A data examination method and device

Abstract

The application provides a data examination method and device, relates to the technical field of data processing, and can not only reduce examination efficiency, but also waste a large amount of human resources. The method comprises the following steps: acquiring cut information and a cut examination model of a target cut type in each cut stage; determining the accuracy of the cut information according to the cut examination model; and if the accuracy of the cut information meets a preset alarm condition, sending alarm information. The application embodiment is used in a data examination process.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a data approval method and device. BACKGROUND

[0002] Currently, when performing network cutting, a complex approval process (for example, cutting pre-approval, cutting execution approval and cutting verification approval) is often required to ensure the safety and effectiveness of the cutting operation. Currently, when approving the cutting process of network cutting, artificial approval is mostly relied on, which not only reduces the approval efficiency, but also wastes a large amount of human resources. SUMMARY

[0003] The present application provides a data approval method and device, which can not only improve the approval efficiency, but also save a large amount of human resources.

[0004] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0005] In a first aspect, the present application provides a data approval method, which comprises:

[0006] Obtaining cutting information and a cutting approval model of a target cutting type at each cutting stage; the target cutting type is a cutting type corresponding to a target network cutting;

[0007] Determining the accuracy of the cutting information according to the cutting approval model;

[0008] If the accuracy of the cutting information meets a preset alarm condition, an alarm information is sent.

[0009] Based on the above technical solutions, the present application provides a data approval method, which obtains cutting information and a cutting approval model of a target cutting type at each cutting stage; determines the accuracy of the cutting information according to the cutting approval model; and if the accuracy of the cutting information meets a preset alarm condition, an alarm information is sent. Through the above method, not only the approval efficiency can be improved, but also a large amount of human resources can be saved.

[0010] Optionally, determining the accuracy of the cutting information according to the cutting approval model comprises:

[0011] Vectorizing the cutting information to obtain cutting data;

[0012] Inputting the cutting data into the cutting approval model to obtain the accuracy of the cutting information.

[0013] Optionally, vectorizing the cutting information to obtain cutting data comprises:

[0014] Extracting target cutting information from the cutting information;

[0015] The target cutover information is subjected to word segmentation processing to obtain a plurality of word segmentation processing results;

[0016] The plurality of word segmentation processing results are subjected to vector conversion to obtain a plurality of word segmentation vectors;

[0017] The plurality of word segmentation vectors are determined as cutover data.

[0018] Optionally, the method further comprises:

[0019] determining training data sets, validation data sets and a convolutional neural network model for each cutover stage of the target cutover type;

[0020] training the convolutional neural network model using the training data sets to obtain an initial cutover approval model;

[0021] adjusting parameters of the initial cutover approval model using the validation data sets to obtain a cutover approval model for the target cutover stage.

[0022] Optionally, the method further comprises:

[0023] acquiring cutover information of a preset first time period as cutover optimization information according to a preset period;

[0024] optimizing the cutover approval model according to the cutover optimization information.

[0025] Optionally, before acquiring the cutover information and the cutover approval model corresponding to each cutover stage of the target cutover type, the method further comprises:

[0026] acquiring current cutover information;

[0027] classifying the current cutover information according to each cutover type contained in the current cutover information to obtain cutover information corresponding to each cutover type;

[0028] storing each cutover type and the cutover information corresponding to each cutover type correspondingly.

[0029] Optionally, storing the cutover information corresponding to each cutover type comprises:

[0030] acquiring current cutover information;

[0031] classifying the current cutover information according to each cutover type contained in the current cutover information to obtain cutover information corresponding to each cutover type;

[0032] for cutover information corresponding to any cutover type, determining cutover information corresponding to each cutover stage of the cutover type;

[0033] storing each cutover stage of the cutover type and the cutover information corresponding to each cutover stage correspondingly.

[0034] In a second aspect, the present application provides a data approval device, and the method comprises:

[0035] An acquisition unit is configured to acquire cutover information and a cutover approval model of a target cutover type at each cutover stage; the target cutover type is a cutover type corresponding to a target network cutover;

[0036] A determination unit is configured to determine the accuracy of the cutover information according to the cutover approval model;

[0037] A sending unit is configured to send an alarm information if the accuracy of the cutover information meets a preset alarm condition.

[0038] Optionally, the determination unit is specifically configured to:

[0039] perform vectorization processing on the cutover information to obtain cutover data;

[0040] input the cutover data into the cutover approval model to obtain the accuracy of the cutover information.

[0041] Optionally, the determination unit is specifically configured to:

[0042] extract target cutover information from the cutover information;

[0043] perform word segmentation processing on the target cutover information to obtain a plurality of word segmentation processing results;

[0044] perform vector conversion on the plurality of word segmentation processing results to obtain a plurality of word segmentation vectors;

[0045] determine the plurality of word segmentation vectors as the cutover data.

[0046] In a third aspect, the present application provides a data approval device, and the device comprises a processor and a communication interface; the communication interface and the processor are coupled, and the processor is configured to run computer programs or instructions to implement the data approval method described in the first aspect and any possible implementation manner of the first aspect.

[0047] In a fourth aspect, the present application provides a computer readable storage medium, and the computer readable storage medium stores instructions, when the instructions run on a terminal, the terminal executes the data approval method described in the first aspect and any possible implementation manner of the first aspect.

[0048] In a fifth aspect, the present application provides a computer program product comprising instructions, when the computer program product runs on a data approval device, the data approval device executes the data approval method described in the first aspect and any possible implementation manner of the first aspect.

[0049] In a sixth aspect, an embodiment of the present application provides a chip, which comprises a processor and a communication interface, the communication interface is coupled with the processor, and the processor is configured to run computer programs or instructions to implement the data approval method as described in the first aspect and any possible implementation manner of the first aspect.

[0050] Specifically, the chip provided in the embodiments of the present application further comprises a memory configured to store the computer programs or instructions. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 A structural block diagram of a data approval method provided by an embodiment of the present application is shown in FIG. 1.

[0052] Figure 2 A flowchart of a data approval method provided by an embodiment of the present application is shown in FIG. 2.

[0053] Figure 3 A flowchart of vectorization processing provided by an embodiment of the present application is shown in FIG. 3.

[0054] Figure 4 A flowchart of obtaining a word segmentation vector provided by an embodiment of the present application is shown in FIG. 4.

[0055] Figure 5 A flowchart of training a cut-off approval model provided by an embodiment of the present application is shown in FIG. 5.

[0056] Figure 6 A flowchart of determining a training data set provided by an embodiment of the present application is shown in FIG. 6.

[0057] Figure 7 A structural diagram of a data approval apparatus provided by an embodiment of the present application is shown in FIG. 7.

[0058] Figure 8 Another possible structural diagram of a data approval apparatus provided by an embodiment of the present application is shown in FIG. 8. DETAILED DESCRIPTION

[0059] The data approval method and apparatus provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0060] The term “and / or” in the present document is merely used to describe an association relationship of associated objects, and can represent three relationships, for example, A and / or B can represent three cases of existence of A alone, existence of A and B simultaneously, and existence of B alone.

[0061] The terms “first” and “second” and the like in the specification and the drawings of the present application are used to distinguish different objects, or to distinguish different processing of the same object, and are not used to describe a specific order of the objects.

[0062] Moreover, the terms "comprising" and "having" and any variations thereof in the present description are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or apparatus that comprises a list of steps or units is not necessarily limited to the listed steps or units, but can optionally further include other steps or units not expressly listed or other steps or units inherent to such process, method, product, or apparatus.

[0063] It should be noted that in the embodiments of the present application, the words "exemplary" or "for example" are used to mean serving as an example, instance, or illustration. Any embodiment or design presented as "exemplary" or "for example" in the embodiments of the present application should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of "exemplary" or "for example" is intended to present concepts in a concrete manner.

[0064] With the advent of the digital era, the network has become an indispensable part of people's work and life. In order to adjust the network structure, add network equipment, replace network equipment, change lines, change device configuration or other network change requirements, network cutting is born.

[0065] However, when performing network cutting, a complex approval process (for example, cutting pre-approval, cutting execution approval, and cutting verification approval) is often required to ensure the safety and effectiveness of the cutting operation. Currently, when approving the cutting process of network cutting, artificial approval is mostly relied on, which not only reduces the approval efficiency, but also wastes a lot of human resources.

[0066] To solve the above technical problems, a data approval method provided by the embodiments of the present application,

[0067] Figure 1 The structural block diagram of the data approval method provided by the embodiments of the present application is shown in Figure 1 The structural block diagram contains a database 101, a server 102, and a user terminal 103. The server 102 can include an information acquisition module and an approval module.

[0068] In the embodiments of the present application, the server 102 can acquire the cutting information and the cutting approval model of the target cutting type at each cutting stage from the database 101 through the information acquisition module. After acquiring the cutting information and the cutting approval model, the server 102 can send the acquired cutting information and the cutting approval model to the approval module through the information acquisition module. The approval module can determine the accuracy of the cutting information according to the cutting approval model. If the accuracy of the cutting information meets the preset alarm condition, the server 102 can send alarm information to the user terminal 103.

[0069] Figure 2 A flowchart of a data approval method provided in an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the method comprises the following steps. Figure 2

[0070] In step S201, the cutover information and the cutover approval model of a target cutover type in each cutover stage are obtained.

[0071] The target cutover type is a cutover type corresponding to a target network cutover. The target network cutover is a network cutover whose cutover time is within a preset time period. The preset time period is a time period set in advance after the current time. For example, assuming that the current time is 7:00, the preset time period can be from 1:00 to 6:00 the next morning, and the target network cutover is a network cutover whose cutover time is between 1:00 and 6:00 the next morning.

[0072] In the embodiment of the present application, the preset time period can be one day (for example, from 1:00 to 6:00 the next morning), one week (7 days) (for example, from 1:00 to 6:00 the next morning every day in a week), or one month (30 days) (for example, from 1:00 to 6:00 the next morning every day in a month), which is not limited in the embodiment of the present application.

[0073] In the embodiment of the present application, the cutover type corresponding to the target network cutover can be new equipment, hardware expansion, software expansion, version upgrade, patch loading, switching and emergency drill, optimization adjustment, equipment offline, or optical cable cutover.

[0074] In the embodiment of the present application, the cutover information can include cutover type, cutover reason, cutover time, cutover location, cutover personnel, cutover scheme, emergency plan, professional type, network element type, cutover user, cutover service, equipment information, and network information.

[0075] The professional type refers to the service type to which each network cutover belongs. For example, when the cutover type of the target network cutover is new equipment, the service type to which the target network cutover belongs can be core network new equipment, bearer network new equipment, or cloud computing new equipment. The network element type refers to the network element type involved in each network cutover, such as a network node or a user server. The cutover user refers to the number of users before each network cutover, the number of users during the cutover, and the number of users after the cutover. The cutover service can include information such as the area, inter-access, optical cable segment, and electrical / optical code involved in each network cutover. The equipment information can include the equipment model, equipment manufacturer, equipment state, and configuration information of the equipment involved in each network cutover. The network information can include information such as the network connection mode, link bandwidth, topology relationship, network traffic, and network state involved in each network cutover.

[0076] ​In the embodiments of the present application, the cutting stage can include a cutting pre-audit stage, a cutting execution stage and a cutting verification stage, and correspondingly, the cutting approval model can include a cutting pre-audit approval model, a cutting execution approval model and a cutting verification approval model.

[0077] The cutting information corresponding to the cutting pre-audit stage can include any one or more of cutting reasons, cutting times, cutting locations, cutting personnel, cutting schemes, emergency plans, professional types, network element types, cutting users, cutting services before cutting, device information before cutting and network information before cutting.

[0078] The cutting information corresponding to the cutting execution stage can include any one or more of cutting services at the time of cutting, device information at the time of cutting and network information at the time of cutting.

[0079] The cutting information corresponding to the cutting verification stage can include any one or more of cutting services after cutting, device information after cutting and network information after cutting.

[0080] Step S202, determining the accuracy of the cutting information according to the cutting approval model.

[0081] Step S203, if the accuracy of the cutting information meets the preset alarm condition, sending an alarm information.

[0082] In an optional implementation, the alarm condition can include but is not limited to an accuracy less than a preset accuracy threshold. The accuracy threshold can be 90%, or 95%, which is not limited in the embodiments of the present application.

[0083] The data approval method provided by the embodiments of the present application will be described below with respect to the cutting pre-audit stage, the cutting execution stage and the cutting verification stage respectively.

[0084] I. Cutting pre-audit stage

[0085] Before performing the target network cutting, the target network cutting can be determined according to the cutting time of each network cutting and a preset time period. After determining the target network cutting, the cutting type corresponding to the target network cutting can be determined as the target cutting type. Then, the cutting information corresponding to the target cutting type in the cutting pre-audit stage and the cutting pre-audit approval model are obtained, and the accuracy of the cutting information corresponding to the cutting pre-audit stage is determined by the cutting pre-audit approval model. If the accuracy of the cutting information is greater than or equal to the preset accuracy threshold, the compliance identifier corresponding to the cutting information is set to 1 (indicating that the cutting information is compliant).

[0086] If the accuracy of the cutover information is less than the preset accuracy threshold, the compliance identifier corresponding to the cutover information is set to 0 (indicating that the cutover information is non-compliant), and an alarm information is sent to the user terminal. After receiving the alarm information, the user terminal can manually review the cutover information according to the stage identifier information of the cutover stage and the cutover information carried in the alarm information. If the operator determines that the cutover information is compliant, the user terminal triggers a compliance operation, and the user terminal sends a compliance instruction to the server in response to the compliance operation of the operator. After receiving the compliance instruction, the server updates the compliance identifier corresponding to the cutover information from 0 to 1. If the operator determines that the cutover information is non-compliant, the cutover information can be manually modified, and the modified cutover information is sent to the server through the user terminal. After receiving the modified cutover information, the server can determine the accuracy of the modified cutover information by the cutover pre-review approval model according to the above steps.

[0087] After setting the compliance identifier of the cutover information corresponding to the cutover pre-review stage in the above manner, when the current time is the execution time of the cutover pre-review stage, the server can first obtain the compliance identifier of the cutover information corresponding to the cutover pre-review stage of the target network cutover. If the compliance identifier indicates that the cutover information corresponding to the cutover pre-review stage is compliant, the cutover information corresponding to the cutover pre-review stage is pre-reviewed and approved. If the compliance identifier indicates that the cutover information corresponding to the cutover pre-review stage is non-compliant, the pre-review and approval of the cutover information corresponding to the cutover pre-review stage is suspended.

[0088] II. Cutover Execution Stage

[0089] In executing the target network cutover, the cutover execution approval model corresponding to the target cutover type in the cutover execution stage can be obtained first, and the cutover information corresponding to the target cutover type in the cutover execution stage can be obtained at the preset at least one execution node. That is, when the current time reaches the execution time of the preset at least one execution node, the cutover business, device information and network information of the current time are obtained. After obtaining the cutover information, the accuracy of the cutover information can be determined by the cutover execution approval model. If the accuracy of the cutover information is greater than or equal to the preset accuracy threshold, the compliance identifier corresponding to the cutover information is set to 1 (indicating that the cutover information is compliant).

[0090] If the accuracy of the cutover information is less than the preset accuracy threshold, the compliance identifier corresponding to the cutover information is set to 0 (indicating that the cutover information is non-compliant), and an alarm information is sent to the user terminal. After receiving the alarm information, the user terminal can manually review the cutover information according to the stage identifier information of the cutover stage and the cutover information carried in the alarm information. If the operator determines that the cutover information is compliant, a compliant operation can be triggered on the user terminal. The user terminal responds to the compliant operation of the operator and sends a compliant instruction to the server. After receiving the compliant instruction, the server updates the compliance identifier corresponding to the cutover information from 0 to 1, and continues to perform the target network cutover. If the operator determines that the cutover information is non-compliant, a non-compliant operation can be triggered on the user terminal. The user terminal responds to the non-compliant operation of the operator and sends a non-compliant instruction to the server. After receiving the non-compliant instruction, the server can stop performing the target network cutover. Alternatively, if the operator determines that the cutover information is non-compliant, the cutover information can be manually modified, and the modified cutover information is sent to the server through the user terminal. After receiving the modified cutover information, the server can determine the accuracy of the modified cutover information according to the above steps by using the cutover execution review model.

[0091] III. Cutover verification stage

[0092] At the end of the target network cutover execution, the cutover information corresponding to the target cutover type in the cutover verification stage and the cutover verification review model can be obtained, i.e. the cutover business, device information and network information after the cutover are obtained. After obtaining the cutover information, the accuracy of the cutover information can be determined by using the cutover verification review model. If the accuracy of the cutover information is greater than or equal to the preset accuracy threshold, the compliance identifier corresponding to the cutover information is set to 1 (indicating that the cutover information is compliant).

[0093] If the accuracy of the cutover information is less than the preset accuracy threshold, the compliance identifier corresponding to the cutover information is set to 0 (indicating that the cutover information is non-compliant), and an alarm information is sent to the user terminal. After receiving the alarm information, the user terminal can manually review the cutover information according to the stage identifier information of the cutover stage and the cutover information carried in the alarm information. If the operator determines that the cutover information is compliant, a compliant operation can be triggered on the user terminal. The user terminal responds to the compliant operation of the operator and sends a compliant instruction to the server. After receiving the compliant instruction, the server updates the compliance identifier corresponding to the cutover information from 0 to 1. If the operator determines that the cutover information is non-compliant, a non-compliant operation can be triggered on the user terminal. The user terminal responds to the non-compliant operation of the operator and sends a non-compliant instruction to the server. After receiving the non-compliant instruction, the server can execute an emergency plan and return to before the target network cutover is executed.

[0094] By the above manner, the cutover information in the cutover pre-auditing stage, the cutover execution stage and the cutover verification stage can be automatically audited without manual auditing by the operation personnel or the auditing personnel, so that the auditing efficiency of the cutover information in the cutover stages can be improved, and a large amount of human resources can be saved.

[0095] In an optional implementation, before the cutover information corresponding to each cutover stage of the target cutover type is acquired, the current cutover information can also be acquired, and the current cutover information is stored according to each cutover type contained in the current cutover information.

[0096] The current cutover information refers to the cutover information of each network cutover in a preset time period. For example, the preset time period is from 1:00 to 6:00 in the morning, and the current cutover information refers to the cutover information of each network cutover in the time period from 1:00 to 6:00 in the morning.

[0097] Specifically, in some embodiments, in the process of storing the current cutover information according to each cutover type contained in the current cutover information, the current cutover information can be acquired, the current cutover information can be classified according to each cutover type contained in the current cutover information, the cutover information corresponding to each cutover type can be obtained, and each cutover type and the cutover information corresponding to each cutover type can be stored correspondingly.

[0098] For example, in an embodiment, assuming that the cutover types in the current cutover information include “new device” and “hardware expansion”, the cutover information corresponding to “new device” and the cutover information corresponding to “hardware expansion” can be acquired from the current cutover information, and then the “new device” and the cutover information corresponding to “new device” can be stored correspondingly, and the “hardware expansion” and the cutover information corresponding to “hardware expansion” can be stored correspondingly.

[0099] In other embodiments, in the process of storing the current cutover information according to each cutover type contained in the current cutover information, the current cutover information can be acquired, the current cutover information can be classified according to each cutover type contained in the current cutover information, the cutover information corresponding to each cutover type can be obtained, for the cutover information corresponding to any cutover type, the cutover information corresponding to each cutover stage of the cutover type can be determined, and each cutover stage of the cutover type and the cutover information corresponding to each cutover stage can be stored correspondingly.

[0100] Exemplarily, in an embodiment, it is assumed that the cutover types in the current cutover information include "new device" and "hardware expansion", and each cutover type corresponds to a cutover stage including three stages of cutover pre-audit stage, cutover execution stage and cutover verification stage. After obtaining the current cutover information, the cutover information corresponding to the "new device" and the cutover information corresponding to the "hardware expansion" can be obtained from the current cutover information respectively, and then the cutover information corresponding to the cutover pre-audit stage, the cutover information corresponding to the cutover execution stage and the cutover information corresponding to the cutover verification stage can be obtained from the cutover information corresponding to the "new device" respectively, and the cutover information corresponding to the cutover pre-audit stage, the cutover information corresponding to the cutover execution stage and the cutover information corresponding to the cutover verification stage can be obtained from the cutover information corresponding to the "hardware expansion" respectively. After obtaining the cutover information corresponding to each cutover stage of each cutover type, each cutover type, each cutover stage and the cutover information corresponding to each cutover stage of each cutover type can be stored correspondingly. For example, [new device, cutover pre-audit stage, cutover information corresponding to the cutover pre-audit stage of the new device], [new device, cutover execution stage, cutover information corresponding to the cutover execution stage of the new device], [new device, cutover verification stage, cutover information corresponding to the cutover verification stage of the new device], [hardware expansion, cutover pre-audit stage, cutover information corresponding to the cutover pre-audit stage of the hardware expansion], [hardware expansion, cutover execution stage, cutover information corresponding to the cutover execution stage of the hardware expansion], [hardware expansion, cutover verification stage, cutover information corresponding to the cutover verification stage of the hardware expansion] are stored.

[0101] In an optional implementation, in the process of performing the above step S202 (determining the accuracy of the cutover information according to the cutover approval model), the cutover information can be subjected to vectorization processing to obtain cutover data, the cutover data is input into the cutover approval model, and the accuracy of the cutover information is obtained.

[0102] Specifically, in some embodiments, in the process of subjecting the cutover information to vectorization processing to obtain cutover data, the method shown in FIG. 6 can be referred to, and the method shown in FIG. 7 can be referred to. Figure 3 Figure 3 The method includes the following steps.

[0103] Step S301: extracting target cutover information from the cutover information.

[0104] Specifically, after extracting the cutover information of the target cutover type in the target cutover stage, the target cutover information can be extracted from the cutover information.

[0105] In an optional implementation, the target cutover information can include one or more of the following information:

[0106] ​Basic cutover information: cutover type, cutover reason, cutover time, cutover location, professional type, cutover user, cutover service and network element type.

[0107] Cutover personnel information: personnel identification information, personnel experience information, and personnel skill information.

[0108] Cutover plan information: cutover preparation information, cutover operation procedure, cutover inspection information, and cutover test information.

[0109] Emergency response plan information: Rewind preparation information, rewind operation procedures, rewind inspection information, and rewind test information.

[0110] Equipment information: equipment model, equipment manufacturer, equipment status, and equipment configuration information.

[0111] Network information: Information such as network connection method, link bandwidth, topology, network traffic and network status.

[0112] In one optional implementation, after extracting the target cutover information in the above manner, data cleaning can be performed on the target cutover information. Data cleaning may include one or more data processing operations such as removing duplicate data, processing missing data, and processing abnormal data.

[0113] By using the above method, after obtaining the cutover information, the target cutover information is extracted from the cutover information, and data cleaning is performed on the target cutover information. This not only reduces the interference of irrelevant features and improves the efficiency and accuracy of data processing, but also improves the quality and usability of the data.

[0114] Step S302: Perform word segmentation on the target cut-off information to obtain multiple word segmentation results.

[0115] Each word segmentation result can include multiple words.

[0116] Specifically, in the process of segmenting the target cutover information, Chinese cutover information can be extracted from the target cutover information, and then segmented according to a preset segmentation algorithm to obtain multiple segmentation results. Here, Chinese cutover information refers to the Chinese characters contained in the target cutover information.

[0117] In this embodiment, the word segmentation algorithm can be either Python's Jieba algorithm or the HanLP algorithm; this embodiment does not limit the specific algorithm used.

[0118] Using the above method, the target cutover information can be segmented according to certain rules and algorithms, and transformed into a series of meaningful words, thereby facilitating subsequent processing and analysis.

[0119] In an optional embodiment, after obtaining the plurality of word segmentation processing results, the word frequency of each word contained in the word segmentation processing results can be counted, and the stop words and low-frequency words contained in the word segmentation processing results can be removed.

[0120] In step S303, the plurality of word segmentation processing results are vector converted to obtain a plurality of word segmentation vectors.

[0121] After obtaining the plurality of word segmentation processing results through step S303, the plurality of word segmentation vectors can be obtained by referring to the method shown in FIG. 3B, as shown in FIG. 3C, the method comprises the following steps. Figure 4 Figure 4 After obtaining the plurality of word segmentation processing results through step S303, the plurality of word segmentation vectors can be obtained by referring to the method shown in FIG. 3B, as shown in FIG. 3C, the method comprises the following steps.

[0122] In step S401, a word segmentation mapping table corresponding to each word segmentation processing result is determined.

[0123] After obtaining the plurality of word segmentation processing results through step S303, the word table corresponding to each word segmentation processing result can be obtained according to the preset relationship mapping table. The preset relationship mapping table contains a plurality of words and word identification information corresponding to each word. The word identification information is used to uniquely identify the corresponding word.

[0124] In the embodiments of the present application, the word identification information corresponding to each word can be a numerical value (for example, an integer value), or a character, which is not limited in the embodiments of the present application.

[0125] Specifically, for any word segmentation processing result, the word identification information corresponding to each word contained in each word segmentation processing result can be obtained according to the relationship mapping table, and the word segmentation mapping table corresponding to each word segmentation processing result can be formed according to the word identification information corresponding to each word in each word segmentation processing result.

[0126] Exemplarily, in an embodiment, assuming that the word identification information is an integer type numerical value, the integer type numerical value corresponding to each word contained in each word segmentation processing result can be obtained according to the relationship mapping table, and the word segmentation mapping table corresponding to each word segmentation processing result can be formed according to the integer type numerical value corresponding to each word in each word segmentation processing result.

[0127] In step S402, a word vector model corresponding to each word segmentation processing result is determined.

[0128] After obtaining the plurality of word segmentation processing results through step S303, the word vector model corresponding to each word segmentation processing result can be obtained by using a preset vector conversion algorithm.

[0129] In the embodiments of the present application, the preset vector conversion algorithm can include but is not limited to the word2vec algorithm, which is not limited in the embodiments of the present application.

[0130] ​Step S403, obtaining the word segmentation vector corresponding to each word segmentation processing result based on the word segmentation mapping table and the word vector model corresponding to the word segmentation processing result.

[0131] After obtaining the word segmentation mapping table and the word vector model corresponding to each word segmentation processing result through step S401 and step S402, the word segmentation vector corresponding to each word segmentation processing result can be determined based on the word segmentation mapping table and the word vector model corresponding to each word segmentation processing result.

[0132] In the embodiment of the present application, when the dimensions of the word segmentation vectors corresponding to each word segmentation processing result are different, the word segmentation vectors can be zero-padded or truncated according to the preset dimension parameter, so that the dimensions of the word segmentation vectors corresponding to each word segmentation processing result are consistent, facilitating the processing of the cutting approval model.

[0133] Step S304, determining the plurality of word segmentation vectors as cutting data.

[0134] After determining the word segmentation vector corresponding to each word segmentation processing result in the above manner, the word segmentation vector corresponding to each word segmentation processing result can be determined as the cutting data of the target cutting type in the target cutting stage.

[0135] Through the above method, the cutting information of the target cutting type in each cutting stage obtained can be subjected to a series of operations such as numerical conversion, data cleaning and vector conversion,

[0136] In an optional embodiment, before obtaining the cutting approval model of the target cutting type in each cutting stage, the convolutional neural network model needs to be trained to obtain the cutting approval model. Specifically, the method shown in Figure 5 may be referred to. Figure 5 A training method of a cutting approval model of a target cutting type in each cutting stage provided by an embodiment of the present application, as shown in Figure 5 , the method comprises:

[0137] Step S501, determining a training data set, a validation data set and a convolutional neural network model of a target cutting type in a target cutting stage.

[0138] The target cutting stage can be a cutting pre-examination stage, a cutting execution stage or a cutting verification stage.

[0139] Specifically, the training data set of the target cutting type in the target cutting stage can be determined by referring to the method shown in Figure 6 , as shown in Figure 6 , the method comprises:

[0140] Step S601, obtaining cutting information of a target cutting type in a target cutting stage in a historical time period.

[0141] In the embodiments of the present application, the historical time period is pre-set, and the historical time period can be one year (e.g., from January 1, 2000 to January 1, 2001) or one month (e.g., from January 1, 2000 to February 1, 2000). The embodiments of the present application do not limit the historical time period.

[0142] Specifically, in some embodiments, all cutover information contained in the historical time period can be acquired first, then the cutover information of the target cutover type is extracted from all the cutover information, and finally the cutover information of the target cutover stage is extracted from the cutover information of the target cutover type.

[0143] For example, in an embodiment, it is assumed that the target cutover type is the “new device” type, the target cutover stage is the cutover pre-audit stage, and the historical time period is from January 1, 2000 to January 1, 2001. All cutover information contained in the period from January 1, 2000 to January 1, 2001 can be acquired first, then the cutover information corresponding to the “new device” type is extracted from all the cutover information, and after the cutover information corresponding to the “new device” type is extracted, the cutover information corresponding to the cutover pre-audit stage is extracted from the cutover information corresponding to the “new device” type.

[0144] Step S602, extracting the target cutover information from the cutover information.

[0145] Step S602 is similar to the execution method of step S301 described above, and will not be described here again.

[0146] Step S603, performing word segmentation processing on the target cutover information to obtain a plurality of word segmentation processing results.

[0147] Step S603 is similar to the execution method of step S302 described above, and will not be described here again.

[0148] Step S604, determining a word segmentation mapping table corresponding to each word segmentation processing result.

[0149] Step S604 is similar to the execution method of step S401 described above, and will not be described here again.

[0150] Step S605, determining a word vector model corresponding to each word segmentation processing result.

[0151] Step S605 is similar to the execution method of step S402 described above, and will not be described here again.

[0152] Step S606, obtaining a word segmentation vector corresponding to each word segmentation processing result based on the word segmentation mapping table and the word vector model corresponding to the word segmentation processing result.

[0153] Step S606 is similar to the execution method of step S403 described above, and will not be described here.

[0154] After obtaining the word segmentation mapping table and the word vector model corresponding to each word segmentation processing result through steps S604 and S605, the word segmentation vector corresponding to each word segmentation processing result can be determined based on the word segmentation mapping table and the word vector model corresponding to each word segmentation processing result.

[0155] After determining the word segmentation vector corresponding to each word segmentation processing result in the above manner, the word segmentation vector corresponding to each word segmentation processing result can be determined as the training data set of the target segmentation type in the target segmentation stage.

[0156] Similarly, the validation data set of the target segmentation type in the target segmentation stage can also be obtained by the method shown in Figure 4 It should be noted that the historical time period for obtaining the validation data set and the historical time period for obtaining the training data set should be different time periods.

[0157] Step S502, training the convolutional neural network model using the training data set to obtain an initial segmentation approval model.

[0158] Specifically, the training data set of the target segmentation type in the target segmentation stage can be input into the initial segmentation approval model for training, and the weight and bias parameters of the convolutional neural network model can be continuously updated through the back propagation algorithm until the value of the loss function is minimized, to obtain the initial segmentation approval model of the target segmentation stage.

[0159] In the embodiments of the present application, the convolutional neural network model can include an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer.

[0160] The convolutional layer can include one or more convolutional kernels. In the embodiments of the present application, the width of the convolutional kernel is the same as the dimension of the word vector, and the height (i.e. the window value) of the convolutional kernel can be adjusted according to the segmentation data of different segmentation types, and generally can be selected between 3-6. The output of the convolutional layer can pass through a ReLU activation function to increase the nonlinearity of the model.

[0161] The pooling layer can filter the feature data extracted in the convolutional layer, perform dimension reduction processing, reduce the number of parameters, avoid overfitting, and thus improve the robustness of the model. In the embodiments of the present application, the pooling layer can use maximum value pooling to obtain a fixed-length input of the fully connected layer from the output of the indefinite-length convolutional layer. After the convolution and pooling operations of the convolutional layer and the pooling layer, the word segmentation vector corresponding to each word segmentation processing result can further extract the feature information of the segmentation data.

[0162] The full connection layer is equivalent to a classifier, which can fuse the output of the convolution layer and the pooling layer to generate a final classification result. The full connection layer can map the feature data into identification information and send the identification information to the output layer.

[0163] The identification information accuracy probability can be output by an activation function in the output layer. In the embodiment of the present application, the activation function can select a sigmoid function. The loss function can select cross-entropy, and the optimizer can select an Adam optimizer. The output probability data is taken as prediction data of the cross-entropy cost function, and the prediction error of the model is minimized by processing through the loss function and the optimizer.

[0164] In step S503, the parameters of the initial cut approval model are adjusted using the verification data set to obtain a cut approval model of the target cut stage.

[0165] Specifically, after obtaining the initial cut approval model through step S502, the verification data set of the target cut type in the target cut stage can be input into the initial cut approval model for verification. If the verification is passed, the initial cut approval model is determined as the cut approval model. If the verification is not passed, the parameters of the initial cut approval model are adjusted using the verification data set to obtain the cut approval model.

[0166] In the embodiment of the present application, the parameters of the initial cut approval model can include learning rate, regularization parameter, network structure and other parameters, which are not limited in the embodiment of the present application.

[0167] Through the above technical solution, the parameters of the initial cut approval model are adjusted through the verification data set, which can improve the generalization ability and performance of the cut approval model.

[0168] In an optional embodiment, after obtaining the cut approval model through the above method, the cut approval model can be optimized. Specifically, the cut information of a preset first time period can be obtained as cut optimization information according to a preset period; and the cut approval model is optimized according to the cut optimization information.

[0169] In the embodiment of the present application, the preset period can be 7 days or 5 days, which is not limited in the embodiment of the present application.

[0170] In the embodiment of the present application, the first time period can be 7 days or 5 days, which is not limited in the embodiment of the present application.

[0171] For example, in an embodiment, assuming that the preset period can be 7 days and the first time period is 5 days, the server can obtain the cut information of the previous 5 days every 7 days, and optimize the cut approval model using the cut information of the 5 days.

[0172] In an optional implementation, after obtaining the cutover approval model in the manner described above, the cutover approval model can be evaluated. If the evaluation fails, the cutover approval model is optimized.

[0173] In the embodiments of the present application, in the process of evaluating the cutover approval model, the cutover approval model can be evaluated by the accuracy, recall rate, F1 score and the like of the cutover approval model, which is not limited in the embodiments of the present application.

[0174] Through the technical solution described above, the cutover approval model can be continuously optimized, thereby enhancing the adaptability of the cutover approval model.

[0175] Figure 7 A structural schematic diagram of a data approval device provided in the embodiments of the present application is shown in FIG. 7, which includes: Figure 7

[0176] The obtaining unit 701 is configured to obtain cutover information and a cutover approval model of a target cutover type at each cutover stage.

[0177] The target cutover type is a cutover type corresponding to a target network cutover;

[0178] The determining unit 702 is configured to determine the accuracy of the cutover information according to the cutover approval model.

[0179] The sending unit 703 is configured to send an alarm information if the accuracy of the cutover information meets a preset alarm condition.

[0180] Optionally, the determining unit 702 is specifically configured to:

[0181] vectorize the cutover information to obtain cutover data;

[0182] input the cutover data into the cutover approval model to obtain the accuracy of the cutover information.

[0183] Optionally, the determining unit is specifically configured to:

[0184] extract target cutover information from the cutover information;

[0185] perform word segmentation processing on the target cutover information to obtain a plurality of word segmentation processing results;

[0186] perform vector conversion on the plurality of word segmentation processing results to obtain a plurality of word segmentation vectors;

[0187] determine the plurality of word segmentation vectors as the cutover data.

[0188] Optionally, the device further includes a processing unit, which is specifically configured to:

[0189] ​determine the target cutover type, the training data set, the validation data set and the convolutional neural network model of the target cutover stage;

[0190] train the convolutional neural network model using the training data set to obtain an initial cutover approval model;

[0191] adjust the parameters of the initial cutover approval model using the validation data set to obtain a cutover approval model of the target cutover stage.

[0192] Optionally, the apparatus further comprises an optimization unit, which is specifically configured to:

[0193] acquire cutover information of a preset first time period as cutover optimization information according to a preset period;

[0194] optimize the cutover approval model according to the cutover optimization information.

[0195] Optionally, the apparatus further comprises a storage unit, which is specifically configured to:

[0196] acquire current cutover information;

[0197] classify the current cutover information according to each cutover type contained in the current cutover information to obtain cutover information corresponding to each cutover type;

[0198] store each cutover type and the cutover information corresponding to each cutover type correspondingly.

[0199] Optionally, the storage unit is further configured to:

[0200] acquire current cutover information;

[0201] classify the current cutover information according to each cutover type contained in the current cutover information to obtain cutover information corresponding to each cutover type;

[0202] for cutover information corresponding to any cutover type, determine cutover information corresponding to each cutover stage of the cutover type;

[0203] store each cutover stage of the cutover type and the cutover information corresponding to each cutover stage correspondingly.

[0204] Figure 8 Another possible structural schematic diagram of the data approval apparatus involved in the above embodiments is shown. The data approval apparatus comprises a processor 801 and a communication interface 802. The processor 801 is configured to control and manage the actions of the data approval apparatus, and the communication interface 802 is configured to support the communication between the data approval apparatus and other network entities. The data approval apparatus can further comprise a memory 803 and a bus 804, and the memory 803 is configured to store the program code and data of the data approval apparatus.

[0205] The memory 803 can be a memory or the like in the data approval device, and can include a volatile memory such as a random access memory, and can also include a non-volatile memory such as a read-only memory, a flash memory, a hard disk, or a solid state disk, and can also include a combination of the above kinds of memories.

[0206] The processor 801 described above can be various exemplary logical blocks, modules and circuits described in combination with the disclosure of the present application. The processor can 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 device, a transistor logic device, a hardware component or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the disclosure of the present application. The processor can also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of DSP and microprocessor, etc.

[0207] The bus 804 can be an extended industry standard architecture (EISA) bus or the like. The bus 804 can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 8 Only one thick line is used in the figure, but it does not mean that there is only one bus or only one type of bus.

[0208] Through the description of the above embodiments, those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of functional modules is exemplified, and in actual application, the above-mentioned functions can be completed by different functional modules according to needs, 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 foregoing method embodiments, which will not be described here.

[0209] The embodiment of the present application provides a computer program product containing instructions, when the computer program product runs on a computer, so that the computer executes the data approval method in the method embodiment described above.

[0210] The embodiment of the present application also provides a computer readable storage medium, and the computer readable storage medium stores instructions, when the instructions run on the computer, so that the computer executes the data approval method in the method flow shown in the method embodiment described above.

[0211] The computer readable storage medium, for example, can be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a register, a hard disk, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing, or any other medium from which a processor can read instructions. An exemplary computer readable storage medium is coupled to the processor such that the processor can read information from, and write information to, the computer readable storage medium. Of course, the computer readable storage medium can be a component of the processor. Suitable processors include, by way of example, both general and special purpose microprocessors. The processor can also be any custom made or off-the-shelf processor capable of executing the methodologies having been described herein, whether provided as standalone device or as part of a combined processor / digital signal processor system.

[0212] An embodiment of the present application provides a computer program product containing instructions which, when executed on a computer, cause the computer to perform the data approval method described in the embodiments of the present application.

[0213] Since the data approval device, the computer readable storage medium and the computer program product in the embodiments of the present application can be applied to the above method, the technical effects they can obtain can be referred to the above method embodiments, which will not be described here again in the embodiments of the present application.

[0214] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented by other means. For example, the above-described device embodiments are only illustrative, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0215] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may also be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0216] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0217] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any change or replacement within the technical scope disclosed in the present application should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A data approval method, characterized by, The method comprises: obtaining current cutover information; classifying the current cutover information according to each cutover type contained in the current cutover information to obtain cutover information corresponding to each cutover type; for cutover information corresponding to any cutover type, determining cutover information corresponding to each cutover stage of the cutover type; storing each cutover stage of the cutover type and the cutover information corresponding to each cutover stage correspondingly; determining a target network cutover according to a cutover time of each network cutover and a preset time period; obtaining cutover information and a cutover approval model of a target cutover type at each cutover stage; the target cutover type is a cutover type corresponding to the target network cutover; determining an accuracy of the cutover information according to the cutover approval model; if the accuracy of the cutover information meets a preset alarm condition, sending alarm information.

2. The method of claim 1, wherein, The method comprises: vectorizing the cutover information to obtain cutover data; inputting the cutover data into the cutover approval model to obtain the accuracy of the cutover information.

3. The method of claim 2, wherein, The method comprises: extracting target cutover information from the cutover information; performing word segmentation processing on the target cutover information to obtain a plurality of word segmentation processing results; performing vector conversion on the plurality of word segmentation processing results to obtain a plurality of word segmentation vectors; determining the plurality of word segmentation vectors as the cutover data.

4. The method of claim 1, wherein, The method further comprises: determining a training data set, a validation data set and a convolutional neural network model of the target cutover type at a target cutover stage; training the convolutional neural network model using the training data set to obtain an initial cutover approval model; adjusting parameters of the initial cutover approval model using the validation data set to obtain a cutover approval model of the target cutover stage.

5. The method according to any one of claims 1 to 4, characterized in that, The method further comprises: obtaining cutover information of a preset first time period as cutover optimization information according to a preset period; optimizing the cutover approval model according to the cutover optimization information.

6. A data approval device characterized by comprising: The device comprises: a storage unit configured to obtain current cutover information, classify the current cutover information according to each cutover type contained in the current cutover information to obtain cutover information corresponding to each cutover type, determine cutover information corresponding to each cutover stage of the cutover type for cutover information corresponding to any cutover type, and store each cutover stage of the cutover type and the cutover information corresponding to each cutover stage correspondingly; a determination unit configured to determine a target network cutover according to a cutover time of each network cutover and a preset time period; an obtaining unit configured to obtain cutover information and a cutover approval model of a target cutover type at each cutover stage; the target cutover type is a cutover type corresponding to the target network cutover; the determination unit is further configured to determine an accuracy of the cutover information according to the cutover approval model; a sending unit configured to send alarm information if the accuracy of the cutover information meets a preset alarm condition.

7. The apparatus of claim 6, wherein, The determination unit is specifically configured to: vectorize the cutover information to obtain cutover data; The cutover data is input into the cutover approval model to obtain the accuracy of the cutover information.

8. The apparatus of claim 7, wherein, The determination unit is specifically configured to: extract target cutover information from the cutover information; perform word segmentation processing on the target cutover information to obtain a plurality of word segmentation processing results; perform vector conversion on the plurality of word segmentation processing results to obtain a plurality of word segmentation vectors; determine the plurality of word segmentation vectors as the cutover data.

9. A data approval device characterized by comprising: comprise: 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 realize the data approval method in any one of claims 1-5.

10. A computer-readable storage medium having stored therein instructions, the computer-readable storage medium comprising: When a computer executes the instruction, the computer executes the data approval method in any one of claims 1-5.

Images