenance.

[0141] Please refer to Figure 7, which schematically shows a flow chart of the module execution capability according to an embodiment of the present application. As shown in Figure 7, the offline module 710 in the network operation and maintenance indicator question and answer system 700 based on the knowledge graph and the big model can execute steps S711 to S713. Step S711 includes obtaining network operation and maintenance data, and triggering the network big model to extract at least one target operation and maintenance data from the entity objects, entity relationships, entity attributes and operation and maintenance events in the network operation and maintenance data. Step S712 includes cleaning the homogeneous and heterogeneous data in the network operation and maintenance data. Step S713 includes labeling the data in the target operation and maintenance data that meets the preset labeling rules to obtain the labeling results; triggering the network big model to convert the labeling results into feature vectors; triggering the network big model to construct a knowledge graph based on the feature vectors.

[0142] The online question and answer module 720 in the network operation and maintenance indicator question and answer system 700 based on the knowledge graph and the big model can execute steps S721 to S722. Step S721 includes triggering the network big model to respond to the knowledge question and answer request, perform intent recognition on the knowledge question and answer request, and obtain the intent recognition result. Step S722 includes triggering the network big model to convert the intent recognition result into a text matrix, and extract the text summary and text classification of the intent recognition result. Step S723 includes triggering the network big model to perform a knowledge graph search operation based on the text matrix, text summary and text classification to obtain search results. Step S724 includes triggering the network big model to encapsulate the search results into a matrix, and convert the matrix into a knowledge question and answer result. Step S725 includes triggering the network big model to generate associated recommended text based on the knowledge question and answer result, and enhance the knowledge graph based on the recommended text. Step S726 includes triggering the network big model to enhance the knowledge graph based on the question and answer knowledge obtained during the knowledge question and answer process.

[0143] Please refer to Figure 8, which schematically shows an application architecture diagram according to an embodiment of the present application. As shown in Figure 8, the network operation and maintenance indicator question and answer system based on the knowledge graph and the big model can be implemented in the application architecture of Figure 8, which defines: a big model application layer, a big model capability layer, a big model base layer, and a big model infrastructure layer. In the big model application layer, scenario applications 810 are defined, specifically including: network planning 811, network construction 812, and network maintenance 813. In the big model capability layer, Model as a Service (MaaS) 820 is defined, specifically including: API 821, intelligent robot 822, model 823, mutually coordinated big model 830 and small model 840. In the big model base layer, data management 851 and knowledge management 852 are defined. In the big model infrastructure layer, computing power 860 is defined, specifically including: basic computing power 861 and resource pool computing power 862.

[0144] Please refer to Figure 9, which schematically shows a block diagram of a network operation and maintenance indicator question and answer device based on a knowledge graph and a large model according to an embodiment of the present application. The network operation and maintenance indicator question and answer device 900 based on a knowledge graph and a large model corresponds to the method shown in Figure 1. As shown in Figure 9, the network operation and maintenance indicator question and answer device 900 based on a knowledge graph and a large model includes:

[0145] The operation and maintenance data acquisition unit 901 is used to acquire network operation and maintenance data and trigger the network big model to extract at least one target operation and maintenance data from the network operation and maintenance data, including entity objects, entity relationships, entity attributes, and operation and maintenance events;

[0146] The knowledge graph construction unit 902 is used to trigger the network large model to construct a knowledge graph based on the target operation and maintenance data;

[0147] The knowledge question answering unit 903 is used to trigger the network model to respond to the knowledge question answering request according to the knowledge graph to generate a knowledge question answering result;

[0148] The knowledge graph enhancement unit 904 is used to trigger the network big model to enhance the knowledge graph based on the knowledge question and answer results. The enhanced knowledge graph is used to assist the network big model in responding to new knowledge question and answer requests.

[0149] It can be seen that the implementation of the device shown in Figure 9 can trigger the network big model to extract at least one target operation and maintenance data from the entity objects, entity relationships, entity attributes and operation and maintenance events in the network operation and maintenance data, so as to build a knowledge graph based on the target operation and maintenance data. The knowledge graph can assist the network big model in responding to knowledge questions and answers, and the knowledge question and answer results can also assist in enhancing the knowledge graph to improve the accuracy of the network big model's response to knowledge questions and answers in the operation and maintenance field. The combination of knowledge graph and network big model is realized, which helps to improve the operation and maintenance question and answer capabilities of the network big model with semantic understanding and reasoning capabilities, improve the efficiency of network operation analysis, and thus help improve the efficiency of communication infrastructure construction and operation. In addition, because the knowledge graph can provide structured knowledge representation, it can also improve the accuracy of the network big model's indicator analysis of knowledge questions and answers in operation and maintenance.

[0150] In an exemplary embodiment of the present application, the knowledge graph construction unit 902 triggers the network macro model to construct a knowledge graph based on the target operation and maintenance data, including:

[0151] Label the data in the target operation and maintenance data that meets the preset labeling rules to obtain the labeling results;

[0152] Trigger the network model to convert the annotation results into feature vectors;

[0153] The trigger network model builds a knowledge graph based on feature vectors.

[0154] It can be seen that by implementing this optional embodiment, operation and maintenance data can be converted into a structured knowledge graph, which can facilitate subsequent knowledge question and answer based on the knowledge graph and help improve the accuracy of knowledge question and answer.

[0155] In an exemplary embodiment of the present application, it further includes:

[0156] The data cleaning unit is used to clean homogeneous and heterogeneous data in network operation and maintenance data.

[0157] It can be seen that implementing this optional embodiment can improve data availability through data cleaning.

[0158] In an exemplary embodiment of the present application, the knowledge question answering unit 903 triggers the network large model to respond to the knowledge question answering request according to the knowledge graph to generate a knowledge question answering result, including:

[0159] Triggering the network large model to respond to the knowledge question and answer request, performing intent recognition on the knowledge question and answer request, and obtaining the intent recognition result;

[0160] The trigger network model converts the intent recognition results into a text matrix and extracts the text summary and text classification of the intent recognition results;

[0161] Trigger the network model to perform knowledge graph search operations based on text matrix, text summary and text classification to obtain search results;

[0162] The network model is triggered to encapsulate the search results into a matrix and convert the matrix into knowledge question answering results.

[0163] It can be seen that the implementation of this optional embodiment can realize operational knowledge question and answer based on the network big model and knowledge graph.

[0164] In an exemplary embodiment of the present application, the knowledge graph enhancement unit 904 triggers the network macro model to enhance the knowledge graph based on the knowledge question and answer results, including:

[0165] Trigger the network model to generate relevant recommended text based on the knowledge question and answer results;

[0166] Enhance knowledge graph based on recommended text.

[0167] It can be seen that by implementing this optional embodiment, other unasked questions related to the knowledge question-answering results (i.e., recommended texts) can be generated for user reference, and the knowledge graph can be enhanced based on the recommended texts, thereby helping to improve the accuracy of the large network model.

[0168] In an exemplary embodiment of the present application, the knowledge graph enhancement unit 904 is also used to trigger the network big model to enhance the knowledge graph based on the question and answer knowledge obtained in the knowledge question and answer process.

[0169] It can be seen that the implementation of this optional embodiment can achieve knowledge graph enhancement based on external question-answering knowledge, which is beneficial to improving the accuracy of large network models.

[0170] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiment of the application, the features and functions of two or more modules or units described above can be concretized in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be concretized.

[0171] Since the various functional modules of the network operation and maintenance indicator question and answer device based on knowledge graph and big model in the example embodiment of the present application correspond to the steps of the example embodiment of the network operation and maintenance indicator question and answer method based on knowledge graph and big model mentioned above, for details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the network operation and maintenance indicator question and answer method based on knowledge graph and big model mentioned above in the present application.

[0172] Please refer to FIG. 10 , which shows a schematic structural diagram of a computer system suitable for implementing an electronic device according to an embodiment of the present application.

[0173] It should be noted that the computer system 1000 of the electronic device shown in FIG10 is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0174] As shown in FIG10 , a computer system 1000 includes a central processing unit (CPU) 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 unit 1008 into a random access memory (RAM) 1003. Various programs and data required for system operation are also stored in the RAM 1003. The CPU 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0175] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, and the like; an output section 1007 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and speakers; a storage section 1008 including a hard disk; and a communication section 1009 including a network interface card such as a LAN card or a modem. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. Removable media 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 1010 as needed, so that computer programs read therefrom can be installed in the storage section 1008 as needed.

[0176] In particular, according to an embodiment of the present application, the process described with reference to the flowchart above can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication portion 1009, and / or installed from a removable medium 1011. When the computer program is executed by the central processing unit (CPU) 1001, the various functions defined in the method and apparatus of the present application are executed.

[0177] As another aspect, the present application further provides a computer-readable medium, which may be included in the electronic device described in the above embodiments, or may exist independently without being incorporated into the electronic device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device implements the method described in the above embodiments.

[0178] It should be noted that the computer-readable medium shown in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wireline, optical fiber cable, RF, or any suitable combination thereof.

[0179] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of the boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0180] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. In some cases, the names of these units do not constitute limitations on the units themselves.

[0181] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered merely as exemplary, and the true scope and spirit of the present application are indicated by the claims.

Claims

1. A network operation and maintenance indicator question-answering method based on knowledge graph and large model, characterized by: include: Acquire network operation and maintenance data, and trigger the network big model to extract at least one target operation and maintenance data from entity objects, entity relationships, entity attributes, and operation and maintenance events in the network operation and maintenance data; Triggering the network macro model to construct a knowledge graph based on the target operation and maintenance data; Triggering the network model to respond to the knowledge question and answer request according to the knowledge graph to generate a knowledge question and answer result; The network big model is triggered to enhance the knowledge graph based on the knowledge question and answer results, and the enhanced knowledge graph is used to assist the network big model in responding to new knowledge question and answer requests.

2. The method according to claim 1, characterized in that Triggering the network macro model to construct a knowledge graph based on the target operation and maintenance data, including: Labeling the data in the target operation and maintenance data that meets the preset labeling rules to obtain a labeling result; Triggering the network macro model to convert the annotation result into a feature vector; The network macro model is triggered to construct a knowledge graph based on the feature vector.

3. The method according to claim 1, characterized in that Also includes: Cleaning homogeneous heterogeneous data in the network operation and maintenance data.

4. The method according to claim 1, wherein Triggering the network model to respond to the knowledge question and answer request according to the knowledge graph to generate a knowledge question and answer result, including: triggering the network large model to respond to the knowledge question and answer request, performing intent recognition on the knowledge question and answer request, and obtaining an intent recognition result; Triggering the network big model to convert the intent recognition result into a text matrix, and extracting the text summary and text classification of the intent recognition result; Triggering the network model to perform a knowledge graph search operation based on the text matrix, text summary, and text classification to obtain search results; The network macro model is triggered to encapsulate the search results into a matrix, and the matrix is ​​converted into a knowledge question answering result.

5. The method according to claim 1, wherein Triggering the network macro model to enhance the knowledge graph based on the knowledge question and answer results includes: Triggering the network macro model to generate associated recommended text based on the knowledge question and answer results; The knowledge graph is enhanced based on the recommended text.

6. The method according to any one of claims 1 to 5, characterized in that Also includes: The network macro model is triggered to enhance the knowledge graph based on the question-answering knowledge acquired during the knowledge question-answering process.

7. A network operation and maintenance indicator question-answering device based on knowledge graph and large model, characterized in that: include: An operation and maintenance data acquisition unit, configured to acquire network operation and maintenance data and trigger the network macro model to extract at least one target operation and maintenance data selected from entity objects, entity relationships, entity attributes, and operation and maintenance events in the network operation and maintenance data; A knowledge graph construction unit, configured to trigger the network macro model to construct a knowledge graph based on the target operation and maintenance data; A knowledge question answering unit, configured to trigger the network macro model to respond to the knowledge question answering request according to the knowledge graph to generate a knowledge question answering result; The knowledge graph enhancement unit is used to trigger the network big model to enhance the knowledge graph based on the knowledge question and answer results, and the enhanced knowledge graph is used to assist the network big model in responding to new knowledge question and answer requests.

8. A network operation and maintenance indicator question-answering system based on knowledge graph and large model, characterized by: include: An offline module is used to obtain network operation and maintenance data and trigger the network big model to extract at least one target operation and maintenance data among entity objects, entity relationships, entity attributes and operation and maintenance events in the network operation and maintenance data; An offline module is used to trigger the network model to build a knowledge graph based on the target operation and maintenance data; An online question-answering module, configured to trigger the network model to respond to the knowledge question-answering request according to the knowledge graph to generate a knowledge question-answering result; The online question-and-answer module is used to trigger the network model to enhance the knowledge graph based on the knowledge question-and-answer results. The enhanced knowledge graph is used to assist the network model in responding to new knowledge question-and-answer requests.

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

10. An electronic device, characterized in that: include: processor; as well as a memory for storing executable instructions of the processor; The processor is configured to perform the method according to any one of claims 1 to 6 by executing the executable instructions.

Images