Data generation method and apparatus, device, and storage medium

Abstract

The application provides a data generation method and device, equipment and a storage medium, and relates to the field of big data. The method comprises the following steps: obtaining identification information of a first data table; taking the identification information as a keyword, searching for a field relationship list corresponding to the first data table from a plurality of field relationship lists corresponding to a plurality of data tables, the field relationship list comprising a field in the first data table and an associated field corresponding to the field in the first data table, and the field relationship list being obtained by performing field similarity and association analysis on the plurality of data tables; based on the field relationship list corresponding to the first data table, constructing a data table association relationship graph between the first data table and a second data table, the second data table being a data table to which the associated field belongs; and based on the data table association relationship graph, assigning values to the field in the first data table and the field in the second data table. Thus, based on the association relationship between the fields of the data tables, the efficiency and accuracy of assigning values to the data tables are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of big data, and in particular to a data generation method and device, equipment and a storage medium. BACKGROUND

[0002] With the increase of data, batch verification of big data needs to be performed to improve the accuracy of massive data.

[0003] In the process of batch verification of big data, the data correlation between data tables cannot be checked and analyzed by the tester, and the tester needs to rely on the logical document provided by the development designer to prepare data. The logical document records the data correlation between data tables.

[0004] However, the logical document may be missing data tables and data correlation, resulting in a large amount of work and low efficiency of the tester preparing data in data testing. SUMMARY

[0005] The present application provides a data generation method, device, equipment and storage medium to solve the problem of large amount of work and low efficiency of data preparation.

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

[0007] obtaining identification information of a first data table to be generated;

[0008] Taking the identification information as a key, a field relationship list corresponding to the first data table is searched from a plurality of field relationship lists corresponding to a plurality of data tables, the field relationship list corresponding to the first data table includes a correlation relationship corresponding to a field in the first data table, the correlation relationship includes a field in the first data table and a corresponding associated field in the first data table, and the plurality of field relationship lists corresponding to the plurality of data tables are obtained by performing field similarity and correlation analysis on the plurality of data tables;

[0009] Based on the field relationship list corresponding to the first data table, a data table correlation graph between the first data table and a second data table is constructed, the second data table being a data table to which the associated field belongs; based on the data table correlation graph, the field in the first data table and the field in the second data table are assigned values, and the value of the field in the first data table is consistent with the value of the associated field corresponding to the field in the first data table.

[0010] In a second aspect, the present application provides a data generation device, comprising:

[0011] The acquisition module is configured to obtain identification information of a first data table to be generated;

[0012] The searching module is configured to search, with the identification information as a keyword, the field relation list corresponding to the first data table from a plurality of field relation lists corresponding to a plurality of data tables, wherein the field relation list corresponding to the first data table includes the association relationship of the fields in the first data table, and the association relationship includes the fields in the first data table and the associated fields corresponding to the fields in the first data table, and the plurality of field relation lists corresponding to the plurality of data tables are obtained by performing field similarity and association analysis on the plurality of data tables.

[0013] The constructing module is configured to construct, based on the field relation list corresponding to the first data table, a data table association relationship graph between the first data table and a second data table, wherein the second data table is a data table to which the associated fields belong.

[0014] The assigning module is configured to assign values to the fields in the first data table and the fields in the second data table based on the data table association relationship graph, wherein the value of the field in the first data table is consistent with the value of the associated field corresponding to the field in the first data table.

[0015] In a third aspect, the present application provides an electronic device, comprising a processor and a memory connected with the processor in communication; the memory stores computer execution instructions; and the processor executes the computer execution instructions stored in the memory to implement the data generation method according to the first aspect.

[0016] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by a processor to implement the data generation method according to the first aspect.

[0017] In a fifth aspect, the present application provides a computer program product, wherein the computer program product comprises a computer program, and the computer program is executed by a processor to implement the data generation method according to the first aspect.

[0018] The data generation method, device, equipment and storage medium provided by the present application, by pre-analyzing the field similarity and correlation of multiple data tables, a field relationship list corresponding to each of the multiple data tables is obtained, including a field relationship list corresponding to a first data table. When preparing to generate data for the first data table, the field relationship list corresponding to the first data table can be queried from the field relationship lists corresponding to the multiple data tables. The field relationship list records the fields in the first data table and the corresponding associated fields in the first data table. Therefore, the data table correlation graph between the first data table and a second data table can be established based on the field relationship list, and the second data table is a data table to which the associated fields belong; the fields in the first data table and the fields in the second data table are assigned based on the data table correlation graph, so as to ensure that the values of the fields in the first data table are consistent with the values of the corresponding associated fields. In this way, the user does not need to manually find other data tables associated with the first data table, and does not need to manually modify the values of the fields in each data table one by one, thereby effectively reducing the workload of data preparation and improving the efficiency of data preparation. BRIEF DESCRIPTION OF DRAWINGS

[0019] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the application.

[0020] Figure 1 An example diagram of an application scenario applicable to an embodiment of the present application;

[0021] Figure 2 A flowchart of a data generation method provided by an embodiment of the present application;

[0022] Figure 3 A flowchart of a generation process of a field relationship list corresponding to each of multiple data tables in a data generation method provided by an embodiment of the present application;

[0023] Figure 4 An example diagram of a data processing flow provided by an embodiment of the present application;

[0024] Figure 5 A structural diagram of a data generation device provided by an embodiment of the present application;

[0025] Figure 6 A structural diagram of an electronic device provided by an embodiment of the present application.

[0026] The above-described drawings have shown the specific embodiments of the present application, and will be described in more detail hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of the present application by any means, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0027] The exemplary embodiments will be described in detail herein with reference to the attached drawings. In the following description, like reference numerals refer to like elements throughout the description. The following exemplary embodiments are not representative of all embodiments consistent with the present application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the present application as detailed in the appended claims.

[0028] It should be noted that the data generation method, device, equipment and storage medium provided by the present application can be used in the field of big data, and can also be used in any field other than the field of big data. The application field of the data generation method, device, equipment and storage medium provided by the present application is not limited.

[0029] The tester has no way to check and analyze the data association between the background data tables, and mainly relies on the logic document provided by the developer for data preparation. However, the logic document may have problems of missing data tables and association, which leads to that the tester needs to check and modify each data table one by one if the tester wants to perform data rework in the actual test process, which is large in workload and low in efficiency; if the tester only generates numbers for each table one by one, the linkage generation of numbers between different tables cannot be realized, which leads to different values of the fields associated between the tables.

[0030] To solve the above problems, the data generation method, device, equipment and storage medium provided by the present application analyze the field similarity and association of a plurality of data tables to obtain a field relationship list corresponding to each of the plurality of data tables, the field relationship list including fields in the data table and associated fields corresponding to the fields; during data preparation, a data table association relationship graph between the data table to be generated and related data tables can be established based on the field relationship list corresponding to the data table to be generated; and linkage generation of numbers between the data table to be generated and the related data tables is performed based on the data table association relationship graph. Thus, it is not necessary to manually generate numbers for each table one by one, which reduces the manual intervention in data preparation, reduces the workload of data preparation, improves the efficiency of data preparation, and ensures the accuracy of number generation in the data preparation process.

[0031] The specific application scenario of the present application is a data processing scenario. Figure 1 An example diagram of an application scenario to which the embodiments of the present application are applicable is shown in FIG. 1. Figure 1 As shown in FIG. 1, the application scenario can involve a data generation device 101, which can be a server or a terminal. Figure 1Taking the server as an example, on the data generation apparatus 101, the linkage assignment between the data tables can be performed based on the field relationship list corresponding to the data table, so as to ensure that the values of the associated fields between the data tables are consistent.

[0032] Optionally, the application scenario can further include an interactive apparatus 102, which can be a terminal. A user can input a number request on the interactive apparatus 102 and input the identification information of the data table to be numbered. The interactive apparatus 102 communicates with the data generation apparatus 101, and the data generation apparatus 101 performs linkage assignment for the data table to be numbered and the data table associated with the data table in response to the request input by the user on the interactive apparatus 102.

[0033] The terminal can be a personal digital assistant (PDA) device, a handheld device (for example, a smart phone or a tablet computer) with a wireless communication function, a computing device (for example, a personal computer (PC)), a wearable device (for example, a smart watch or a smart bracelet), and a smart home device (for example, a smart speaker or a smart display device), and the like. The server can be a standalone server or a server cluster, and can be a local server or a cloud server.

[0034] The technical solutions of the present application and how the technical solutions solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0035] Figure 2 A flowchart of a data generation method provided by an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the data generation method includes the following steps: Figure 2

[0036] S201, obtaining identification information of a first data table to be numbered.

[0037] Among the plurality of data tables, the identification information of the first data table uniquely identifies the first data table.

[0038] Optionally, the identification information of the first data table includes a table name of the first data table. The table name of the first data table can include at least one language description of the table name of the first data table, such as a Chinese table name or an English table name.

[0039] Optionally, the identification information of the first data table includes a number of the first data table.

[0040] ​The first data table to be created is a field in the first data table to be assigned a value. For a data table, multiple data can be included, and one field can correspond to multiple values.

[0041] The first data table to be created can be one or more data tables.

[0042] In this embodiment, the identification information of the first data table input by the user can be obtained; or the identification information of the first data table to be created can be obtained from the user's request for creating; or the identification information of the first data table to be created can be obtained from the configured data table creation task.

[0043] S202, taking the identification information of the first data table as a keyword, searching for the field relationship list corresponding to the first data table from the field relationship lists respectively corresponding to multiple data tables.

[0044] The field relationship list corresponding to the first data table includes the association relationship of the field in the first data table, and the association relationship includes the field in the first data table and the associated field corresponding to the field in the first data table. In the field relationship list corresponding to the first data table, one field in the first data table can correspond to one or more association relationships; different association relationships corresponding to one field can include different associated fields. For example, in the field relationship list corresponding to the first data table, one or more association relationships corresponding to the first field in the first data table, one or more association relationships corresponding to the second field in the first data table, and so on are included; if the first field corresponds to multiple association relationships, the first association relationship in the multiple association relationships can include the first associated field corresponding to the first field, the second association relationship can include the second associated field corresponding to the first field, and so on.

[0045] Taking table A, table B, and table C as examples, in the field relationship list corresponding to table A, multiple association relationships corresponding to the first field of table A, multiple association relationships corresponding to the second field of table A, and so on are included. Among them, in one association relationship corresponding to the first field of table A, the first field of table A is associated with the first field of table B, and in another association relationship corresponding to the first field of table A, the first field of table A is associated with the second field of table C. It can be seen that in the field relationship list corresponding to table A, multiple association relationships respectively corresponding to each field in table A can be described, and in different association relationships corresponding to one field, the field corresponds to different associated fields.

[0046] The field relationship list corresponding to each of the plurality of data tables is obtained by performing field similarity and correlation analysis on the plurality of data tables, so as to improve the accuracy of the correlation relationship of each field in the field relationship list. Specifically, in the field relationship list, the similarity of the fields in each data table is determined to obtain the similar fields corresponding to the fields in each data table; then, for each data table in the plurality of data tables, the correlation analysis is performed on the fields in the data table and the similar fields corresponding to the fields in the data table, and the correlation fields corresponding to the fields in the data table are determined from the similar fields corresponding to the fields in the data table.

[0047] In this embodiment, the correspondence between the data table and the field relationship list can be reflected as the correspondence between the identification information of the data table and the field relationship list. Therefore, the field relationship list corresponding to the first data table is found in the field relationship lists corresponding to the plurality of data tables, with the identification information of the first data table as the key.

[0048] S203, based on the field relationship list corresponding to the first data table, a data table correlation relationship graph between the first data table and a second data table is constructed, and the second data table is a data table to which the correlation field belongs.

[0049] The second data table can be one or more data tables. In the case of multiple second data tables, it indicates that there is a field-level correlation relationship between the first data table and the plurality of other data tables. For example, the first data table is table A, and the second data table can include table B and table C.

[0050] In this embodiment, in the field relationship list corresponding to the first data table, the correlation relationship can include the identification information of the second data table to which the correlation field belongs, and the second data table can be imported based on the identification information of the second data table; based on the correlation relationship between the fields in the first data table and the fields in the second data table in the field relationship list corresponding to the first data table, a data table correlation relationship graph between the first data table and the second data table is established. The data table correlation relationship graph can be presented in the form of a list, and the correlation relationship of each field in the first data table is described in the list; the data table correlation relationship graph can be presented in the form of a graph, and specifically, the correlation relationship between the fields in the first data table and the fields in the second data table can be identified by the nodes corresponding to the first data table, the nodes corresponding to the fields in the first data table, the nodes corresponding to the second data table, the nodes corresponding to the fields in the second data table, and the connection lines between the nodes. Compared with the field relationship list, the data table correlation relationship graph can clearly show which second data tables are associated with the first data table, and can also clearly show which fields in the first data table are associated with which fields in the second data table.

[0051] S204, based on the data table association relationship graph, values are assigned to the fields in the first data table and the fields in the second data table, and the value of the field in the first data table is consistent with the value of the associated field corresponding to the field in the first data table.

[0052] In this embodiment, based on the data table association relationship graph, the same value is assigned to the field in the first data table and the corresponding associated field in the second data table, so as to ensure that the value of the field in the first data table is consistent with the value of the associated field corresponding to the field in the first data table. In this way, the user only needs to assign values to the fields in the first data table, without assigning values to the second data table one by one, thereby reducing the workload of the user in data generation and improving the efficiency.

[0053] In the embodiment of the present application, the field similarity and association of multiple data tables are analyzed to obtain a field relationship list corresponding to each of the multiple data tables. Based on the field relationship list corresponding to the data table to be generated, a data table association relationship graph is established between the data table to be generated and the related data tables, and based on the data table association relationship graph, the data table to be generated and the related data tables are linked to generate data. Thus, it is not necessary to manually generate data for each table one by one, thereby reducing the manual intervention in data preparation, and through automatic analysis of the association relationship between the fields of the data tables and automatic linkage generation of data, the workload of data preparation is reduced, the efficiency of data preparation is improved, and the accuracy of data generation in the data preparation process is ensured.

[0054] Figure 3 A flowchart of the generation process of the field relationship list corresponding to each of the multiple data tables in the data generation method provided in the embodiment of the present application is shown in FIG. Figure 3 As shown in the figure, the generation process of the field relationship list includes the following steps:

[0055] S301, obtaining data table information of multiple data tables, the data table information including table structure and table data.

[0056] Among the data table information of the multiple data tables, the table structure corresponding to each of the multiple data tables and the table data corresponding to each of the multiple data tables are included. The table structure corresponding to the data table mainly reflects the business corresponding to the data table and the fields of the data table, and the table data corresponding to the data table refers to the values of the fields of the data table, specifically, the multiple rows of data in the data table.

[0057] In this embodiment, the data table information of the multiple data tables can be obtained from the production environment of the data table. The production environment of the data table, such as the development environment, the business environment, etc., can obtain example values of the table data of the data table or actual values of the table data of the data table, and the table data can truly reflect the association between the fields.

[0058] In a possible implementation, the data table information of the plurality of data tables can be acquired from the production environment of the data tables according to the data acquisition parameters according to the configured acquisition parameters, so as to improve the information acquisition accuracy based on the acquisition parameters.

[0059] Optionally, the acquisition parameters include a timing time parameter. The data table information of the plurality of data tables can be acquired from the production environment of the data tables periodically according to the timing time parameter. In addition, the acquisition parameters can further include identification information of the data tables, so as to acquire the data table information of specific data tables. In addition to periodically acquiring the data table information of the plurality of data tables, the data table information of the plurality of data tables can also be acquired at random according to actual needs.

[0060] S302, feature extraction is performed on the table structure of the plurality of data tables to obtain table features corresponding to the plurality of data tables respectively.

[0061] Optionally, one data table can correspond to one or more table features.

[0062] In this embodiment, the table features related to the business and the fields of the data tables can be extracted from the table structure corresponding to the data tables, because the table structure corresponding to the data tables mainly reflects the business and the fields of the data tables.

[0063] In a possible implementation, S302 can include: for each data table in the plurality of data tables, performing feature extraction on the table structure of the data table according to a configured feature type to obtain the table feature of the data table under the feature type. The feature type includes at least one of the following: data table name, business type to which the data table belongs, data table field name, and data table field type. The data table name usually includes words related to the business, so the data table name and the business type to which the data table belongs are related to the business of the data table; the data table field name and the data table field type are related to the fields of the data table. Therefore, the table features related to the business and / or the fields of the data table can be extracted.

[0064] Optionally, one data table can correspond to one or more data table names.

[0065] Optionally, the data table name includes table names in a plurality of languages, such as Chinese table names and English table names.

[0066] Optionally, one field can correspond to one or more data table field names.

[0067] Optionally, the data table field name includes field names in a plurality of languages, such as Chinese field names and English field names.

[0068] The data table belongs to a business type such as a deposit business, a withdrawal business, and the like, and a data table field type such as a text type, a character type, a date type, a currency type, and the like.

[0069] In the present implementation, for each data table in the plurality of data tables, the table structure of the data table is analyzed and refined, and different information refined is classified according to a configured feature type to obtain table features of the data table under the feature type. The feature type includes at least one of the following: a data table name, a business type to which the data table belongs, a data table field name, and a data table field type. The extraction process of the table features can include at least one of the following: extracting the table name of the data table from the table structure of the data table, and classifying the table name of the data table as a table feature under the feature type of the data table name; extracting the business type to which the data table belongs from the table structure of the data table, and classifying the business type to which the data table belongs as a table feature under the feature type of the business type to which the data table belongs; extracting the field name of each field in the data table from the table structure of the data table, and classifying the field name of each field in the data table as a table feature under the feature type of the data table field name; and extracting the field type of each field in the data table from the table structure of the data table, and classifying the field type of each field in the data table as a table feature under the feature type of the data table field type.

[0070] Optionally, after obtaining the table features corresponding to the plurality of data tables respectively, the table features corresponding to the plurality of data tables respectively can be stored in a feature database for subsequent use.

[0071] S303, similarity analysis is performed on the table features corresponding to the plurality of data tables respectively to obtain an initial relationship list corresponding to the plurality of data tables respectively.

[0072] For each data table in the plurality of data tables, the initial relationship list includes a similarity relationship corresponding to a field in the data table, and the similarity relationship includes the field in the data table, a similar field corresponding to the field in the data table, and a similarity degree between the field in the data table and the similar field.

[0073] For each data table in the plurality of data tables, the field in the data table can be one or more. Taking one field as an example, in the initial relationship list corresponding to the data table, the one field in the data table can correspond to one or more similarity relationships, and one similarity relationship includes one similar field. In the plurality of similarity relationships corresponding to the one field, different similarity relationships include different similar fields. Thus, it can be understood that in the initial relationship list corresponding to the data table, the one field in the data table can correspond to one or more similar fields, and the one field in the data table and each similar field can form a similarity relationship respectively.

[0074] As an example, the initial relation list corresponding to table A includes: a plurality of similar relations corresponding to the first field in table A, a plurality of similar relations corresponding to the second field in table A, and so on. The first similar relation corresponding to the first field in table A includes the first field in table A, the first field in table B, and the similarity between the first field in table A and the first field in table B; the second similar relation corresponding to the first field in table A includes the first field in table A, the second field in table C, and the similarity between the first field in table A and the second field in table C. The first field in table B and the second field in table C are similar fields corresponding to the first field in table A.

[0075] In this embodiment, for a data table, since the table features corresponding to the data table are related to the business of the data table and the fields of the data table, if the businesses of two data tables are similar and the fields of the two data tables are similar, it can be determined that the fields of the two data tables are likely to be similar fields. Therefore, the similar fields corresponding to the fields in each data table and the similarity between the fields in each data table and the similar fields can be obtained by performing similarity analysis on the table features corresponding to the plurality of data tables respectively, and the initial relation list corresponding to each data table can be obtained in combination with the similar fields corresponding to the fields in each data table and the similarity between the fields in each data table and the similar fields.

[0076] In a possible implementation, as shown in FIG. 3, Figure 3 S303 includes: S3031, for each data table in the plurality of data tables, combining the plurality of table features corresponding to the data table according to a feature combination mode to obtain a combined feature corresponding to the data table; S3032, performing field similarity identification based on the plurality of table features corresponding to the plurality of data tables respectively and the combined features corresponding to the plurality of data tables respectively to obtain field similarity between the plurality of data tables; and S3033, obtaining an initial relation list according to the field similarity between the plurality of data tables. Thus, the table features and the combined features obtained by combining the table features provide more feature data for field similarity identification between data tables and provide accuracy of field similarity.

[0077] In S3031, the feature combination mode is, for example, two-by-two combination, combination of three table features, combination of four table features, and the like. The table features for combination can be randomly selected or specified. The combination of table features is performed separately for each data table in the plurality of data tables to avoid confusion of table features between different data tables. For each data table, the table features for combination are selected from the plurality of table features corresponding to the data table according to the feature combination mode; and the table features for combination are combined to obtain the combined feature corresponding to the data table.

[0078] Optionally, the feature combination manners are multiple, and different feature combination manners correspond to different combined features. In other words, the multiple table features are combined according to different feature combination manners, and multiple combined features can be obtained. Based on this, one possible implementation of S3031 comprises: for each data table in the multiple data tables, combining, according to a feature combination manner, the table features belonging to different feature types in the multiple table features corresponding to the data table, to obtain multiple combined features corresponding to the data table. In this way, the richness of the combined features is improved through multiple feature combination manners.

[0079] Further, the multiple table features include a data table name, a business type to which the data table belongs, a data table field name, and a data table field type, and there are four types of table features. The combination of table features can include: selecting two types of features for combination, selecting three types of features for combination, and selecting four types of features for combination.

[0080] Further, if the data table name includes a data table English name and a data table Chinese name, and the data table field name includes a data table field English name and a data table field Chinese name, there are six types of features. The combination of table features can include: selecting two types of features for combination, selecting three types of features for combination, selecting four types of features for combination, selecting five types of features for combination, and selecting six types of features for combination.

[0081] Further, the table features of the feature type of data table name and the table features of the feature type of data table field name are included in different combined features. Since the data table name usually contains business information, such as a deposit business statistical table and a loan business statistical table, the table features of the feature type of data table name can reflect the business to which the data table belongs, and the table features of the feature type of data table field name reflect the field name of the data table. If the table features of the feature type of data table field name of two tables are similar, but the table features of the feature type of data table name of the two tables are not similar, the fields with similar names in the two tables are only similar in name, and are not actually similar fields. For example, the balance in the deposit business statistical table and the balance in the loan business statistical table are not related. If the table features of the feature type of data table field name of two tables are similar, and the table features of the feature type of data table name of the two tables are similar, the fields with similar names in the two tables are similar fields under similar businesses, and are likely to be related fields. Therefore, by combining the table features of the feature type of data table name and the table features of the feature type of data table field name, the accuracy of identifying similar fields can be improved.

[0082] In S3032, for each data table, the field can be multiple, so the table features of the feature type of the data table field name can be multiple. To determine the similarity of the fields between two data tables, the similarity between the table features of the data table field name of the two data tables needs to be considered. For example, the similarity between the first table feature of the data table field name of data table A (e.g., the first field name of data table A) and the first table feature of the data table field name of data table B (e.g., the first field name of data table B), and the similarity between the first table feature of the data table field name of data table A and the second table feature of the data table name of data table B (e.g., the second field name of data table B). Therefore, in multiple data tables, when determining the similarity of the fields between two data tables, the table features of the two data tables under the same feature type can be matched according to the feature type to obtain the similarity between the table features of the two data tables under the same feature type, wherein the similarity between the table features of the two data tables under the same feature type can include the similarity between the table features of the data table field name of the two data tables; the combined features of the two data tables under the same feature combination mode can be matched according to the feature combination mode to obtain the similarity between the combined features of the two data tables under the same feature combination mode; then, the similarity between the table features of the two data tables under the data table field name, the similarity between the table features of the two data tables under the remaining feature types in the multiple feature types except the data table field name, and the similarity between the combined features of the two data tables under the feature combination mode can be weighted to obtain the final similarity between the table features of the two data tables under the data table field name; and based on the final similarity between the table features of the two data tables under the data table field name, the field similarity between the two data tables can be determined.

[0083] In a possible implementation of S3032, the multiple table features corresponding to the multiple data tables and the multiple combined features corresponding to the multiple data tables can be input into a text similarity recognition algorithm for similarity recognition to obtain the field similarity between the multiple data tables. The text similarity recognition algorithm can be a deep learning algorithm, and the table features and the combined features can be regarded as texts, so that the accuracy of the similarity recognition is improved by using the text similarity recognition algorithm.

[0084] In the present implementation, in the similarity recognition process, for the multiple data tables, the multiple table features corresponding to each two data tables and the multiple combined features corresponding to each two data tables can be input into a text similarity recognition algorithm, and a deep learning model is used in the text similarity recognition algorithm to perform the above similarity recognition process to obtain the field similarity between each two data tables. In this way, the field similarity between the multiple data tables can be obtained.

[0085] In S3033, after the similarity between the fields of the plurality of data tables is obtained, for each data table in the plurality of data tables, the fields of the data table can be filtered from the fields of the remaining data tables in the plurality of data tables based on the similarity between the fields of the data table and the fields of the remaining data tables. For example, the fields with the top 10% similarity to the fields of the data table are filtered as the similar fields corresponding to the fields of the data table. In this way, one or more similar fields corresponding to the fields in the data table and the similarity between the fields in the data table and the similar fields can be obtained, and the fields in the data table, the similar fields corresponding to the fields in the data table, and the similarity between the fields in the data table and the similar fields are combined to obtain the initial relationship list corresponding to the data table. In addition to the percentage filtering, the filtering can also be based on a similarity threshold.

[0086] S304, based on the intra-table data of the plurality of data tables, verifying and filtering the similar relationships corresponding to the fields in the initial relationship list corresponding to the plurality of data tables respectively to obtain the field relationship list corresponding to the plurality of data tables respectively.

[0087] The intra-table data in the plurality of data tables is the value of the field in the plurality of data tables, i.e., the instance data. Whether the fields are related is reflected in the similarity of the data table business type and the field name, and also reflected in the similarity of the field values, so after the initial relationship list corresponding to the data table is obtained based on the structural characteristics of the data table in the foregoing step, the similar relationships corresponding to the fields in the initial relationship list corresponding to the plurality of data tables respectively can be verified and filtered based on the intra-table data of the plurality of data tables to obtain the field relationship list corresponding to the plurality of data tables respectively, so as to improve the accuracy of the association relationship of the fields in the field relationship list.

[0088] In this embodiment, for each data table in the plurality of data tables, a plurality of values of the field and a plurality of values of the similar field corresponding to the field in the initial relationship list corresponding to the data table can be obtained from the intra-table data of the plurality of data tables, and the association degree of the field and the similar field in the values is judged based on the plurality of values of the field and the plurality of values of the similar field. In this way, if the field corresponds to a plurality of similar fields, the association degree of the field and each similar field in the plurality of similar fields in the values can be obtained, the plurality of similar fields are filtered based on the association degree of the field and each similar field in the plurality of similar fields in the values to obtain the associated field corresponding to the field, and the field relationship list corresponding to the data table is obtained based on the associated field corresponding to the field. In this way, the field relationship list corresponding to the plurality of data tables respectively is obtained.

[0089] In one possible implementation manner, as Figure 3As shown, S304 can include: S3041, for each data table in the plurality of data tables, obtaining, from the intra-table data of the plurality of data tables, a plurality of values of a field in the initial relationship list corresponding to the data table and a plurality of values of a similar field corresponding to the field; S3042, performing overlap comparison on the plurality of values of the field and the plurality of values of the similar field corresponding to the field to obtain a numerical value overlap ratio between the field and the similar field corresponding to the field; S3043, determining an association score between the field and the similar field corresponding to the field of the data table in combination with the overlap ratio between the field and the similar field corresponding to the field of the data table and the similarity between the field and the similar field corresponding to the field of the data table; S3044, based on the association score between the field and the similar field corresponding to the field of the data table, filtering out an associated field corresponding to the field of the data table from the similar field corresponding to the field of the data table; and S3045, generating the field relationship list corresponding to the data table in combination with the field of the data table, the associated field corresponding to the field of the data table, and the association score between the field and the associated field corresponding to the field of the data table. Thus, by performing overlap comparison on the values between the field and the similar field, the accuracy of verifying the similar relationship is improved, the field is filtered in combination with the overlap ratio between the values of the field and the similar field and the similarity between the field and the similar field, the accuracy of the field filtering is improved, and thus the accuracy of the associated field corresponding to the field in the field relationship list is effectively improved.

[0090] In S3041, for each data table in the plurality of data tables, the field in the initial relationship list corresponding to the data table is included in the data table, and the value of the field in the data table can be obtained from the intra-table data of the data table; the similar field corresponding to the field in the initial relationship list corresponding to the data table is included in the data table, and the value of the similar field can be obtained from the intra-table data of the data table to which the similar field belongs. For example, the value of the field in the first data table can be obtained from the intra-table data of the first data table, and the value of the field in the second data table can be obtained from the intra-table data of the second data table.

[0091] In S3042-S3043, for each data table in the plurality of data tables, in the initial relationship list corresponding to the data table, one field in the data table can correspond to one or more similar relationships, that is, one field in the data table can correspond to one or more similar fields, and for each similar relationship, the following verification process can be performed: the values of the field (the field in the data table) and the values of the similar field are compared for overlap to obtain the number of overlaps between the values of the field and the values of the similar field, and the number of overlaps is compared with the number of values of the field to obtain the value overlap ratio between the field and the corresponding similar field, for example, the field has 100 values in total, and 20 values overlap with 20 values of the similar field, so the overlap ratio is 20%, and the higher the overlap ratio, the higher the correlation; then, the correlation score between the field of the data table and the corresponding similar field is determined in combination with the overlap ratio between the field of the data table and the corresponding similar field and the similarity between the field of the data table and the corresponding similar field, and the correlation score reflects the degree of correlation between the field and the similar field; based on the correlation score between the field of the data table and the corresponding similar field, the associated field corresponding to the field of the data table is selected from the similar field corresponding to the field of the data table to select the similar field with a higher degree of correlation as the associated field.

[0092] In S3043, in the process of determining the correlation score between the field of the data table and the corresponding similar field in combination with the overlap ratio between the field of the data table and the corresponding similar field and the similarity between the field of the data table and the corresponding similar field, the overlap ratio between the field of the data table and the corresponding similar field and the similarity between the field of the data table and the corresponding similar field can be weighted to obtain the correlation score between the field of the data table and the corresponding similar field. Thus, on the one hand, the numerical correlation of the field and the similarity of the field in business and name are reasonably integrated, and on the other hand, the weights of the overlap ratio and the similarity can be adjusted to improve the accuracy of the correlation score.

[0093] Optionally, the overlap ratio between the field of the data table and the corresponding similar field and the similarity between the field of the data table and the corresponding similar field can be input into a scoring model (such as a deep learning model) to obtain the correlation score output by the scoring model, and the scoring model can be pre-trained.

[0094] In S3044, the associated field corresponding to the field of the data table can be selected from the similar field corresponding to the field of the data table in the order from high to low of the correlation score between the field of the data table and the corresponding similar field.

[0095] Optionally, the similar fields corresponding to the fields in the data table are filtered from the similar fields corresponding to the fields in the data table in the order of the association scores from high to low, and a preset number or a preset proportion of the similar fields are selected as the association fields. Alternatively, the similar fields with the association scores greater than a score threshold are filtered from the similar fields corresponding to the fields in the data table as the association fields.

[0096] In S3045, in addition to the fields in the data table and the association fields corresponding to the fields in the data table, the association relationship list corresponding to the data table can further include the association scores between the fields in the data table and the corresponding association fields, so that the field relationship list corresponding to the data table can be generated in combination with the fields in the data table, the association fields corresponding to the fields in the data table, and the association scores between the fields in the data table and the corresponding association fields.

[0097] Optionally, after obtaining the initial relationship lists corresponding to the plurality of data tables respectively, the initial relationship lists corresponding to the plurality of data tables respectively can be stored in the data table similarity information library, and the initial relationship list corresponding to the data table can be obtained from the data table similarity information library when verification and filtering are performed subsequently.

[0098] In the embodiments of the present application, in the process of generating the field relationship lists corresponding to the plurality of data tables respectively, the similarity analysis is first performed based on the table features corresponding to the plurality of data tables respectively to obtain the initial relationship lists corresponding to the plurality of data tables respectively, and the similar fields corresponding to the fields in the data table and the similarity between the fields and the similar fields are recorded in the initial relationship lists; then, the similar relationship of the fields in the initial relationship lists corresponding to the plurality of data tables respectively is verified and filtered according to the data in the plurality of data tables, and the association fields corresponding to the fields are selected therefrom to obtain the association relationship tables corresponding to the plurality of data tables respectively. In this way, the accuracy of the association relationship table is effectively improved, and the dependence on other data providers for providing the association relationship of the fields in the data table is reduced, that is, the dependence on the data providers is reduced.

[0099] As an example, Figure 4 A data processing flow example diagram is provided for the embodiments of the present application. As shown in Figure 4 The data table information of the plurality of data tables can be obtained by the data table information acquisition module 1 first, and the information is stored in the data table information library; then, the feature extraction, feature combination, and similarity identification of the fields are performed on the plurality of data tables by the data table information feature extraction module 2, and the obtained initial relationship list can be stored in the data table similarity information library; then, the initial relationship list in the data table similarity information library is verified and filtered by the data table correlation score module 3, and finally the association relationship list is generated, and the association relationship list is stored in the data table similarity information library. After that, the intelligent data generation module 4 automatically generates data for the data table by using the association relationship list.

[0100] In some embodiments, the value assignment for the fields in the first data table and the fields in the second data table based on the data table association graph comprises: obtaining a data table generation mode; if the data table generation mode is that the user fills in the data by himself / herself, obtaining the value input by the user for the fields in the first data table, and assigning values to the fields in the second data table based on the data table association graph and the value input by the user for the fields in the first data table; if the data table generation mode is an automatic number generation mode, obtaining the in-table data of the plurality of data tables, and assigning values to the fields in the first data table and the fields in the second data table based on the in-table data and the data table association graph. Thus, the user can flexibly select the data table generation mode, and when the data table generation mode is that the user fills in the data by himself / herself, the value assignment for the fields in the second data table can be automatically linked based on the value input by the user for the fields in the first data table, thereby reducing the workload of the user in this mode and improving the number generation efficiency; when the data table generation mode is the automatic number generation mode, the in-table data of the plurality of data tables is used to assign values to the fields in the first data table and the fields in the second data table, without the need for the user to fill in the data, thereby improving the number generation efficiency, reducing the manual intervention in the data preparation process, and achieving the effect of automatic number generation.

[0101] In the embodiment, if the user selects the data filling mode by himself / herself, the user only needs to fill in the data of the associated fields of one table, and then the data of the fields can be automatically filled into the other fields having the association relationship, which comprises: filling the value input by the user in the fields of the first data table into the corresponding associated fields in the second data table; and / or filling the value input by the user in the fields of the second data table into the corresponding associated fields in the first data table. If the user selects the automatic number generation mode, the in-table data of the first data table and the in-table data of the second data table can be obtained from the in-table data of the plurality of data tables, and the values of the fields in the first data table and the fields in the second data table are assigned according to the in-table data of the first data table and the in-table data of the second data table, while ensuring that the values of the fields in the first data table and the corresponding associated fields in the second data table are consistent. As for the non-associated fields in the second data table (i.e., the fields that are not associated with the fields in the first data table), the user can manually fill in the data, or the second data table can be automatically assigned values according to the in-table data of the plurality of data tables (the in-table data of the plurality of data tables includes the in-table data of the second data table, i.e., the instance data of the second data table).

[0102] Figure 5 A structural schematic diagram of a data generation device provided by the embodiment of the application is shown in FIG. 5. Figure 5 As shown in FIG. 5, the data generation device 500 provided by the embodiment of the application comprises:

[0103] The obtaining module 501 is configured to obtain the identification information of the first data table to be generated.

[0104] The searching module 502 is configured to search, with the identification information as a keyword, the field relation list corresponding to the first data table from the field relation lists corresponding to the plurality of data tables, the field relation list corresponding to the first data table including the association relationship of the fields in the first data table, the association relationship including the fields in the first data table and the associated fields corresponding to the fields in the first data table, and the field relation lists corresponding to the plurality of data tables being obtained by performing field similarity and association analysis on the plurality of data tables.

[0105] The constructing module 503 is configured to construct, based on the field relation list corresponding to the first data table, a data table association relationship graph between the first data table and a second data table, the second data table being a data table to which the associated fields belong.

[0106] The assigning module 504 is configured to assign, based on the data table association relationship graph, values to the fields in the first data table and the fields in the second data table, the value of the field in the first data table being consistent with the value of the associated field corresponding to the field in the first data table.

[0107] In some embodiments, the field relation lists corresponding to the plurality of data tables are generated in the following manner: obtaining data table information of the plurality of data tables, the data table information including table structure and table data; performing feature extraction on the table structure of the plurality of data tables to obtain table features corresponding to the plurality of data tables; performing similarity analysis on the table features corresponding to the plurality of data tables to obtain initial relation lists corresponding to the plurality of data tables, wherein for each data table in the plurality of data tables, the initial relation list includes a similarity relationship of the fields in the data table, the similarity relationship including the fields in the data table, similar fields corresponding to the fields in the data table, and a similarity degree between the fields in the data table and the similar fields; verifying and screening the similarity relationship of the fields in the initial relation lists corresponding to the plurality of data tables based on the table data of the plurality of data tables to obtain the field relation lists corresponding to the plurality of data tables.

[0108] In some embodiments, the feature extraction on the table structure of the plurality of data tables to obtain the table features corresponding to the plurality of data tables includes: for each data table in the plurality of data tables, performing feature extraction on the table structure of the data table according to a configured feature type to obtain table features of the data table under the feature type, wherein the feature type includes at least one of the following: data table name, business type to which the data table belongs, data table field name, and data table field type.

[0109] In some embodiments, similarity analysis is performed on table features corresponding to the plurality of data tables respectively to obtain an initial relationship list corresponding to the plurality of data tables respectively, including: for each data table in the plurality of data tables, combining the plurality of table features corresponding to the data table according to a feature combination mode to obtain a combined feature corresponding to the data table; performing field similarity identification based on the plurality of table features corresponding to the plurality of data tables respectively and the combined features corresponding to the plurality of data tables respectively to obtain field similarity between the plurality of data tables; and obtaining the initial relationship list according to the field similarity between the plurality of data tables.

[0110] In some embodiments, the feature combination mode is multiple, different feature combination modes correspond to different combined features, and different combined features each include a table feature of a feature type of a data table table name and a table feature of a feature type of a data table field name; for each data table in the plurality of data tables, combining the plurality of table features corresponding to the data table according to the feature combination mode to obtain the combined feature corresponding to the data table, including: for each data table in the plurality of data tables, combining the table features of different feature types in the plurality of table features corresponding to the data table according to the feature combination mode to obtain a plurality of combined features corresponding to the data table.

[0111] In some embodiments, the field similarity between the plurality of data tables is obtained by performing field similarity identification based on the plurality of table features corresponding to the plurality of data tables respectively and the combined features corresponding to the plurality of data tables respectively, including: inputting the plurality of table features corresponding to the plurality of data tables respectively and the plurality of combined features corresponding to the plurality of data tables respectively into a text similarity identification algorithm for similarity identification to obtain the field similarity between the plurality of data tables.

[0112] In some embodiments, the similar relationship of a field in the initial relationship list corresponding to the plurality of data tables is verified and filtered based on the data in the table of the plurality of data tables to obtain a field relationship list corresponding to the plurality of data tables, including: for each data table in the plurality of data tables, obtaining a plurality of values of the field and a plurality of values of a similar field corresponding to the field from the data in the table of the data table; performing overlap comparison on the plurality of values of the field and the plurality of values of the similar field corresponding to the field to obtain a numerical overlap ratio between the field and the similar field corresponding to the field; determining an association score between the field of the data table and the similar field corresponding to the field based on the overlap ratio between the field of the data table and the similar field corresponding to the field and the similarity between the field of the data table and the similar field corresponding to the field; filtering an associated field corresponding to the field of the data table from the similar field corresponding to the field of the data table based on the association score between the field of the data table and the similar field corresponding to the field; and generating a field relationship list corresponding to the data table based on the field of the data table, the associated field corresponding to the field of the data table, and the association score between the field of the data table and the associated field corresponding to the field.

[0113] In some embodiments, the assigning module 504 is specifically configured to: acquire a data table generation mode; if the data table generation mode is that the user fills in data by himself / herself, acquire the values input by the user for the fields in the first data table, and assign values to the fields in the second data table based on the data table association graph and the values input by the user for the fields in the first data table; and if the data table generation mode is an automatic number generation mode, acquire the intra-table data of the plurality of data tables, and assign values to the fields in the first data table and the fields in the second data table based on the intra-table data and the data table association graph.

[0114] Figure 6 A structural schematic diagram of an electronic device is provided in the embodiments of the present application. As shown in the figure, the electronic device 600 can include at least one processor 601 and a memory 602. Figure 6 As shown in the figure, the electronic device is taken as an example of one processor. Figure 6

[0115] The memory 602 is configured to store the program of the processor 601. Specifically, the program can include program code, and the program code includes computer operation instructions.

[0116] The memory 602 can include a high-speed RAM memory, and can also include a non-volatile memory, for example, at least one disk memory.

[0117] The processor 601 is configured to execute the computer program stored in the memory 602 to implement each step in the data generation method in each method embodiment.

[0118] The processor 601 can be a central processing unit (CPU) or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0119] ​Optionally, the memory 602 can be independent or integrated with the processor 601. When the memory 602 is independent of the processor 601, the electronic device 600 can further include a bus 603 for connecting the processor 601 and the memory 602. The bus can be an industry standard architecture (ISA) bus, a peripheral component (PCI) bus, or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc., but it does not mean that there is only one bus or one type of bus.

[0120] Optionally, in a specific implementation, if the memory 602 and the processor 601 are integrated on a chip, the memory 602 and the processor 601 can communicate through an internal interface.

[0121] The application further provides a computer readable storage medium, which can include a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes, and specifically, the computer readable storage medium stores computer execution instructions. When at least one processor of the electronic device executes the computer execution instructions, the electronic device executes the steps of the data generation method provided by the various embodiments.

[0122] The application further provides a computer program product, which includes a computer program stored in a readable storage medium. At least one processor of the electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program to make the electronic device implement the steps of the data generation method provided by the various embodiments.

[0123] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. The specification and examples given are considered exemplary only, and the true scope and spirit of the application is indicated by the following claims.

[0124] It is to be understood that the application is not limited to the precise construction already described above and shown in the drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the application. The scope of the application should only be limited by the claims appended hereto.

Claims

1. A data generating method characterized by comprising: The method comprises: obtaining identification information of a first data table to be created; using the identification information as a key, searching for a field relationship list corresponding to the first data table from a plurality of field relationship lists corresponding to a plurality of data tables, the field relationship list corresponding to the first data table comprising an association relationship between a field in the first data table and an associated field corresponding to the field in the first data table, the plurality of field relationship lists being obtained by performing field similarity and association analysis on the plurality of data tables; based on the field relationship list corresponding to the first data table, constructing a data table association relationship graph between the first data table and a second data table, the second data table being a data table to which the associated field belongs; based on the data table association relationship graph, assigning values to the field in the first data table and the field in the second data table, the value of the field in the first data table being consistent with the value of the associated field corresponding to the field in the first data table. The plurality of field relationship lists are generated by: obtaining data table information of the plurality of data tables, the data table information comprising table structure and data in the table; performing feature extraction on the table structure of the plurality of data tables to obtain table features corresponding to the plurality of data tables; performing similarity analysis on the table features corresponding to the plurality of data tables to obtain an initial relationship list corresponding to the plurality of data tables, wherein for each data table in the plurality of data tables, the initial relationship list comprises a similarity relationship between a field in the data table and a similar field corresponding to the field in the data table, and the similarity relationship comprises the field in the data table, the similar field corresponding to the field in the data table, and a similarity degree between the field in the data table and the similar field; based on the data in the table of the plurality of data tables, verifying and screening the similarity relationship between the fields in the initial relationship list corresponding to the plurality of data tables to obtain the field relationship list corresponding to the plurality of data tables.

2. The data generating method according to claim 1, characterized by, The feature extraction on the table structure of the plurality of data tables to obtain the table features corresponding to the plurality of data tables comprises: for each data table in the plurality of data tables, performing feature extraction on the table structure of the data table according to a configured feature type to obtain table features of the data table under the feature type, wherein the feature type comprises at least one of the following: data table name, business type to which the data table belongs, data table field name, and data table field type.

3. The data generating method according to claim 2, characterized by, The similarity analysis on the table features corresponding to the plurality of data tables to obtain the initial relationship list corresponding to the plurality of data tables comprises: for each data table in the plurality of data tables, combining a plurality of table features corresponding to the data table according to a feature combination mode to obtain combined features corresponding to the data table; based on the plurality of table features corresponding to the plurality of data tables and the combined features corresponding to the plurality of data tables, performing similarity recognition on the fields to obtain field similarity between the plurality of data tables; obtaining the initial relationship list according to the field similarity between the plurality of data tables.

4. The data generating method according to claim 3, characterized by, The feature combination manners are multiple, different feature combination manners correspond to different combined features, and the different combined features all include a table feature of a feature type of a data table table name and a table feature of a feature type of a data table field name; The feature combination manner is used to combine the multiple table features corresponding to the data table to obtain the combined feature corresponding to the data table, including: The feature combination manner is used to combine the table features belonging to different feature types in the multiple table features corresponding to the data table to obtain the multiple combined features corresponding to the data table.

5. The data generating method according to claim 3, characterized by, The field similarity between the multiple data tables is obtained by performing field similarity identification based on the multiple table features corresponding to the multiple data tables respectively and the combined features corresponding to the multiple data tables respectively, including: The field similarity between the multiple data tables is obtained by inputting the multiple table features corresponding to the multiple data tables respectively and the multiple combined features corresponding to the multiple data tables respectively into a text similarity identification algorithm to perform similarity identification.

6. The data generating method according to any one of claims 1 to 5, characterized by, The field relationship list corresponding to the multiple data tables respectively is obtained by verifying and screening the similar relationship corresponding to the field in the initial relationship list based on the table data of the multiple data tables, including: For each data table in the multiple data tables, multiple values of the field and multiple values of the similar field corresponding to the field are obtained from the table data; The value overlap ratio between the field and the similar field corresponding to the field is obtained by performing overlap comparison on the multiple values of the field and the multiple values of the similar field corresponding to the field; The association score between the field and the similar field corresponding to the field is determined in combination with the overlap ratio between the field and the similar field corresponding to the field and the similarity between the field and the similar field corresponding to the field; Based on the association score between the field and the similar field corresponding to the field, the associated field corresponding to the field of the data table is screened out from the similar field corresponding to the field of the data table; The field relationship list corresponding to the data table is generated in combination with the field of the data table, the associated field corresponding to the field of the data table, and the association score between the field and the associated field corresponding to the field of the data table.

7. The data generating method according to any one of claims 1 to 5, characterized by, The field in the first data table and the field in the second data table are valued based on the data table association relationship graph, including: An acquisition module is configured to acquire identification information of a first data table to be numbered. An acquisition module is configured to acquire identification information of a first data table to be numbered. ​ 8. A data generating apparatus characterized by comprising: ​ ​ The searching module is configured to search, with the identification information as a keyword, the field relation list corresponding to the first data table from the field relation lists corresponding to the plurality of data tables respectively, wherein the field relation list corresponding to the first data table comprises the association relationship between the field in the first data table and the associated field corresponding to the field in the first data table, and the field relation lists corresponding to the plurality of data tables are obtained by performing field similarity and association analysis on the plurality of data tables; The constructing module is configured to construct, based on the field relation list corresponding to the first data table, a data table association relationship graph between the first data table and a second data table, wherein the second data table is a data table to which the associated field belongs; The assigning module is configured to assign values to the field in the first data table and the field in the second data table based on the data table association relationship graph, wherein the value of the field in the first data table is consistent with the value of the associated field corresponding to the field in the first data table. The field relation lists corresponding to the plurality of data tables are generated in the following manner: obtaining data table information of the plurality of data tables, wherein the data table information comprises table structure and data in the table; performing feature extraction on the table structure of the plurality of data tables to obtain table features corresponding to the plurality of data tables respectively; performing similarity analysis on the table features corresponding to the plurality of data tables respectively to obtain initial relation lists corresponding to the plurality of data tables respectively, wherein for each data table in the plurality of data tables, the initial relation list comprises a similar relationship between the field in the data table and the similar field corresponding to the field in the data table, and the similar relationship comprises the field in the data table, the similar field corresponding to the field in the data table, and the similarity between the field in the data table and the similar field corresponding to the field in the data table; verifying and screening the similar relationship between the fields in the initial relation lists corresponding to the plurality of data tables respectively based on the data in the tables of the plurality of data tables to obtain the field relation lists corresponding to the plurality of data tables respectively.

9. An electronic device, comprising: comprise: a processor, and a memory connected to the processor in communication; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the data generation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by the processor to implement the data generation method according to any one of claims 1 to 7.

Images