Network topology entity extraction method and device, storage medium and program product

Abstract

The invention provides a network topology entity extraction method and device, a storage medium and a program product, relates to the technical field of communication, and can solve the problem of poor entity extraction effect in related technologies. The method comprises the following steps: acquiring an original network topology data text; inputting the original network topology data text and the task guidance information into an entity extraction model to obtain network topology entity data; the network topology entity data is composed of network topology fields in a key value pair format; the network topology field comprises an entity type in the network topology and corresponding entity content. According to the invention, the entity extraction effect of the network topology data can be improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a network topology entity extraction method, device, storage medium and program product. Background Art

[0002] With the rapid progress in the field of artificial intelligence and the vigorous development of high-performance parallel computing technology, deep neural network technology has ushered in an unprecedented research boom and significant progress. Among them, large models have almost reshaped some common industry methods in many fields due to their excellent performance and wide application capabilities.

[0003] However, in the network topology entity extraction scenario, the original data usually contains a large number of specific professional terms, rare words and other non-standard words and symbols. Due to the recognition limitations of the current large model, the entity extraction effect for network topology data is poor. Summary of the invention

[0004] The present application provides a network topology entity extraction method, device, storage medium and program product, which can improve the entity extraction effect of network topology data.

[0005] In order to achieve the above objectives, this application adopts the following technical solutions:

[0006] In the first aspect, the present application provides a network topology entity extraction method, which includes: obtaining original network topology data text; inputting the original network topology data text and task guidance information into an entity extraction model to obtain network topology entity data; the network topology entity data consists of network topology fields in a key-value pair format; the network topology fields include entity types in the network topology and corresponding entity content.

[0007] Based on the above technical solution, the entity extraction device can obtain the original network topology data text, and input the original network topology data text and task guidance information into the entity extraction model to obtain network topology entity data. Among them, the network topology entity data is composed of network topology fields in key-value pair format, and the network topology fields include entity types in the network topology and corresponding entity contents. In this way, the present application can extract messy original data into structured data corresponding to entity types and entity contents, thereby improving the entity extraction effect of network topology data.

[0008] In combination with the first aspect above, in a possible implementation manner, the task guidance information includes at least one of a key-value pair format corresponding to the network topology field to be extracted and a prompt word.

[0009] In combination with the first aspect above, in a possible implementation manner, the prompt word includes at least one of a task objective, a data text description, and task precautions.

[0010] In combination with the first aspect above, in a possible implementation method, the training method of the entity extraction model includes the following steps: inputting the data training set and the prompt words into the entity extraction model to obtain the model output results, and updating the model parameters of the entity extraction model based on the model output results; when the model training rounds are greater than or equal to the preset training rounds, obtaining the current updated entity extraction model; performing model performance evaluation on the updated entity extraction model through the data verification set; when the model performance of the updated entity extraction model does not reach the preset performance threshold, adjusting the hyperparameters of the updated entity extraction model and re-training the model; when the model performance of the updated entity extraction model reaches the preset performance threshold, determining that the entity extraction model training is completed.

[0011] In combination with the first aspect above, in a possible implementation method, the training method of the entity extraction model further includes the following steps: performing a model test on the trained entity extraction model through a data test set to determine the model performance of the trained entity extraction model.

[0012] In combination with the first aspect above, in a possible implementation, the data training set, the data verification set and the data test set are obtained by dividing an initial data set, and the initial data set includes unprocessed original network topology data text and corresponding network topology entity data.

[0013] In the second aspect, the present application provides a network topology entity extraction device, which includes: a processing unit and a communication unit; the communication unit is used to obtain original network topology data text; the processing unit is used to input the original network topology data text and task guidance information into an entity extraction model to obtain network topology entity data; the network topology entity data is composed of network topology fields in key-value pair format; the network topology field includes entity types in the network topology and corresponding entity content.

[0014] In combination with the second aspect above, in a possible implementation manner, the task guidance information includes at least one of a key-value pair format corresponding to the network topology field to be extracted and a prompt word.

[0015] In combination with the above second aspect, in a possible implementation manner, the prompt word includes at least one of a task objective, a data text description, and task precautions.

[0016] In combination with the above-mentioned second aspect, in a possible implementation method, the processing unit is also used to: input the data training set and the prompt word into the entity extraction model to obtain the model output result, and update the model parameters of the entity extraction model based on the model output result; when the model training round is greater than or equal to the preset training round, obtain the current updated entity extraction model; evaluate the model performance of the updated entity extraction model through the data verification set; when the model performance of the updated entity extraction model does not reach the preset performance threshold, adjust the hyperparameters of the updated entity extraction model and re-train the model; when the model performance of the updated entity extraction model reaches the preset performance threshold, determine that the entity extraction model training is completed.

[0017] In combination with the above-mentioned second aspect, in a possible implementation method, the processing unit is further used to: perform model testing on the trained entity extraction model through a data test set to determine the model performance of the trained entity extraction model.

[0018] In combination with the above-mentioned second aspect, in a possible implementation method, the data training set, the data verification set and the data test set are obtained by dividing the initial data set, and the initial data set includes unprocessed original network topology data text and corresponding network topology entity data.

[0019] In a third aspect, the present application provides a network topology entity extraction device, which includes: a processor and a communication interface; the communication interface and the processor are coupled, and the processor is used to run a computer program or instruction to implement the method described in any of the above embodiments.

[0020] In a fourth aspect, the present application provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal, the terminal executes the method described in any of the above embodiments.

[0021] In a fifth aspect, the present application provides a computer program product, which includes computer program instructions. When the computer program instructions are executed by a processor, the method described in any of the above embodiments is implemented.

[0022] In a sixth aspect, the present application provides a chip, the chip comprising a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run a computer program or instruction to implement the method described in any of the above embodiments.

[0023] Specifically, the chip provided in the present application also includes a memory for storing computer programs or instructions.

[0024] It should be noted that the above computer instructions may be stored in whole or in part on a computer-readable storage medium, wherein the computer-readable storage medium may be packaged together with the processor of the device, or may be packaged separately from the processor of the device, which is not limited in this application.

[0025] The descriptions of the second to sixth aspects of the present application can refer to the detailed description of the first aspect; and the beneficial effects of the descriptions of the second to sixth aspects can refer to the beneficial effect analysis of the first aspect, which will not be repeated here.

[0026] In this application, the name of the network topology entity extraction device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they fall within the scope of the claims of this application and their equivalent technologies.

[0027] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 An architecture diagram of a network topology entity extraction system provided in an embodiment of the present application;

[0029] Figure 2 An architecture diagram of another network topology entity extraction system provided in an embodiment of the present application;

[0030] Figure 3 A flowchart of a network topology entity extraction method provided in an embodiment of the present application;

[0031] Figure 4 A flowchart of a training method for entity extraction provided in an embodiment of the present application;

[0032] Figure 5 A flowchart of another entity extraction training method provided in an embodiment of the present application;

[0033] Figure 6 A flowchart of another entity extraction training method provided in an embodiment of the present application;

[0034] Figure 7 A structural diagram of a network topology entity extraction device provided in an embodiment of the present application;

[0035] Figure 8 A structural diagram of another network topology entity extraction device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0037] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0038] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0039] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0040] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0041] In the description of the present application, unless otherwise specified, “plurality” means two or more.

[0042] With the rapid progress in the field of artificial intelligence and the vigorous development of high-performance parallel computing technology, deep neural network technology has ushered in an unprecedented research boom and significant progress. Among them, large models have almost reshaped some common industry methods in many fields due to their excellent performance and wide application capabilities.

[0043] However, the current large model has the following problems in entity extraction model:

[0044] 1. The relevant solutions for multi-language mixed texts have the problem of low extraction accuracy.

[0045] 2. The relevant solutions for the extraction of specific professional entities cannot provide comprehensive coverage, resulting in insufficient entity recognition.

[0046] In the network topology entity extraction scenario, the original data usually contains a large number of specific professional terms, rare words and other non-standard words and symbols. Due to the recognition limitations of the current large model, the entity extraction effect for network topology data is poor.

[0047] In view of this, the present application provides a network topology entity extraction method, a network topology entity extraction device (hereinafter referred to as the entity extraction device) can obtain the original network topology data text, and input the original network topology data text and task guidance information into the entity extraction model to obtain the network topology entity data. Among them, the network topology entity data is composed of a network topology field in a key-value pair format, and the network topology field includes an entity type in the network topology and the corresponding entity content. In this way, the present application can extract the messy original data into structured data corresponding to the entity type and the entity content, thereby improving the entity extraction effect of the network topology data.

[0048] The implementation of the embodiments of the present application will be described in detail below in conjunction with the accompanying drawings.

[0049] Figure 1 This is an architecture diagram of a network topology entity extraction system provided in an embodiment of the present application. Figure 1 As shown, the network topology entity extraction system includes: an entity extraction device 101 and a communication node 102.

[0050] The entity extraction device 101 and the communication node 102 are connected via a communication link, which may be a wired communication link or a wireless communication link, which is not limited in this application. The number of the communication node 102 in this application may be one or more.

[0051] The technical solutions of the embodiments of the present application can be applied to various communication systems, such as code division multiple access (CDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal frequency division multiple access (OFDMA), single carrier frequency division multiple access (SC-FDMA) and other systems. The term "system" can be interchangeable with "network". The CDMA system can implement wireless technologies such as universal terrestrial radio access (UTRA) and CDMA2000. UTRA can include wideband CDMA (WCDMA) technology and other CDMA variant technologies. CDMA2000 can cover interim standard (IS) 2000 (IS-2000), IS-95 and IS-856 standards. The TDMA system can implement wireless technologies such as global system for mobile communication (GSM). The OFDMA system can implement wireless technologies such as evolved universal radio terrestrial access (evolved UTRA, E-UTRA), ultra mobile broadband (ultra mobile broadband, UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE802.20, Flash OFDMA, etc. UTRA and E-UTRA are UMTS and UMTS evolution versions. 3GPP's long term evolution (long term evolution, LTE) and various versions based on LTE evolution are new versions of UMTS using E-UTRA. The communication system can also be a 5G communication system, a new radio (new radio, NR) 5G-A (5.5G) system, and a 6G communication system. In addition, the communication system can also be applicable to future-oriented communication technologies, and the technical solutions provided in the embodiments of the present application are applicable.

[0052] The network topology entity extraction method provided in the embodiment of the present application can be applied to the above-mentioned entity extraction device 101, and the above-mentioned entity extraction device 101 can be an independent communication device, such as a communication device deployed on the access network side or the core network side. The above-mentioned entity extraction device 101 can also be a functional module coupled in a communication device, such as a communication network element. The above-mentioned entity extraction device 101 can also be a computer program (application, APP) for executing the network topology entity extraction method. The above-mentioned entity extraction device 101 can also be a server connected to the communication device.

[0053] Exemplarily, when the entity extraction device 101 is a server connected to the communication device, the entity extraction device 101 includes:

[0054] Processor, the processor can be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present application.

[0055] Transceiver, a transceiver can be a device using any transceiver type for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0056] Memory, the memory may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may exist independently and be connected to the processor through a communication line. The memory may also be integrated with the processor.

[0057] The communication node 102 may be a terminal device, such as a terminal. The terminal is a device with wireless communication function, which can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted. It can also be deployed on the water surface (such as ships, etc.). It can also be deployed in the air (such as airplanes, balloons and satellites, etc.). The terminal is also called user equipment (UE), mobile station (MS), mobile terminal (MT) and terminal equipment, etc., which is a device that provides voice and / or data connectivity to users. For example, the terminal includes a handheld device with wireless connection function, a vehicle-mounted device, etc. At present, the terminal can be: a mobile phone, a tablet computer, a laptop computer, a PDA, a mobile internet device (MID), a wearable device (such as a smart watch, a smart bracelet, a pedometer, etc.), an in-vehicle device (such as a car, a bicycle, an electric car, an airplane, a ship, a train, a high-speed rail, etc.), a virtual reality (VR) device, an augmented reality (AR) device, a wireless terminal in industrial control, a smart home device (such as a refrigerator, a TV, an air conditioner, an electric meter, etc.), an intelligent robot, a workshop device, a wireless terminal in self-driving, a wireless terminal in remote medical surgery, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, or a wireless terminal in a smart home, a flying device (such as an intelligent robot, a hot air balloon, a drone, an airplane), etc. In a possible application scenario of the present application, the terminal device is a terminal device that often works on the ground, such as an in-vehicle device. In this application, for the sake of convenience, chips deployed in the above-mentioned devices, such as system-on-a-chip (SOC), baseband chips, etc., or other chips with communication functions may also be referred to as terminals.

[0058] The communication node 102 may also be an access network device, which includes but is not limited to: an access point (AP) in a WiFi system, such as a home gateway, a router, a server, a switch, a bridge, etc., an evolved NodeB (eNB), a radio network controller (RNC), a NodeB (NB), a base station controller (BSC), a base transceiver station (BTS), a home base station (e.g., home evolved NodeB, or home NodeB, HNB), a baseband unit (BBU), a wireless relay node, a wireless backhaul node, a transmission point (TRP or transmission point, TP), etc. It may also be a 5G base station, such as a gNB in ​​a new radio (NR) system, or a transmission point (TRP or TP), one or a group of antenna panels (including multiple antenna panels) of a base station in a 5G system, or it may also be a network node constituting a gNB or a transmission point, such as a baseband unit (BBU), or a distributed unit (distributed unit). The access network equipment also includes base stations in different networking modes, such as master evolved NodeB (MeNB), secondary eNB (SeNB, or secondary gNB, SgNB). The access network equipment also includes different types, such as ground base stations, aerial base stations, and satellite base stations.

[0059] The communication node 102 may also be a core network device, which may be a physical device or a virtual device. The core network devices of the 4G network include, but are not limited to, a mobility management entity (MME), a serving gateway (SGW), and a public data network gateway (PGW). The core network devices of the 5G network include, but are not limited to, an access and mobility management function (AMF), a session management function (SMF), and a user plane function (UPF).

[0060] As a possible embodiment, combining Figure 1 The network topology entity extraction system shown in Figure 2 As shown, the entity extraction device 101 includes: a data normalization module 1011, a training module 1012, and an entity extraction module 1013.

[0061] The data normalization module 1011 is used to perform preprocessing operations on the original network topology data.

[0062] Exemplarily, the data normalization module 1011 may perform pre-processing operations such as cleaning, formatting, and denoising on the collected original network topology data to ensure the quality and consistency of the data.

[0063] During the model training phase, the data normalization module 1011 is also used to normalize the preprocessed original network topology data to obtain the entity type and entity content in the data for subsequent model training.

[0064] The training module 1012 is used to train the entity extraction model. Exemplarily, the training module 1012 can establish an appropriate prompt pattern to assist the entity extraction model in analysis, and then input the data processed by the data normalization module 1011 into the entity extraction model as training data, thereby updating the parameters of the entity extraction model until the final trained entity extraction model is obtained.

[0065] The entity extraction module 1013 is used to extract entities according to the required entity types for the new original network topology data text, extract features and context information based on the entity extraction model trained by the training module 1012, and identify entities in the network topology.

[0066] It should be pointed out that the various embodiments of the present application can learn from or refer to each other. For example, the same or similar steps, method embodiments, system embodiments and device embodiments can all refer to each other without limitation.

[0067] Figure 3 A flowchart of a network topology entity extraction method provided in an embodiment of the present application. Figure 3 As shown, the method comprises the following steps:

[0068] Step 301: The entity extraction device obtains original network topology data text.

[0069] The original network topology data text is usually a mixture of many symbols, numbers, and multiple languages, including professional terms in the field of network topology and other non-standard words. The following is an exemplary original network topology data text:

[0070] "Assuming a given topology, devices [PE1] and [CE1] are connected, and devices [PE2] and [CE2] are connected. To configure an MPLS VPN from device [PE1] to device [PE2], and use the [vrf-test] label, where the RD value is [3:300], and the IRT and ERT values ​​are [3:300]; please perform VRF-related configuration on devices [PE1] and [PE2], and configure static routes between [PE1] and [CE1], and between [PE2] and [CE2] respectively."

[0071] Step 302: The entity extraction device inputs the original network topology data text and task guidance information into the entity extraction model to obtain network topology entity data.

[0072] The network topology entity data consists of network topology fields in a key-value pair format, and the network topology fields include entity types in the network topology and corresponding entity contents.

[0073] Exemplarily, in combination with the above example, the extracted network topology entity data is as follows:

[0074]

[0075] "ceName":"CE2"

[0076] } ]

[0078] }

[0079] The network topology entity data is in a key-value pair format, where the key is the entity type in the network topology and the value is the entity content.

[0080] In a possible implementation, the entity extraction device can perform optimization operations such as verification, deduplication, and merging on the entities and relationships in the initially extracted network topology entity data to improve the accuracy of the extraction results.

[0081] Based on the above technical solution, the entity extraction device can obtain the original network topology data text, and input the original network topology data text and task guidance information into the entity extraction model to obtain network topology entity data. Among them, the network topology entity data is composed of network topology fields in key-value pair format, and the network topology fields include entity types in the network topology and corresponding entity contents. In this way, the present application can extract messy original data into structured data corresponding to entity types and entity contents, thereby improving the entity extraction effect of network topology data.

[0082] As a possible embodiment, the task guidance information includes at least one of a key-value pair format corresponding to the network topology field to be extracted and a prompt word.

[0083] Among them, by inputting the key-value pair format corresponding to the network topology field to be extracted, the entity extraction model can be instructed to extract the specified network topology entity, and the prompt word can be used to guide the entity extraction model to output a specific type of text.

[0084] In some embodiments, the prompt word includes at least one of a task objective, a data text description, and task precautions.

[0085] Prompts refer to the use of short text prompts in natural language to guide the model to generate specific types of text. By designing well-defined prompt text, the model can be guided to produce the desired type of text output. These prompts are usually phrases or sentences that can provide hints and context about the desired output.

[0086] When using prompt words, users no longer need to describe the task in detail as in traditional programming or rule writing, but can simply provide one or more prompts to let the model automatically learn how to generate the required text. For entity extraction tasks in the field of network topology, appropriate prompt words are created to input the entity extraction model. Prompt words can make the entity extraction model fit this type of task to the greatest extent.

[0087] For example, the prompt words may include: an overview of the task to be completed by the model, an interpretation of the input, the format of the output, and emphasizing the confusing issues in the text. The following is an exemplary prompt word text:

[0088] "Based on the questions entered by the user, the Chinese and English mixed text is extracted, the network topology entities in the text are output in json format. Note that both - and _ will exist in the text name, so do not confuse them."

[0089] Based on the above technical solution, the entity extraction device in the embodiment of the present application can input the key-value pair format and / or prompt words corresponding to the network topology field to be extracted into the entity extraction model, thereby guiding the entity extraction model to output the specified network topology entity and making the entity extraction model fit such tasks to the greatest extent, thereby further improving the entity extraction effect of the network topology data.

[0090] The following introduces the training process of the entity extraction model.

[0091] As a possible embodiment of the present application, Figure 4 As shown, the training method of the entity extraction model includes the following steps:

[0092] Step 401: The entity extraction device inputs the data training set and the prompt word into the entity extraction model to obtain the model output result, and updates the model parameters of the entity extraction model based on the model output result.

[0093] For example, Figure 5 As shown, the entity extraction device can select a suitable general large model as the base model and perform model configuration, that is, initialize the hyperparameters in the model, such as the hyperparameters including learning rate, batch size, training rounds, etc.

[0094] The entity extraction device can input the data training set and the prompt words into the base model for model training. The model training process includes operations such as forward propagation, loss calculation, back propagation and parameter update.

[0095] Step 402: When the model training round is greater than or equal to the preset training round, the entity extraction device obtains the currently updated entity extraction model.

[0096] After repeated iterations, when the model training rounds reach the preset training rounds, the currently updated entity extraction model is the entity extraction model that has been preliminarily trained, and the entity extraction device can evaluate the model to determine whether the trained model meets the standards.

[0097] Step 403: The entity extraction device evaluates the model performance of the updated entity extraction model through the data verification set.

[0098] When the model performance of the updated entity extraction model does not reach the preset performance threshold, the entity extraction apparatus executes the following step 404 .

[0099] Step 404: adjust the hyperparameters of the updated entity extraction model and retrain the model.

[0100] When the model performance of the updated entity extraction model reaches a preset performance threshold, the entity extraction apparatus executes the following step 405 .

[0101] Step 405: Determine whether the entity extraction model training is completed.

[0102] Based on the above technical solution, the entity extraction device can perform model training on the entity extraction model. When the model performance does not reach the preset performance threshold, it means that the entity extraction effect of the entity extraction model obtained at this time is poor. The entity extraction device can fine-tune the hyperparameters of the model configuration and re-train the model based on the fine-tuned hyperparameters to prevent overfitting problems and ensure the extraction effect of the entity extraction model. Correspondingly, when the model performance reaches the preset performance threshold, it means that the entity extraction effect of the entity extraction model obtained at this time is good. The entity extraction device can determine that the current entity extraction model is a trained model and can be used for subsequent network topology entity extraction operations.

[0103] After the model training is completed, the entity extraction device can also perform model testing on the entity extraction model.

[0104] As a possible embodiment of the present application, Figure 4 The embodiment shown, as Figure 6 As shown, the method further includes the following steps 601.

[0105] Step 601: The entity extraction device performs a model test on the trained entity extraction model through a data test set to determine the model performance of the trained entity extraction model.

[0106] In some embodiments, the data training set, the data validation set, and the data test set are obtained by dividing an initial data set, and the initial data set includes unprocessed original network topology data text and corresponding network topology entity data.

[0107] Exemplarily, the entity extraction device may obtain an initial data set by labeling, thereby performing data division to obtain a data training set, a data verification set, and a data test set.

[0108] It should be understood that since the above-mentioned data training set and data verification set are involved in the process of model parameter updating and hyperparameter adjustment, it is difficult to determine the actual model performance of the trained entity extraction model through the data training set and the data verification set. At this time, the entity extraction device performs model testing on the trained entity extraction model through the data test set, so as to obtain the actual model performance of the entity extraction model and ensure the model quality.

[0109] The embodiment of the present application can divide the network topology entity extraction device into functional modules or functional units according to the above method example. For example, each functional module or functional unit can be divided corresponding to each function, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware or in the form of software functional modules or functional units. Among them, the division of modules or units in the embodiment of the present application is schematic, which is only a logical function division, and there may be other division methods in actual implementation.

[0110] For example, Figure 7 FIG. 7 shows a structural diagram of a network topology entity extraction device 70. The network topology entity extraction device 70 includes:

[0111] The communication unit 702 is used to obtain original network topology data text.

[0112] Processing unit 701 is used to input the original network topology data text and task guidance information into the entity extraction model to obtain network topology entity data; the network topology entity data consists of network topology fields in key-value pair format; the network topology fields include entity types in the network topology and corresponding entity contents.

[0113] In a possible implementation manner, the task guidance information includes at least one of a key-value pair format corresponding to the network topology field to be extracted and a prompt word.

[0114] In a possible implementation, the prompt word includes at least one of a task objective, a data text description, and task precautions.

[0115] In one possible implementation, the processing unit 701 is also used to: input the data training set and the prompt word into the entity extraction model to obtain the model output result, and update the model parameters of the entity extraction model based on the model output result; when the model training round is greater than or equal to the preset training round, obtain the current updated entity extraction model; evaluate the model performance of the updated entity extraction model through the data verification set; when the model performance of the updated entity extraction model does not reach the preset performance threshold, adjust the hyperparameters of the updated entity extraction model and re-train the model; when the model performance of the updated entity extraction model reaches the preset performance threshold, determine that the entity extraction model training is completed.

[0116] In a possible implementation, the processing unit 701 is further used to: perform a model test on the trained entity extraction model through a data test set to determine the model performance of the trained entity extraction model.

[0117] In a possible implementation, the data training set, the data verification set, and the data test set are obtained by dividing an initial data set, and the initial data set includes unprocessed original network topology data text and corresponding network topology entity data.

[0118] When implemented by hardware, the communication unit in the embodiment of the present application can be integrated on the communication interface, and the processing unit can be integrated on the processor. Figure 8 shown.

[0119] Figure 8 Another possible structural diagram of the network topology entity extraction device involved in the above embodiment is shown. The network topology entity extraction device 80 includes: a processor 802 and a communication interface 803. The processor 802 is used to control and manage the actions of the network topology entity extraction device 80. The communication interface 803 is used to support the communication between the network topology entity extraction device 80 and other network entities. The network topology entity extraction device 80 may also include a memory 801 and a bus 804, and the memory 801 is used to store program codes and data of the network topology entity extraction device 80.

[0120] Among them, the memory 801 can be the memory in the network topology entity extraction device 80, etc. The memory may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a read-only memory, a flash memory, a hard disk or a solid-state drive; the memory may also include a combination of the above types of memory.

[0121] The processor 802 may be a processor that implements or executes various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present application. The processor 802 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present application. The processor 802 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0122] The bus 804 may be an Extended Industry Standard Architecture (EISA) bus, etc. The bus 804 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0123] Figure 8The network topology entity extraction device 80 in the embodiment can also be a chip. The chip includes one or more than two (including two) processors 802 and a communication interface 803 .

[0124] In some embodiments, the chip further includes a memory 801, which may include a read-only memory and a random access memory, and provides operation instructions and data to the processor 802. A portion of the memory 801 may also include a non-volatile random access memory (NVRAM).

[0125] In some implementations, the memory 801 stores the following elements, execution modules or data structures, or a subset thereof, or an extended set thereof.

[0126] In the embodiment of the present application, the corresponding operation is performed by calling the operation instruction stored in the memory 801 (the operation instruction may be stored in the operating system).

[0127] Through the description of the above implementation methods, technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0128] An embodiment of the present application provides a computer program product including instructions. When the computer program product is run on a computer, the computer is enabled to execute the network topology entity extraction method in the above method embodiment.

[0129] An embodiment of the present application also provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a computer, the computer executes the network topology entity extraction method in the method flow shown in the above method embodiment.

[0130] Among them, 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 (a non-exhaustive list) include: 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), a register, a hard disk, 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, or any other form of computer readable storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application-specific integrated circuit (ASIC). In the embodiments of the present application, a computer-readable storage medium may be any tangible medium that contains or stores a program, which may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0131] Since the network topology entity extraction device, computer-readable storage medium, and computer program product in the embodiments of the present application can be applied to the above method, the technical effects that can be obtained can also refer to the above method embodiments, and the embodiments of the present application will not be repeated here.

[0132] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0133] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0134] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0135] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A network topology entity extraction method, characterized in that: The method comprises: Get the original network topology data text; The original network topology data text and task guidance information are input into the entity extraction model to obtain network topology entity data; the network topology entity data is composed of network topology fields in key-value pair format; the network topology fields include entity types in the network topology and corresponding entity contents.

2. The method according to claim 1, characterized in that The task guidance information includes at least one of a key-value pair format corresponding to the network topology field to be extracted and a prompt word.

3. The method according to claim 2, characterized in that The prompt word includes at least one of a task objective, a data text description, and task precautions.

4. The method according to any one of claims 1 to 3, characterized in that: The training method of the entity extraction model includes the following steps: Inputting the data training set and the prompt word into the entity extraction model to obtain a model output result, and updating the model parameters of the entity extraction model based on the model output result; When the model training round is greater than or equal to the preset training round, obtain the currently updated entity extraction model; Evaluate the model performance of the updated entity extraction model through the data validation set; When the model performance of the updated entity extraction model does not reach a preset performance threshold, adjusting the hyperparameters of the updated entity extraction model and re-training the model; When the model performance of the updated entity extraction model reaches a preset performance threshold, it is determined that the training of the entity extraction model is completed.

5. The method according to claim 4, characterized in that The training method of the entity extraction model further comprises the following steps: The trained entity extraction model is tested using a data test set to determine the model performance of the trained entity extraction model.

6. The method according to claim 5, characterized in that The data training set, the data verification set and the data test set are obtained by dividing an initial data set, and the initial data set includes unprocessed original network topology data text and corresponding network topology entity data.

7. A network topology entity extraction device, characterized in that: including a processing unit and a communication unit; The communication unit is used to obtain the original network topology data text; The processing unit is used to input the original network topology data text and task guidance information into the entity extraction model to obtain network topology entity data; the network topology entity data is composed of network topology fields in key-value pair format; the network topology fields include entity types in the network topology and corresponding entity content.

8. A network topology entity extraction device, characterized in that: include: A processor and a communication interface; the communication interface is coupled to the processor, and the processor is used to run a computer program or instruction to implement the network topology entity extraction method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions. When a computer executes the instructions, the computer executes the network topology entity extraction method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product comprises computer program instructions, and when the computer program instructions are executed by a processor, the network topology entity extraction method according to any one of claims 1 to 6 is implemented.

Images