Data generation method, device and equipment, readable storage medium and program product

Through the automated data generation method, the target configuration model and big data model are used to generate initial data matching the business data type, which solves the problems of poor data quality and low efficiency in traditional data generation methods, and achieves higher quality and efficient data generation .

CN118939531APending Publication Date: 2024-11-12CHINA CONSTRUCTION BANK +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410915960.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-09
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

Traditional data generation methods rely on testers to manually fill in configuration information and preset conditions, resulting in poor data quality and low efficiency, and cannot be closely matched with business logic.

Method used

By obtaining multiple field information, inputting it into the target configuration model, generating initial data and filtering target data based on the field configuration information, and using an automated process to generate initial data matching the business data type to improve data quality.

Benefits of technology

Improve the quality and efficiency of data generation, avoid the poor data quality problems caused by manual filling in configuration information and preset conditions, and the generated data is closer to the business logic.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118939531A_ABST
    Figure CN118939531A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of big data, and particularly discloses a data generation method and device, equipment, a readable storage medium and a program product. The method comprises the steps of obtaining multiple pieces of field information, inputting each piece of field information into a target configuration model, obtaining field configuration information corresponding to each piece of field information output by the target configuration model, generating initial data according to each business data type and a big data model, the target data corresponding to each piece of field information can be screened out from the initial data according to each piece of field configuration information and each piece of business data type, and each piece of field information corresponds to different business data types. By adopting the method, the data quality of the generated data can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to a data generation method, apparatus, device, readable storage medium and program product. Background Art

[0002] Data testing is a key step in ensuring data quality and the correctness of the data processing system. Data testing can verify the integrity, accuracy, consistency and validity of the data, ensure that the data is not missing or damaged, and conform to the expected data model and specification requirements.

[0003] Data generation provides the necessary input data for data testing. The quality of data generation methods often affects data testing. A good data generation method is important for improving the comprehensiveness and effectiveness of data testing. Traditional data generation methods usually require testers to manually fill in the configuration information and preset conditions required to generate data, and then generate data through random numbers.

[0004] Therefore, the data generated by traditional data generation methods has the problem of poor data quality. Summary of the invention

[0005] Based on this, it is necessary to provide a data generation method, device, equipment, readable storage medium and program product that can improve data quality in response to the above technical problems.

[0006] In a first aspect, the present application provides a data generation method, comprising:

[0007] Get multiple field information, each field information corresponds to a different business data type;

[0008] Input each field information into the target configuration model to obtain the field configuration information corresponding to each field information output by the target configuration model;

[0009] Generate initial data according to each business data type and the big data model, and filter out target data corresponding to each field information from the initial data according to each field configuration information and each business data type.

[0010] In one of the embodiments, the field configuration information includes at least one of repetition configuration information, numbering level configuration information, and grouping field configuration information.

[0011] In one embodiment, the method further comprises:

[0012] Acquire multiple sample field information, and preprocess each sample field information to obtain multiple sample character fragments;

[0013] The configuration information label of each sample character segment is obtained, and the initial configuration model is iteratively trained according to each sample character segment and the configuration information label of each sample character segment to obtain a target configuration model.

[0014] In one embodiment, each sample field information is preprocessed to obtain a plurality of sample character segments, including:

[0015] Perform word segmentation on the sample field information to obtain multiple candidate sample character fragments;

[0016] A plurality of sample character segments are screened out from each candidate sample character segment according to preset conditions.

[0017] In one embodiment, a plurality of sample character segments are selected from each candidate sample character segment according to a preset condition, including:

[0018] Calculating the compactness value corresponding to each candidate sample character segment, the compactness value is used to represent the compactness between each character included in the candidate sample character segment;

[0019] Calculate the correlation value corresponding to each candidate sample character segment, the correlation value is used to represent the correlation degree between the candidate sample character segment and other candidate sample character segments;

[0020] If the closeness value and the association value of the candidate sample character segment meet the preset conditions, the candidate sample character segment is used as the sample character segment.

[0021] In one embodiment, the initial configuration model is iteratively trained according to each sample character segment and the configuration information label of each sample character segment to obtain a target configuration model, including:

[0022] For one iterative training process, each sample character fragment is input into the intermediate configuration model to obtain multiple sample configuration information output by the intermediate configuration model, wherein in the first iterative training process, the intermediate configuration model is the initial configuration model, and in the non-first iterative training process, the intermediate configuration model is the configuration model obtained after the last iterative training ends;

[0023] Determine whether the intermediate configuration model has converged based on each sample configuration information and each configuration information label;

[0024] If the intermediate configuration model converges, the intermediate configuration model is used as the target configuration model.

[0025] In one embodiment, generating initial data according to each business data type and a big data model includes:

[0026] Input each business data type into the big data model to obtain the initial data corresponding to each business data type. The big data model is trained based on the business data of different business data types.

[0027] In one embodiment, according to each field configuration information and each business data type, target data corresponding to each field information is screened out from the initial data, including:

[0028] For a field information, determine candidate data corresponding to the field information from the initial data according to the business data type corresponding to the field information;

[0029] The candidate data is screened according to the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

[0030] In a second aspect, the present application provides a data generating device, comprising:

[0031] The acquisition module is used to obtain multiple field information, each field information corresponds to a different business data type;

[0032] A configuration module, used to input each field information into the target configuration model, and obtain field configuration information corresponding to each field information output by the target configuration model;

[0033] The generation module is used to generate initial data according to various business data types and the big data model, and to filter out target data corresponding to each field information from the initial data according to each field configuration information and each business data type.

[0034] In one of the embodiments, the field configuration information includes at least one of repetition configuration information, numbering level configuration information, and grouping field configuration information.

[0035] In one embodiment, the device further comprises:

[0036] A preprocessing module is used to obtain multiple sample field information and preprocess each sample field information to obtain multiple sample character fragments;

[0037] The training module is used to obtain the configuration information label of each sample character segment, and iteratively train the initial configuration model according to each sample character segment and the configuration information label of each sample character segment to obtain a target configuration model.

[0038] In one embodiment, the preprocessing module includes:

[0039] A word segmentation unit is used to perform word segmentation processing on the sample field information to obtain multiple candidate sample character fragments;

[0040] The sample screening unit is used to screen out a plurality of sample character segments from each candidate sample character segment according to preset conditions.

[0041] In one embodiment, the sample screening unit is specifically used for:

[0042] Calculating the compactness value corresponding to each candidate sample character segment, the compactness value is used to represent the compactness between each character included in the candidate sample character segment;

[0043] Calculate the correlation value corresponding to each candidate sample character segment, the correlation value is used to represent the correlation degree between the candidate sample character segment and other candidate sample character segments;

[0044] If the closeness value and the association value of the candidate sample character segment meet the preset conditions, the candidate sample character segment is used as the sample character segment.

[0045] In one embodiment, the training module includes:

[0046] An input unit, for inputting each sample character segment into an intermediate configuration model for an iterative training process, and obtaining a plurality of sample configuration information output by the intermediate configuration model, wherein in the first iterative training process, the intermediate configuration model is an initial configuration model, and in the non-first iterative training process, the intermediate configuration model is a configuration model obtained after the last iterative training ends;

[0047] A judging unit, used to judge whether the intermediate configuration model has converged according to each sample configuration information and each configuration information label;

[0048] The convergence unit is used to use the intermediate configuration model as the target configuration model if the intermediate configuration model converges.

[0049] In one embodiment, the generating module includes:

[0050] The initial generation unit is used to input various business data types into the big data model to obtain initial data corresponding to each business data type. The big data model is trained based on business data of different business data types.

[0051] In one embodiment, the generating module further comprises:

[0052] A candidate determination unit, for determining, for a field information, candidate data corresponding to the field information from the initial data according to the business data type corresponding to the field information;

[0053] The target screening unit is used to screen the candidate data according to the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

[0054] In a third aspect, an embodiment of the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method of the first aspect when executing the computer program.

[0055] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of the first aspect described above.

[0056] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the method in the first aspect are implemented.

[0057] The above-mentioned data generation method, device, equipment, readable storage medium and program product can obtain the field configuration information corresponding to each field information output by the target configuration model by acquiring multiple field information and inputting each field information into the target configuration model, generate initial data according to each business data type and the big data model, and according to each field configuration information and each business data type, can filter out the target data corresponding to each field information from the initial data, wherein each field information corresponds to a different business data type. In this way, according to the business data type and the big data model, the initial data that fits the business data type and matches the field information is automatically generated, and then the initial data is screened once using the configuration information corresponding to each field information automatically generated according to the field information to obtain the target data that meets the requirements, thereby avoiding the problem of poor data quality in the traditional technology that randomly generates data according to the configuration information and preset conditions required for the tester to manually fill in the data generation. The data generated by the data generation method provided by the technical solution of the present application has higher data quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 An application environment diagram of a data generation method in an embodiment;

[0060] Figure 2 A schematic diagram of a flow chart of a data generation method in an embodiment;

[0061] Figure 3 A schematic diagram of a process for obtaining a target configuration model in another embodiment;

[0062] Figure 4 A schematic diagram of a flow chart of preprocessing each sample field information to obtain multiple sample character segments in another embodiment;

[0063] Figure 5 A flowchart of an exemplary data generation method in one embodiment;

[0064] Figure 6 is a structural block diagram of a data generating device in one embodiment;

[0065] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. It should be noted that in the embodiments of the present application, certain software, groups, models and other existing solutions in the industry may be mentioned, which should be considered as exemplary, and their purpose is only to illustrate the feasibility of the implementation of the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0067] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations. The acquisition, storage, use, and processing of data in the technical solution of this application are in compliance with relevant provisions of national laws and regulations.

[0068] Data testing is a key step in ensuring data quality and the correctness of the data processing system. Data testing can verify the integrity, accuracy, consistency and validity of the data, ensure that the data is not missing or damaged, and conform to the expected data model and specification requirements.

[0069] Data generation provides the necessary input data for data testing. The quality of data generation methods often affects data testing. A good data generation method is important for improving the comprehensiveness and effectiveness of data testing. Traditional data generation methods usually require testers to manually fill in the attributes and types of data fields required to generate data, and then generate data through random numbers.

[0070] Existing data generation methods require users to spend a lot of time manually filling in a series of configuration information and presets, which not only wastes human resources, but also has problems such as low accuracy and poor efficiency. The test data finally generated has a large degree of randomness and cannot be closely aligned with the business logic, resulting in poor data quality.

[0071] In view of this, the present application provides a data generation method, device, equipment, readable storage medium and program product, by obtaining multiple field information, and inputting each field information into the target configuration model, the field configuration information corresponding to each field information output by the target configuration model can be obtained, and initial data can be generated according to each business data type and the big data model, and according to each field configuration information and each business data type, the target data corresponding to each field information can be screened out from the initial data, wherein each field information corresponds to a different business data type. In this way, according to the business data type and the big data model, the initial data that fits the business data type and matches the field information is automatically generated, and then the initial data is screened once using the configuration information corresponding to each field information automatically generated according to the field information to obtain the target data that meets the requirements, avoiding the problem of poor data quality in the traditional technology that is randomly generated according to the configuration information and preset conditions required for the tester to manually fill in the data generation. The data generated by the data generation method provided by the technical solution of the present application has higher data quality.

[0072] The data generation method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the data storage system can store the data that the server 101 needs to process. The data storage system can be integrated on the server 101, or it can be placed on the cloud or other network servers. The server 101 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0073] In an exemplary embodiment, Figure 2 As shown, a data generation method is provided, which is applied to Figure 1 The server 101 in the example is used for explanation. It is understandable that the method can also be applied to a terminal, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. The following steps 201 to 203 are included. Among them:

[0074] Step 201, obtaining multiple field information.

[0075] A field can be a data item used to generate data, such as name, age, institution number, etc. Different fields correspond to different field information, and the field information may include field type, field length, field name, and remarks, etc.

[0076] In an embodiment of the present application, the server can obtain multiple field information through different methods. Optionally, the server can obtain field information input by the user through an external device; optionally, the server can obtain field information stored at a preset storage address through the address; optionally, the server can obtain table data of the target database and perform automated analysis, for example, inputting the table data into a pre-trained field analysis model to obtain multiple field information output by the field analysis model.

[0077] Each field of information corresponds to a different business data type. The business data type is linked to the business data, such as organization number, personnel number, business date, etc. The server can classify each field of information into different business data types. For example, the first organization number and the second organization number are both organization numbers, and the ID number and personnel code are both personnel numbers. Different business data types correspond to different business data.

[0078] Optionally, the server can obtain the mapping relationship input by the user through an external input device, and the mapping relationship can be used to characterize the relationship between field information and business data type. The server can determine the business data type corresponding to each field information based on each mapping relationship; optionally, the server can determine the business data type corresponding to each field information based on different mapping rules; optionally, the server can input each field information into a pre-trained mapping model to obtain the business data type corresponding to each field information output by the mapping model.

[0079] Step 202: input each field information into the target configuration model to obtain the field configuration information corresponding to each field information output by the target configuration model.

[0080] The field configuration information can be set according to needs, including at least one of repeated configuration information, numbering level configuration information, and grouped field configuration information. The repeated configuration information is used to indicate whether repeated data can exist under the field corresponding to the field information. The numbering level configuration information is used to indicate the numbering level to which the data under the field corresponding to the field information should correspond. The grouped field configuration information is used to indicate whether there is a field associated with the field corresponding to the field information.

[0081] In a possible implementation, the field configuration information may also include value range information, custom logic information, and the like.

[0082] The server can input each field information into the target configuration model, thereby automatically obtaining the field configuration information corresponding to each field information output by the target configuration model. The target configuration model is used to automatically configure the field configuration information for each field information.

[0083] Step 203, generating initial data according to each business data type and the big data model, and filtering out target data corresponding to each field information from the initial data according to each field configuration information and each business data type.

[0084] The big data model can be a model trained in advance based on business data. The server inputs the business data type into the big data model, and the big data model can randomly generate business data associated with the business data type based on the business data type. In an embodiment of the present application, the server can input each business data type into the big data model to obtain initial data corresponding to each business data type. The big data model is trained based on business data of different business data types, and the business data is unscreened business data.

[0085] After obtaining a plurality of initial data, in order to improve data quality, the server may also perform a round of screening on the initial data according to configuration information of each field and types of each business data to obtain target data that meets the requirements.

[0086] In an embodiment of the present application, the process of server screening to obtain target data may include the following steps: for a field information, the server may determine the candidate data corresponding to the field information from the initial data based on the business data type corresponding to the field information, and then the server may screen the candidate data based on the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

[0087] Exemplarily, the field is the first organization number, and its corresponding business data type is the organization number. The server can use the data with the business data type of the organization number in the initial data as candidate data. Then, the server can obtain the field configuration information corresponding to the first organization number, such as not allowing duplication, data length, etc. The server can filter the candidate data according to the field configuration information, and select the candidate data that meets the field configuration information as the target data.

[0088] In the above embodiment, initial data that fits the business data type and matches the field information is automatically generated according to the business data type and the big data model, and then the initial data is screened once using the configuration information corresponding to each field information automatically generated according to the field information to obtain target data that meets the requirements. This avoids the problem of poor data quality in traditional technologies where data is randomly generated based on the configuration information and preset conditions required for data generation manually filled in by testers. The data generated by the data generation method provided by the technical solution of the present application has higher data quality.

[0089] In one embodiment, based on the above Figure 2 The embodiment shown, see Figure 3 , this embodiment involves the process of obtaining the target configuration model. Figure 3 As shown, the process may include step 301 and step 302 .

[0090] Step 301: Acquire multiple sample field information, and pre-process each sample field information to obtain multiple sample character segments.

[0091] The sample field information may be data used to train the configuration model to obtain the target configuration model. Optionally, the server may obtain the sample field information input by the user through an external input device; optionally, the server may obtain the sample field information based on data table analysis in a database.

[0092] After obtaining the sample field information, the server can preprocess the sample field information to facilitate subsequent model training and improve the efficiency of model training. The preprocessing can include word segmentation, screening according to preset conditions, standardization, missing value processing, etc., so as to obtain multiple sample character fragments.

[0093] Step 302, obtaining the configuration information label of each sample character segment, and iteratively training the initial configuration model according to each sample character segment and the configuration information label of each sample character segment to obtain a target configuration model.

[0094] After obtaining multiple sample character fragments, the server needs to obtain the configuration information tag of each sample character fragment. The configuration information tag is the actual field configuration information corresponding to each sample character fragment. The configuration information tag may include at least one of repeated configuration information, numbering level configuration information, and grouped field configuration information.

[0095] Optionally, the server can obtain the configuration information tags corresponding to each sample character fragment input by the user through an external input device; optionally, the server can obtain the configuration information file stored at a preset storage address through the address, and the configuration information file contains the configuration information tags corresponding to each sample character fragment; optionally, the server can add a configuration information tag to each sample character fragment according to preset conditions.

[0096] Exemplarily, for repeated configuration information, the server can classify and mark it. The server can vector process each candidate sample character fragment to obtain multiple word vectors, and then input each word vector into a pre-trained classification model. The output result of the classification model is the repeated configuration information corresponding to each field information. The information may include 0 and 1, 0 represents that the data under the field information can be repeated, and 1 represents that the data under the field information cannot be repeated; for numbering level configuration information, the server can also classify and mark each field information through a pre-trained classification model, and different numbers correspond to different levels.

[0097] Regarding the grouped field configuration information, in an embodiment of the present application, the server can calculate the cosine similarity between the characters included in each candidate sample character fragment, and determine whether the grouped field configuration information of each candidate sample character fragment is grouped or not based on the cosine similarity corresponding to each candidate sample character fragment.

[0098] Regarding the calculation process of cosine similarity, for a candidate sample character segment, the server can first vectorize each character included in the candidate sample character segment to obtain the word vector corresponding to each character. Taking the character length of each candidate sample character segment as 2 as an example, the server can calculate the cosine similarity of each candidate sample character segment according to the following formula:

[0099] (1)

[0100] A and B are two word vectors, and cos(θ) is the cosine similarity.

[0101] In an embodiment of the present application, the server can set the grouping configuration information of the candidate sample character segments whose cosine similarity is greater than or equal to the preset similarity threshold as grouping, and set the grouping configuration information of the candidate sample character segments whose cosine similarity is less than the preset similarity threshold as not grouping.

[0102] After obtaining the configuration information label of each sample character fragment, the server can iteratively train the initial configuration model according to each sample character fragment and the configuration information label of each sample character fragment to obtain a target configuration model. The basic model of the initial configuration model can be a neural network model such as a pre-trained model. The server can obtain the target configuration model by iteratively training the initial configuration model.

[0103] Regarding the process of the server training the initial configuration model to obtain the target configuration model, in a possible implementation, the following steps may be included: for an iterative training process, the server inputs each sample character fragment into the intermediate configuration model to obtain multiple sample configuration information output by the intermediate configuration model; determines whether the intermediate configuration model converges based on each sample configuration information and each configuration information label; if the intermediate configuration model converges, the intermediate configuration model is used as the target configuration model.

[0104] Among them, in the first iterative training process, the intermediate configuration model is the initial configuration model, and in the non-first iterative training process, the intermediate configuration model is the configuration model obtained after the last iterative training is completed.

[0105] Regarding the process of the server judging whether the intermediate model has converged, optionally, the server may match each sample configuration information with each configuration information label to obtain the degree of match between each sample configuration information and each configuration information label, and the server may compare the degree of match with the degree of match threshold, and if the degree of match reaches the degree of match threshold, the server may determine that the intermediate configuration model of the current iteration process has converged, otherwise it has not converged; optionally, the server may count the number of iterative trainings through a counter, and when the number of iterative trainings reaches a preset number threshold, the server may determine that the intermediate configuration model of the current iteration process has converged, otherwise it has not converged.

[0106] In the above embodiment, a target configuration model is automatically trained to set field configuration information for each field information, thereby avoiding the problem of low configuration efficiency in the traditional technology of manually filling in the configuration information and preset conditions required to generate data.

[0107] In one embodiment, based on the above Figure 3 The embodiment shown, see Figure 4 This embodiment involves a process of preprocessing each sample field information to obtain multiple sample character segments. Figure 4 As shown, the process may include step 401 and step 402 .

[0108] Step 401, perform word segmentation processing on sample field information to obtain multiple candidate sample character segments.

[0109] Step 402, screening out a plurality of sample character segments from each candidate sample character segment according to a preset condition.

[0110] In an embodiment of the present application, the sample field information may include data such as names and notes that characterize the data characteristics generated by the sample field information, such as "region name", "city number", etc. The server may perform word segmentation on these data in the sample field information. Optionally, the server may perform word segmentation on these data according to a preset character length, thereby obtaining multiple candidate sample character segments of a preset character length. For example, the preset character length may be 2, and the word segmentation of "region name" may be performed to obtain candidate sample character segments such as "region", "name", and "district name".

[0111] After obtaining multiple candidate sample character fragments, since these candidate sample character fragments may include some meaningless candidate sample character fragments, the server needs to screen these candidate sample character fragments to reduce the computational complexity of the iterative training process and further reduce the waste of computing resources.

[0112] In a possible implementation, the server may input each candidate sample character segment into a pre-trained recognition model, thereby obtaining a sample character segment that meets the requirements and is output by the recognition model.

[0113] In another possible implementation, the server may screen candidate sample character segments according to a plurality of preset conditions to obtain sample character segments output by the recognition model that meet the preset conditions.

[0114] Based on this, in one embodiment, the process of screening and obtaining multiple sample character fragments may include the following steps: calculating the closeness value corresponding to each candidate sample character fragment; calculating the association value corresponding to each candidate sample character fragment; if the closeness value and the association value of the candidate sample character fragment meet preset conditions, the candidate sample character fragment is used as the sample character fragment.

[0115] The compactness value can be used to characterize the compactness between the characters included in the candidate sample character segment. Taking the character length corresponding to the candidate sample character segment as 2 as an example, the server can obtain the compactness value corresponding to each candidate sample character segment by calculating formula (2).

[0116] (2)

[0117] Where x and y represent the characters in the candidate sample character segments respectively, P(x, y) is the probability that the two characters appear simultaneously in all candidate sample character segments, P(x) is the probability that the x character appears, and P(y) is the probability that the y character appears.

[0118] The association value can be used to represent the degree of association between the candidate sample character segment and other candidate sample character segments. Taking the character length of the candidate sample character segment as 2 as an example, the server can obtain the association value corresponding to each candidate sample character segment by calculating formula (3).

[0119] (3)

[0120] Among them, x is any candidate sample character fragment, yn represents the candidate sample character fragment of its left or right neighbor, and Y represents the set of candidate sample character fragments of its left or right neighbor. For example, the unsegmented field information is "other city number" and "city name", and after word segmentation, the candidate sample character fragments "other", "city", "number", "name" and so on can be obtained. Then, the candidate sample character fragments yn of the left or right neighbor of the candidate sample character fragment "city" are "other", "number", and "name", and the set Y of the candidate sample character fragments of its left or right neighbor is {"other", "number", "name"}.

[0121] After calculating the closeness value and the association value corresponding to each candidate sample character fragment, the server can select each candidate sample character fragment according to the preset conditions and the closeness value and the association value of each candidate sample character fragment. The server can obtain the closeness threshold and the association threshold. The preset conditions are that the closeness value is greater than the closeness threshold and the association value is greater than the association threshold. If the closeness value and the association value of the selected sample character fragment meet the preset conditions, the server can use the candidate sample character fragment as the sample character fragment.

[0122] In the above embodiment, by performing word segmentation processing on each candidate sample character segment and screening to obtain sample character segments for subsequent iterative training, the sample size of subsequent iterative training is reduced and the training efficiency is improved.

[0123] In one embodiment, referring to Figure 5 , provides an exemplary data generation method that can be applied to Figure 1 In the implementation environment shown.

[0124] Step 501: perform word segmentation processing on sample field information to obtain multiple candidate sample character segments.

[0125] Step 502, calculating the compactness value corresponding to each candidate sample character segment.

[0126] The closeness value is used to represent the closeness between the characters included in the candidate sample character segment.

[0127] Step 503, calculating the association value corresponding to each candidate sample character segment.

[0128] The correlation value is used to represent the correlation degree between the candidate sample character segment and other candidate sample character segments.

[0129] Step 504: If the closeness value and the association value of the candidate sample character segment meet the preset conditions, the candidate sample character segment is used as the sample character segment.

[0130] Step 505: Obtain the configuration information tag of each sample character segment.

[0131] Step 506, for one iterative training process, each sample character segment is input into the intermediate configuration model to obtain a plurality of sample configuration information output by the intermediate configuration model.

[0132] Among them, in the first iterative training process, the intermediate configuration model is the initial configuration model, and in the non-first iterative training process, the intermediate configuration model is the configuration model obtained after the last iterative training is completed.

[0133] Step 507: determine whether the intermediate configuration model has converged according to each sample configuration information and each configuration information label.

[0134] Step 508: If the intermediate configuration model converges, the intermediate configuration model is used as the target configuration model.

[0135] Step 509, obtaining multiple field information.

[0136] Among them, each field information corresponds to a different business data type.

[0137] Step 510: Input each field information into the target configuration model to obtain the field configuration information corresponding to each field information output by the target configuration model.

[0138] Step 511, input each business data type into the big data model to obtain initial data corresponding to each business data type.

[0139] Among them, the big data model is trained based on business data of different business data types.

[0140] Step 512: for a field information, determine candidate data corresponding to the field information from the initial data according to the business data type corresponding to the field information.

[0141] Step 513: Screen the candidate data according to the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

[0142] The field configuration information includes at least one of repetition configuration information, numbering level configuration information, and grouping field configuration information.

[0143] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0144] Based on the same inventive concept, the embodiment of the present application also provides a data generation device for implementing the data generation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the one or more data generation device embodiments provided below can refer to the limitations on the data generation method above, and will not be repeated here.

[0145] In an exemplary embodiment, Figure 6 As shown, a data generation device is provided, including: an acquisition module 601, a configuration module 602 and a generation module 603, wherein:

[0146] The acquisition module 601 is used to acquire multiple field information, each field information corresponds to a different business data type;

[0147] Configuration module 602, used to input each field information into the target configuration model, and obtain field configuration information corresponding to each field information output by the target configuration model;

[0148] The generation module 603 is used to generate initial data according to each business data type and the big data model, and filter out target data corresponding to each field information from the initial data according to each field configuration information and each business data type.

[0149] In one embodiment, the field configuration information includes at least one of repetition configuration information, numbering level configuration information, and grouping field configuration information.

[0150] In one embodiment, the apparatus further comprises:

[0151] A preprocessing module is used to obtain multiple sample field information and preprocess each sample field information to obtain multiple sample character fragments;

[0152] The training module is used to obtain the configuration information label of each sample character segment, and iteratively train the initial configuration model according to each sample character segment and the configuration information label of each sample character segment to obtain a target configuration model.

[0153] In one embodiment, the pre-processing module includes:

[0154] A word segmentation unit is used to perform word segmentation processing on the sample field information to obtain multiple candidate sample character fragments;

[0155] The sample screening unit is used to screen out a plurality of sample character segments from each candidate sample character segment according to preset conditions.

[0156] In one embodiment, the sample screening unit is specifically used for:

[0157] Calculating the compactness value corresponding to each candidate sample character segment, the compactness value is used to represent the compactness between each character included in the candidate sample character segment;

[0158] Calculate the correlation value corresponding to each candidate sample character segment, the correlation value is used to represent the correlation degree between the candidate sample character segment and other candidate sample character segments;

[0159] If the closeness value and the association value of the candidate sample character segment meet the preset conditions, the candidate sample character segment is used as the sample character segment.

[0160] In one embodiment, the training module includes:

[0161] An input unit, for inputting each sample character segment into an intermediate configuration model for an iterative training process, and obtaining a plurality of sample configuration information output by the intermediate configuration model, wherein in the first iterative training process, the intermediate configuration model is an initial configuration model, and in the non-first iterative training process, the intermediate configuration model is a configuration model obtained after the last iterative training ends;

[0162] A judging unit, used to judge whether the intermediate configuration model has converged according to each sample configuration information and each configuration information label;

[0163] The convergence unit is used to use the intermediate configuration model as the target configuration model if the intermediate configuration model converges.

[0164] In one embodiment, the generation module 603 includes:

[0165] The initial generation unit is used to input various business data types into the big data model to obtain initial data corresponding to each business data type. The big data model is trained based on business data of different business data types.

[0166] In one embodiment, the generating module 603 further includes:

[0167] A candidate determination unit, for determining, for a field information, candidate data corresponding to the field information from the initial data according to the business data type corresponding to the field information;

[0168] The target screening unit is used to screen the candidate data according to the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

[0169] Each module in the above data generator device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module above.

[0170] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 7 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data generation party data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a data generation party method is implemented.

[0171] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0172] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0173] Acquire multiple field information, each field information corresponding to a different business data type;

[0174] Input each of the field information into the target configuration model to obtain the field configuration information corresponding to each of the field information output by the target configuration model;

[0175] Initial data is generated according to each of the business data types and the big data model, and target data corresponding to each of the field information is screened out from the initial data according to each of the field configuration information and each of the business data types.

[0176] In one embodiment, the field configuration information includes at least one of repetition configuration information, numbering level configuration information, and grouping field configuration information.

[0177] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0178] Acquire multiple sample field information, and preprocess each sample field information to obtain multiple sample character fragments;

[0179] The configuration information label of each of the sample character segments is obtained, and the initial configuration model is iteratively trained according to each of the sample character segments and the configuration information label of each of the sample character segments to obtain a target configuration model.

[0180] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0181] Performing word segmentation processing on the sample field information to obtain multiple candidate sample character fragments;

[0182] The plurality of sample character segments are screened out from each of the candidate sample character segments according to a preset condition.

[0183] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0184] Calculating a compactness value corresponding to each of the candidate sample character segments, wherein the compactness value is used to characterize the compactness between the characters included in the candidate sample character segment;

[0185] Calculating a correlation value corresponding to each of the candidate sample character segments, wherein the correlation value is used to represent the degree of correlation between the candidate sample character segment and other candidate sample character segments;

[0186] If the closeness value and the association value of the candidate sample character segment meet the preset conditions, the candidate sample character segment is used as the sample character segment.

[0187] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0188] For one iterative training process, each of the sample character segments is input into the intermediate configuration model to obtain a plurality of sample configuration information output by the intermediate configuration model, wherein in the first iterative training process, the intermediate configuration model is the initial configuration model, and in the non-first iterative training process, the intermediate configuration model is the configuration model obtained after the last iterative training ends;

[0189] Determining whether the intermediate configuration model converges according to each of the sample configuration information and each of the configuration information labels;

[0190] If the intermediate configuration model converges, the intermediate configuration model is used as the target configuration model.

[0191] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0192] Each of the business data types is input into the big data model to obtain the initial data corresponding to each of the business data types. The big data model is trained based on business data of different business data types.

[0193] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0194] For one of the field information, determining candidate data corresponding to the field information from the initial data according to the business data type corresponding to the field information;

[0195] The candidate data is screened according to the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

[0196] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0197] Acquire multiple field information, each field information corresponding to a different business data type;

[0198] Input each of the field information into the target configuration model to obtain the field configuration information corresponding to each of the field information output by the target configuration model;

[0199] Initial data is generated according to each of the business data types and the big data model, and target data corresponding to each of the field information is screened out from the initial data according to each of the field configuration information and each of the business data types.

[0200] In one embodiment, the field configuration information includes at least one of repetition configuration information, numbering level configuration information, and grouping field configuration information.

[0201] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0202] Acquire multiple sample field information, and preprocess each sample field information to obtain multiple sample character fragments;

[0203] The configuration information label of each of the sample character segments is obtained, and the initial configuration model is iteratively trained according to each of the sample character segments and the configuration information label of each of the sample character segments to obtain a target configuration model.

[0204] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0205] Performing word segmentation processing on the sample field information to obtain multiple candidate sample character fragments;

[0206] The plurality of sample character segments are screened out from each of the candidate sample character segments according to a preset condition.

[0207] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0208] Calculating a compactness value corresponding to each of the candidate sample character segments, wherein the compactness value is used to characterize the compactness between the characters included in the candidate sample character segment;

[0209] Calculating a correlation value corresponding to each of the candidate sample character segments, wherein the correlation value is used to represent the degree of correlation between the candidate sample character segment and other candidate sample character segments;

[0210] If the closeness value and the association value of the candidate sample character segment meet the preset conditions, the candidate sample character segment is used as the sample character segment.

[0211] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0212] For one iterative training process, each of the sample character segments is input into the intermediate configuration model to obtain a plurality of sample configuration information output by the intermediate configuration model, wherein in the first iterative training process, the intermediate configuration model is the initial configuration model, and in the non-first iterative training process, the intermediate configuration model is the configuration model obtained after the last iterative training ends;

[0213] Determining whether the intermediate configuration model converges according to each of the sample configuration information and each of the configuration information labels;

[0214] If the intermediate configuration model converges, the intermediate configuration model is used as the target configuration model.

[0215] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0216] Each of the business data types is input into the big data model to obtain the initial data corresponding to each of the business data types. The big data model is trained based on business data of different business data types.

[0217] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0218] For one of the field information, determining candidate data corresponding to the field information from the initial data according to the business data type corresponding to the field information;

[0219] The candidate data is screened according to the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

[0220] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0221] Acquire multiple field information, each field information corresponding to a different business data type;

[0222] Input each of the field information into the target configuration model to obtain the field configuration information corresponding to each of the field information output by the target configuration model;

[0223] Initial data is generated according to each of the business data types and the big data model, and target data corresponding to each of the field information is screened out from the initial data according to each of the field configuration information and each of the business data types.

[0224] In one embodiment, the field configuration information includes at least one of repetition configuration information, numbering level configuration information, and grouping field configuration information.

[0225] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0226] Acquire multiple sample field information, and preprocess each sample field information to obtain multiple sample character fragments;

[0227] The configuration information label of each of the sample character segments is obtained, and the initial configuration model is iteratively trained according to each of the sample character segments and the configuration information label of each of the sample character segments to obtain a target configuration model.

[0228] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0229] Performing word segmentation processing on the sample field information to obtain multiple candidate sample character fragments;

[0230] The plurality of sample character segments are screened out from each of the candidate sample character segments according to a preset condition.

[0231] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0232] Calculating a compactness value corresponding to each of the candidate sample character segments, wherein the compactness value is used to characterize the compactness between the characters included in the candidate sample character segment;

[0233] Calculating a correlation value corresponding to each of the candidate sample character segments, wherein the correlation value is used to represent the degree of correlation between the candidate sample character segment and other candidate sample character segments;

[0234] If the closeness value and the association value of the candidate sample character segment meet the preset conditions, the candidate sample character segment is used as the sample character segment.

[0235] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0236] For one iterative training process, each of the sample character segments is input into the intermediate configuration model to obtain a plurality of sample configuration information output by the intermediate configuration model, wherein in the first iterative training process, the intermediate configuration model is the initial configuration model, and in the non-first iterative training process, the intermediate configuration model is the configuration model obtained after the last iterative training ends;

[0237] Determining whether the intermediate configuration model converges according to each of the sample configuration information and each of the configuration information labels;

[0238] If the intermediate configuration model converges, the intermediate configuration model is used as the target configuration model.

[0239] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0240] Each of the business data types is input into the big data model to obtain the initial data corresponding to each of the business data types. The big data model is trained based on business data of different business data types.

[0241] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0242] For one of the field information, determining candidate data corresponding to the field information from the initial data according to the business data type corresponding to the field information;

[0243] The candidate data is screened according to the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

[0244] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0245] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0246] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A data generation method, characterized in that: The method comprises: Acquire multiple field information, each field information corresponding to a different business data type; Input each of the field information into the target configuration model to obtain the field configuration information corresponding to each of the field information output by the target configuration model; Initial data is generated according to each of the business data types and the big data model, and target data corresponding to each of the field information is screened out from the initial data according to each of the field configuration information and each of the business data types.

2. The method according to claim 1, characterized in that The field configuration information includes at least one of repetition configuration information, numbering level configuration information, and grouping field configuration information.

3. The method according to claim 1, characterized in that The method further comprises: Acquire multiple sample field information, and preprocess each sample field information to obtain multiple sample character fragments; The configuration information label of each of the sample character segments is obtained, and the initial configuration model is iteratively trained according to each of the sample character segments and the configuration information label of each of the sample character segments to obtain a target configuration model.

4. The method according to claim 3, characterized in that The preprocessing of each sample field information to obtain a plurality of sample character segments includes: Performing word segmentation processing on the sample field information to obtain multiple candidate sample character fragments; The plurality of sample character segments are screened out from each of the candidate sample character segments according to a preset condition.

5. The method according to claim 4, characterized in that The step of selecting the plurality of sample character segments from each of the candidate sample character segments according to a preset condition comprises: Calculating a compactness value corresponding to each of the candidate sample character segments, wherein the compactness value is used to characterize the compactness between the characters included in the candidate sample character segment; Calculating a correlation value corresponding to each of the candidate sample character segments, wherein the correlation value is used to represent the degree of correlation between the candidate sample character segment and other candidate sample character segments; If the closeness value and the association value of the candidate sample character segment meet preset conditions, the candidate sample character segment is used as the sample character segment.

6. The method according to claim 3, characterized in that The iterative training of the initial configuration model according to each of the sample character segments and the configuration information label of each of the sample character segments to obtain the target configuration model includes: For one iterative training process, each of the sample character segments is input into the intermediate configuration model to obtain a plurality of sample configuration information output by the intermediate configuration model, wherein in the first iterative training process, the intermediate configuration model is the initial configuration model, and in the non-first iterative training process, the intermediate configuration model is the configuration model obtained after the last iterative training ends; Determining whether the intermediate configuration model converges according to each of the sample configuration information and each of the configuration information labels; If the intermediate configuration model converges, the intermediate configuration model is used as the target configuration model.

7. The method according to claim 1, characterized in that The generating of initial data according to each of the business data types and the big data model includes: Each of the business data types is input into the big data model to obtain the initial data corresponding to each of the business data types. The big data model is trained based on business data of different business data types.

8. The method according to claim 1, characterized in that The step of filtering out target data corresponding to each field information from the initial data according to each field configuration information and each business data type includes: For one of the field information, determining candidate data corresponding to the field information from the initial data according to the business data type corresponding to the field information; The candidate data is screened according to the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

9. A data generating device, characterized in that: The device comprises: An acquisition module, used to acquire multiple field information, each field information corresponding to a different business data type; A configuration module, used to input each of the field information into a target configuration model, and obtain field configuration information corresponding to each of the field information output by the target configuration model; A generation module is used to generate initial data according to each of the business data types and the big data model, and to filter out target data corresponding to each of the field information from the initial data according to each of the field configuration information and each of the business data types.

10. The device according to claim 9, characterized in that The field configuration information includes at least one of repetition configuration information, numbering level configuration information, and grouping field configuration information.

11. The device according to claim 9, characterized in that The device also includes: A preprocessing module, used for acquiring a plurality of sample field information, and preprocessing each of the sample field information to obtain a plurality of sample character fragments; The training module is used to obtain the configuration information label of each of the sample character segments, and iteratively train the initial configuration model according to each of the sample character segments and the configuration information label of each of the sample character segments to obtain a target configuration model.

12. The device according to claim 11, characterized in that The preprocessing module comprises: A word segmentation unit, used to perform word segmentation processing on the sample field information to obtain a plurality of candidate sample character fragments; The sample screening unit is used to screen out the multiple sample character segments from each of the candidate sample character segments according to preset conditions.

13. The device according to claim 12, characterized in that The sample screening unit is specifically used for: Calculating a compactness value corresponding to each of the candidate sample character segments, wherein the compactness value is used to characterize the compactness between the characters included in the candidate sample character segment; Calculating a correlation value corresponding to each of the candidate sample character segments, wherein the correlation value is used to represent the degree of correlation between the candidate sample character segment and other candidate sample character segments; If the closeness value and the association value of the candidate sample character segment meet preset conditions, the candidate sample character segment is used as the sample character segment.

14. The device according to claim 11, characterized in that The training module includes: An input unit, for inputting each of the sample character segments into an intermediate configuration model for an iterative training process, to obtain a plurality of sample configuration information output by the intermediate configuration model, wherein in the first iterative training process, the intermediate configuration model is the initial configuration model, and in the non-first iterative training process, the intermediate configuration model is the configuration model obtained after the last iterative training ends; A judging unit, configured to judge whether the intermediate configuration model has converged according to each of the sample configuration information and each of the configuration information labels; A convergence unit is used to use the intermediate configuration model as the target configuration model if the intermediate configuration model converges.

15. The device according to claim 9, characterized in that The generation module comprises: The initial generation unit is used to input each of the business data types into the big data model to obtain the initial data corresponding to each of the business data types. The big data model is trained based on the business data of different business data types.

16. The device according to claim 9, characterized in that The generation module also includes: a candidate determination unit, configured to determine, for one of the field information, candidate data corresponding to the field information from the initial data according to the type of business data corresponding to the field information; A target screening unit is used to screen the candidate data according to the field configuration information corresponding to the field information to obtain the target data corresponding to the field information.

17. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

18. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

19. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.