Resource data bill generation method and device, and server

Abstract

The invention provides a resource data bill generation method and device and a server, and belongs to the technical field of big data. The method comprises the steps of obtaining an original data table; the original data table comprises original data values and original classification items corresponding to the original data values; processing the original data table by using a first target classification model to obtain a first target classification result; the first target classification result is matched with a preset template of target software; the target software is used for generating a resource data bill according to the preset template; verifying the first target classification result to obtain a target verification result; filling a preset template of the target software with the first target classification result according to a target verification result under the condition that the target verification result is determined to be passed; and utilizing the target software to generate a resource data bill according to the preset template of the filled target software. Based on the method, the efficiency of manufacturing the resource data bill can be improved.

Description

Technical Field

[0001] This specification belongs to the field of big data technology, and in particular relates to a method, device, and server for generating a resource data bill. Background Art

[0002] In the field of big data technology, electronic payroll is an important basis for online banks to pay wages on behalf of others, and is also an important business carrier for linking B-end and C-end customers. It plays a very important role in commercial banks' customer acquisition, retention, and activation. In the prior art, online banks provide customers with preset electronic payroll templates. If customers want to use online banks to generate electronic payrolls, they need to fill in the internal payroll data by column into the preset electronic payroll template, and then upload the completed electronic payroll template to the online bank's server. The online bank's server generates an electronic payroll based on the filled electronic payroll template. This process requires users to manually organize and match the payroll data, so the efficiency of making electronic payrolls is low.

[0003] Currently, no effective solution has been proposed for the above technical problems. Summary of the invention

[0004] The present application provides a method, device, and server for generating a resource data bill, which can quickly generate a resource data bill.

[0005] The purpose of the embodiment of the present application is to provide a method for generating a resource data bill, including:

[0006] Obtaining an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values;

[0007] The original data table is processed using a first target classification model to obtain a first target classification result; wherein the first target classification result matches a preset template of the target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software;

[0008] Verifying the first target classification result to obtain a target verification result;

[0009] According to the target verification result, if it is determined that the target verification result is passed, filling the first target classification result into a preset template of the target software;

[0010] The target software is used to generate a resource data bill according to the preset template of the populated target software.

[0011] Furthermore, in another embodiment of the method, the using the first target classification model to process the original data table to obtain the first target classification result includes:

[0012] Using the first target classification model, the original classification items are divided into target classification items according to the semantic information of the original classification items; wherein the target classification items include: identity identification, basic information, income sub-items, expenditure sub-items, and resource data amount;

[0013] Combine the target classification item with the original data value corresponding to the target classification item to obtain a first target classification result; wherein the first target classification result includes: identity identification and its corresponding original data value, basic information and its corresponding original data value, income sub-items and its corresponding original data value, expenditure sub-items and its corresponding original data value, resource data amount and its corresponding original data value.

[0014] Furthermore, in another embodiment of the method, the first target classification model is obtained by: obtaining a training data set; wherein the training data set includes historical classification items and historical classification results corresponding to the historical classification items; and using the training data set to train an initial machine learning model to obtain the first target classification model.

[0015] Furthermore, in another embodiment of the method, the training data set also includes historical data values; the historical data values ​​have a corresponding relationship with the historical classification items;

[0016] The method further includes: extracting semantic features of the historical classification items and data features of the historical data values.

[0017] Furthermore, in another embodiment of the method, verifying the first target classification result to obtain a target verification result includes:

[0018] Performing format verification on the original data value corresponding to the identity identifier to obtain a first verification result;

[0019] When it is determined that the first verification result is passed, numerical verification is performed on the original data value corresponding to the income sub-item, the original data value corresponding to the expenditure sub-item, and the original data value corresponding to the resource data amount to obtain a second verification result;

[0020] When it is determined that the second verification result is passed, feature verification is performed on the original data value corresponding to the basic information, the original data value corresponding to the income sub-item, and the original data value corresponding to the expenditure sub-item to obtain a third verification result;

[0021] When it is determined that the third verification result is passed, the target verification result is determined to be passed.

[0022] Further, in another embodiment of the method, the performing of numerical verification on the original data values ​​corresponding to the income sub-items, the original data values ​​corresponding to the expenditure sub-items, and the original data values ​​corresponding to the resource data amounts to obtain a second verification result includes:

[0023] The original data values ​​corresponding to the income sub-items and the original data values ​​corresponding to the expenditure sub-items are summed up respectively to obtain the total income data value and the total expenditure data value;

[0024] The subtraction result of the total income item data value and the total expenditure item data value is taken as the first result;

[0025] Detecting whether the first result is equal to the original data value corresponding to the resource data amount;

[0026] When it is determined that the first result is equal to the original data value corresponding to the resource data amount, the second verification result is determined to be passed.

[0027] Furthermore, in another embodiment of the method, the feature verification is performed on the original data values ​​corresponding to the basic information, the original data values ​​corresponding to the income sub-items, and the original data values ​​corresponding to the expenditure sub-items to obtain a third verification result, including:

[0028] Extracting semantic features of original classification items and data features of original data values;

[0029] Calculate the semantic similarity between the semantic features of the original classification items and the semantic features of the historical classification items;

[0030] Determining a unique matching historical classification item for the original classification item according to the semantic similarity;

[0031] Check whether the data characteristics of the original data value corresponding to the original classification item and the data characteristics of the historical data value corresponding to the uniquely matched historical classification item are consistent;

[0032] When it is determined that the data feature of the original data value corresponding to the original classification item is consistent with the data feature of the historical data value corresponding to the uniquely matched historical classification item, the third verification result is determined to be passed.

[0033] Furthermore, in another embodiment of the method, the method further comprises:

[0034] When it is determined that the first verification result is not passed, and / or when it is determined that the second verification result is not passed, and / or when it is determined that the third verification result is not passed, sending a prompt message to the client;

[0035] receiving and executing the modification instruction sent by the client, reclassifying the original classification items and the original data values, and obtaining a second target classification result;

[0036] Filling the second target classification result into the preset template of the target software;

[0037] The target software is used to generate a resource data bill according to the preset template of the populated target software.

[0038] Furthermore, in another embodiment of the method, the method further comprises:

[0039] Adding the second target classification result to the training data set to obtain an expanded training data set;

[0040] The first target classification model is trained using the expanded training data set to obtain a second target classification model.

[0041] The purpose of the embodiment of the present application is to provide a device for generating a resource data bill, the device comprising:

[0042] An acquisition module, configured to acquire an original data table, wherein the original data table includes original data values ​​and original classification items corresponding to the original data values;

[0043] A classification module, used for processing the original data table using a first target classification model to obtain a first target classification result; wherein the first target classification result matches a preset template of a target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software;

[0044] A verification module, used to verify the first target classification result to obtain a target verification result;

[0045] A filling module, configured to fill the first target classification result into a preset template of the target software according to the target verification result, if it is determined that the target verification result is passed;

[0046] A generation module is used to generate a resource data bill using the target software according to a preset template of the filled target software.

[0047] The purpose of an embodiment of the present application is to provide a computer-readable storage medium having computer instructions stored thereon, and the computer-readable storage medium implements the above-mentioned method for generating a resource data bill when the instructions are executed by a processor.

[0048] The purpose of an embodiment of the present application is to provide a server, comprising at least one processor and a memory for storing processor executable instructions, wherein the processor implements the above-mentioned method for generating a resource data bill when executing the instructions.

[0049] Based on the method for generating a resource data bill provided by the present application, by obtaining an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values; using a first target classification model to process the original data table, obtaining a first target classification result; wherein the first target classification result matches a preset template of a target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software; the first target classification result is verified to obtain a target verification result; according to the target verification result, when it is determined that the target verification result is passed, the first target classification result is filled into the preset template of the target software; using the target software, a resource data bill is generated according to the preset template of the filled target software. Resource data bills can be generated efficiently and automatically to avoid manual operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of this specification, the drawings required for use in the embodiments will be briefly introduced below. The drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A schematic diagram of an embodiment of the process of a method for generating a resource data bill provided in an embodiment of this specification;

[0052] Figure 2 A flowchart of a method for generating a resource data bill in a specific scenario example provided in this specification;

[0053] Figure 3 A schematic diagram of the execution process of a system for generating a resource data bill provided in this specification;

[0054] Figure 4 A schematic diagram of the execution process of a system for generating a resource data bill provided in this specification;

[0055] Figure 5 A schematic diagram of an embodiment of a device for generating a resource data bill provided in an embodiment of this specification;

[0056] Figure 6A schematic diagram of an embodiment of the structure of a server provided in an embodiment of this specification. DETAILED DESCRIPTION

[0057] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0058] In the field of big data technology, electronic payroll is an important basis for online banks to pay wages on behalf of others, and is also an important business carrier for linking B-end and C-end customers. It plays a very important role in commercial banks' customer acquisition, retention, and activation. In the prior art, online banks provide customers with preset electronic payroll templates. If customers want to use online banks to generate electronic payrolls, they need to fill in the internal payroll data by column into the preset electronic payroll template, and then upload the completed electronic payroll template to the online bank's server. The online bank's server generates an electronic payroll based on the filled electronic payroll template. This process requires users to manually organize and match the payroll data, so the efficiency of making electronic payrolls is low and requires certain labor costs.

[0059] In view of the above-mentioned problems existing in the existing methods and the specific reasons for the above-mentioned problems, this application considers proposing a method for generating resource data bills to achieve efficient and rapid generation of resource data bills and reduce labor costs.

[0060] Based on the above ideas, this specification proposes a method for generating a resource data bill, which method includes: obtaining an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values; using a first target classification model to process the original data table to obtain a first target classification result; wherein the first target classification result matches a preset template of the target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software; verifying the first target classification result to obtain a target verification result; based on the target verification result, when it is determined that the target verification result is passed, filling the first target classification result into the preset template of the target software; using the target software, generating a resource data bill according to the preset template of the filled target software.

[0061] It should be noted that the information and data related to users involved in this application are obtained and used with the knowledge and consent of the users. In addition, the acquisition, storage, use and processing of the above information and data are in compliance with the relevant provisions of national laws and regulations.

[0062] See also Figure 1 As shown, this specification proposes a method for generating a resource data bill. In specific implementation, the method may include the following contents.

[0063] S101: Acquire an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values.

[0064] In some embodiments, the resource data bill is specifically an electronic payroll. The original data value is the data value required to make an electronic payroll. The original classification item is the header of the original data table, which indicates the data type of the original data value. For example, the original classification item can be: ID number, employee name, employee department, basic salary, bonus, withholding insurance, withholding provident fund, etc. The original data value is the specific numerical value corresponding to the original classification item. This application does not limit the type of original classification item.

[0065] In some embodiments, the original data values ​​and the original classification items constitute an original data table. The original data table is an Excel table in a custom format by the enterprise. In the prior art, if you want to generate a payroll based on the original data table, you need to manually fill in the original data values ​​in the preset template of the target software according to the format of the preset template of the target software. The target software is provided by the online bank, and the target software can generate an electronic payroll based on the filled preset template. The target software can only recognize the data format of the preset template, and cannot recognize the original Excel data table in the custom format of the enterprise.

[0066] In some embodiments, the data format of the preset template of the target software is: column 1 is identity identification, columns 2 to 25 are basic information, column 26 is resource data amount, columns 27 to 52 are income items, and columns 53 to 78 are expenditure items; the original data value of each employee is filled in by row, and the original data value of each employee occupies one row.

[0067] In some embodiments, the identity identifier is used to uniquely identify the identity of the employee and corresponds to the employee one-to-one. The identity identifier can be the employee's ID number or salary card number.

[0068] In some embodiments, the basic information is basic information that is not related to salary accounts, such as: employee department, employee length of service, salary payment date, employee number, etc.

[0069] In some embodiments, the income items are multiple income components in the salary, such as basic salary, performance salary, bonus, etc.

[0070] In some embodiments, the expense items are multiple expense parts in the salary, such as: withholding insurance, withholding provident fund, taxes, etc.

[0071] In some embodiments, the resource data amount is the actual amount of wages paid, and the resource data amount = the sum of the income sub-items - the sum of the expenditure sub-items.

[0072] S102: Process the original data table using a first target classification model to obtain a first target classification result; wherein the first target classification result matches a preset template of the target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software.

[0073] Using the first target classification model, multiple original classification items can be divided into five target classification items: identity identification, basic information, income sub-items, expenditure sub-items, and resource data amounts. Therefore, the original data table can be automatically adjusted to a format that matches the preset template of the target software, avoiding the manual matching and classification process.

[0074] In some embodiments, processing the original data table using the first target classification model to obtain a first target classification result specifically includes:

[0075] S1: using a first target classification model, according to the semantic information of the original classification items, the original classification items are divided into target classification items; wherein the target classification items include: identity identification, basic information, income sub-items, expenditure sub-items, and resource data amount;

[0076] S2: Combine the target classification item with the original data value corresponding to the target classification item to obtain a first target classification result; wherein the first target classification result includes: identity identification and its corresponding original data value, basic information and its corresponding original data value, income sub-items and its corresponding original data value, expenditure sub-items and its corresponding original data value, resource data amount and its corresponding original data value.

[0077] In some embodiments, the first target classification model is obtained by:

[0078] S1: Obtain a training data set; wherein the training data set includes historical classification items and historical classification results corresponding to the historical classification items;

[0079] S2: Use the training data set to train an initial machine learning model to obtain a first target classification model.

[0080] The following is an explanation of the method for dividing based on the first target classification model. First, a training data set is obtained. The training data set includes historical classification items. The historical classification items can specifically be identity card number, salary card number, employee department, employee length of service, salary payment date, employee number, withholding medical insurance, withholding pension insurance, withholding unemployment insurance, withholding provident fund, taxes, etc. The historical classification items should include as many data types of the original data values ​​as possible. The training data set also includes historical classification results, that is, the results after the historical classification items are divided into identity identification, basic information, income sub-items, expenditure sub-items, and resource data amounts. The initial machine learning model can extract the semantic features of multiple historical classification items, and construct the association between the semantic features of the historical classification items and the target classification items (identity identification, basic information, income sub-items, expenditure sub-items, and resource data amounts), and the first target classification model is obtained. After receiving the original classification items, the first target classification model extracts the semantic features of the original classification items, compares the semantic features of the original classification items with the semantic features included in the target classification items, and divides the original classification items into the target classification items with the closest semantic features. For example, the original classification item is withholding provident fund, the historical classification item is withholding housing provident fund, and the withholding housing provident fund is classified into the expenditure sub-item, so the expenditure sub-item has the semantic characteristics of withholding housing provident fund. In the comparison process, it is determined that the semantic characteristics of withholding provident fund are closest to the semantic characteristics of withholding housing provident fund included in the expenditure sub-item, so the withholding provident fund is classified into the expenditure sub-item, and the original data value corresponding to the withholding provident fund is also classified into the expenditure sub-item. Based on the above method, multiple original classification items can be divided into 5 target classification items (identity identification, basic information, income sub-item, expenditure sub-item, resource data amount), each original classification item can only be divided into one of identity identification, basic information, income sub-item, expenditure sub-item, resource data amount, and the original data value corresponding to the original classification item is also divided into the corresponding target classification item, so the target classification item and the original data value have a corresponding relationship; for some classification items that do not exist in the historical classification items, accurate classification can also be achieved according to their semantic characteristics.

[0081] In some embodiments, the training data set also includes historical data values; the historical data values ​​have a corresponding relationship with the historical classification items; the method also includes: extracting semantic features of the historical classification items and data features of the historical data values. For example, when the historical classification item is withholding housing provident fund, the historical data value of withholding housing provident fund should be greater than 0 and less than the upper limit of the deposit stipulated by the policy, so the data feature of the historical data value of withholding housing provident fund is its value range.

[0082] S103: Verify the first target classification result to obtain a target verification result.

[0083] In some embodiments, verifying the first target classification result to obtain a target verification result includes:

[0084] S1: Perform format verification on the original data value corresponding to the identity identifier to obtain a first verification result;

[0085] S2: When it is determined that the first verification result is passed, numerical verification is performed on the original data value corresponding to the income sub-item, the original data value corresponding to the expenditure sub-item, and the original data value corresponding to the resource data amount to obtain a second verification result;

[0086] S3: When it is determined that the second verification result is passed, feature verification is performed on the original data value corresponding to the basic information, the original data value corresponding to the income sub-item, and the original data value corresponding to the expenditure sub-item to obtain a third verification result;

[0087] S4: When it is determined that the third verification result is passed, determine that the target verification result is passed.

[0088] In some embodiments, since the original data value corresponding to the identity identifier is an ID card number or a salary card number, the number of digits of the digits is fixed, and the number of digits of the ID card number or the salary card number can be checked. For the ID card number, the 7th to 14th digits are combined into a time value, and the time value is located in the interval from 19000101 to the verification time, so the 7th to 14th digits of the ID card number can be format checked. For the salary card number, some of its digits are usually fixed, so it can be checked whether some of the salary card number are fixed digits. In the case where the identity identifier includes the ID card number, if the number of digits of the ID card number is equal to 18 and the time value composed of the 7th to 14th digits meets the preset conditions, then the first test result is determined to be passed. In the case where the identity identifier includes the salary card number, if the number of digits of the salary card number meets the preset conditions, then the first test result is determined to be passed.

[0089] In some embodiments, numerical verification is performed on the original data values ​​corresponding to the income sub-items, the original data values ​​corresponding to the expenditure sub-items, and the original data values ​​corresponding to the resource data amounts to obtain a second verification result, specifically including:

[0090] S1: Sum the original data values ​​corresponding to the income sub-items and the original data values ​​corresponding to the expenditure sub-items to obtain the total income data value and the total expenditure data value;

[0091] S2: taking the subtraction result of the total income item data value and the total expenditure item data value as the first result;

[0092] S3: Detect whether the first result is equal to the original data value corresponding to the resource data amount;

[0093] S4: When it is determined that the first result is equal to the original data value corresponding to the resource data amount, the second verification result is determined to be passed.

[0094] In some embodiments, the original data values ​​of N employees can be randomly selected, and the original data values ​​corresponding to the income sub-items of each employee can be summed to obtain N total income data values ​​of the N employees; the original data values ​​corresponding to the expenditure sub-items of each employee can be summed to obtain N total expenditure data values ​​of the N employees; the total income data values ​​of the N employees are subtracted from the total expenditure data values ​​to obtain N first results; it is detected whether the original data values ​​corresponding to the N first results and the N resource data amounts (i.e., the actual salary amounts paid) are equal; if they are equal, it means that there is no error in the division results of the income sub-items, expenditure sub-items, and resource data amounts in the first target classification results, so the second verification result is passed; if they are not equal, it means that there is an error in the division results of the income sub-items, expenditure sub-items, and resource data amounts in the first target classification results, and the second verification result is failed; wherein, N is less than the total number of employees, and N can be taken as 20% or 30% of the total number of employees. This application does not limit the value of N.

[0095] In some embodiments, feature verification is performed on the original data values ​​corresponding to the basic information, the original data values ​​corresponding to the income sub-items, and the original data values ​​corresponding to the expenditure sub-items to obtain a third verification result, which specifically includes:

[0096] S1: Extract the semantic features of the original classification items and the data features of the original data values;

[0097] S2: Calculate the semantic similarity between the semantic features of the original classification items and the semantic features of the historical classification items;

[0098] S3: Determine a unique matching historical classification item for the original classification item according to the semantic similarity;

[0099] S4: Detect whether the data features of the original data value corresponding to the original classification item and the data features of the historical data value corresponding to the uniquely matched historical classification item are consistent;

[0100] S5: When it is determined that the data feature of the original data value corresponding to the original classification item is consistent with the data feature of the historical data value corresponding to the uniquely matched historical classification item, the third verification result is determined to be passed.

[0101] In a specific scenario example, the original classification item is withholding housing provident fund, the semantic features of "withholding housing provident fund" are extracted, and the data features of the original data value of withholding housing provident fund are determined to be in the value range of [0,2000]; the semantic features of multiple historical classification items are calculated, and the semantic similarity between the semantic features of "withholding housing provident fund" and the semantic features of the historical classification items is calculated, the maximum value of the semantic similarity is screened out, and it is detected whether the maximum value of the semantic similarity exceeds the semantic similarity threshold. When it is determined that the maximum value of the semantic similarity exceeds the semantic similarity threshold, the historical classification item withholding housing provident fund corresponding to the maximum value of the semantic similarity is taken as the historical classification item that only matches withholding housing provident fund, indicating that withholding housing provident fund and withholding housing provident fund are synonyms; at the same time, the data features of the historical data values ​​corresponding to withholding housing provident fund are in the value range of [0,2000], which is consistent with the data features of the original data value of withholding housing provident fund, so it can be determined that it is correct to classify "withholding housing provident fund" into the same expenditure sub-item as "withholding housing provident fund". Of course, it should be noted that the data features need to reflect the distribution patterns of the original data values ​​or historical data values. Other features, such as average values, may also be used as data features. This application does not limit the types of data features.

[0102] In another specific scenario example, the original classification item is transportation subsidy, and the historical classification item closest to transportation subsidy is housing subsidy. Therefore, transportation subsidy can be classified into the income sub-item to which housing subsidy belongs. However, since the semantic similarity between the semantic features of transportation subsidy and housing subsidy does not exceed the semantic similarity threshold, transportation subsidy and housing subsidy are not synonyms, and there is no need to detect whether their data features are consistent. Only when the semantic similarity exceeds the semantic similarity threshold will the unique matching historical classification item be determined for the original classification item, and then their data features will be tested for consistency.

[0103] In some embodiments, if the data features of the original data values ​​corresponding to the plurality of original classification items are consistent with the data features of the historical data values ​​corresponding to the uniquely matched historical classification items, the third verification result is determined to be passed. If the data features of the original data values ​​corresponding to the original classification items are inconsistent with the data features of the historical data values ​​corresponding to the uniquely matched historical classification items, the third verification result is determined to be failed.

[0104] S104: Based on the target verification result, when it is determined that the target verification result is passed, the first target classification result is filled into a preset template of the target software.

[0105] In some embodiments, when it is determined that the target inspection result is passed, it indicates that the format of the first target classification result matches the preset template of the target software, and thus the first target classification result can be directly filled into the preset template of the target software.

[0106] S105: Using the target software, a resource data bill is generated according to the preset template of the filled target software.

[0107] In some embodiments, the target software is installed in the online banking server, and the online banking server can open an interface. The user uses the target software through the interface, and the target software generates a resource data bill (i.e., an electronic payslip) according to the preset template of the filled target software; specifically, an electronic payslip can be generated for each employee based on the identity identification; then, the electronic payslip is sent to the employee's reserved mailbox.

[0108] In some embodiments, the user can also send the filled preset template of the target software to the online banking server through the client. After the online banking server generates the electronic pay slip, it sends the electronic pay slip to the client; the user is specifically the financial personnel of the enterprise; the financial personnel then use the client to forward the electronic pay slip to the corresponding employee according to the mailbox reserved by the employee.

[0109] In some embodiments, the method further comprises:

[0110] S1: When it is determined that the first verification result is failed, and / or when it is determined that the second verification result is failed, and / or when it is determined that the third verification result is failed, sending a prompt message to the client;

[0111] S2: receiving and executing the modification instruction sent by the client, reclassifying the original classification items and the original data values, and obtaining a second target classification result;

[0112] S3: Filling the second target classification result into the preset template of the target software;

[0113] S4: Using the target software, a resource data bill is generated according to the preset template of the filled target software.

[0114] In some embodiments, if the first verification result is failed, and / or the second verification result is failed, and / or the third verification result is failed, it means that there is an error in the division result of the original data value and the original classification item using the first target classification model. At this time, an error prompt message is sent to the client held by the user, and the user is specifically a financial personnel of the enterprise; after the user receives the error prompt message, the user manually modifies the first target classification result and sends a modification instruction through the client; according to the modification instruction sent by the client, the original classification item and the original data value are re-divided to obtain a second target classification result; the second target classification result is filled into the preset template of the target software; and using the target software, a resource data bill is generated according to the preset template of the filled target software.

[0115] In some embodiments, the method further comprises:

[0116] S1: adding the second target classification result to the training data set to obtain an expanded training data set;

[0117] S2: Use the expanded training data set to train the first target classification model to obtain a second target classification model.

[0118] In some embodiments, the second target classification result can be added to the training data set, and the division results of the modified original classification items in the second target classification result can be used to expand the training data set. Then, the first target classification model can be trained using the expanded training data set to improve the classification accuracy of the first target classification model and obtain a second target classification model with more accurate classification.

[0119] In some embodiments, synonyms and near-synonyms of historical classification items may also be collected to expand the training data set and retrain the first target classification model.

[0120] In some embodiments, historical electronic payrolls that have been successfully generated may also be collected, and the classification results of the original data items in the historical electronic payrolls may be used to expand the training data set and retrain the first target classification model.

[0121] In some embodiments, after the original data table is obtained, data preprocessing may be performed on the original data values, for example, the time value in the original data value may be converted into a format specified by a preset template of the target software.

[0122] In some embodiments, the income sub-items in the preset template of the target software have a total of 26 columns. If the original data values ​​divided into the income sub-items in the original sub-items are less than 26 columns, the vacant positions can be filled with 0. If the original data values ​​divided into the income sub-items in the original sub-items exceed 26 columns, a prompt message can be sent to the user to remind the user to merge some original classification items to reduce the number of original classification items.

[0123] For a specific scenario example, see Figure 2 As shown, the resource data bill can be generated based on the following methods, including:

[0124] Step 1: Obtain an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values; the original classification table includes at least one of the employee's ID card number and salary card number;

[0125] Step 2: Use the first target classification model to process the original data table, divide the original classification items and their corresponding original data values ​​into different target data items, and obtain the first target classification result; the target data items include: identity identification, basic information, income sub-items, expenditure sub-items, and resource data amounts; the first target classification result includes: identity identification and its corresponding original data value, basic information and its corresponding original data value, income sub-items and its corresponding original data value, expenditure sub-items and its corresponding original data value, resource data amount and its corresponding original data value; the first target classification model is trained using the training data set;

[0126] Step 3: Verify the first target classification result to obtain a target verification result; wherein the verification step specifically includes: format verification of the original data value corresponding to the identity identifier; value verification of the original data value corresponding to the income sub-item, the original data value corresponding to the expenditure sub-item, and the original data value corresponding to the resource data amount; feature verification of the original data value corresponding to the basic information, the original data value corresponding to the income sub-item, and the original data value corresponding to the expenditure sub-item;

[0127] Step 4: Determine whether the target verification result is passed; if it is passed, jump to step 5; if it is not passed, jump to step 6-step 8;

[0128] Step 5: Fill the first target classification result into the preset template of the target software, use the target software to generate a resource data bill according to the preset template of the filled target software; push the resource data bill to the corresponding employee;

[0129] Step 6: Send an error prompt message to the client, reminding the user holding the client to manually modify the erroneous classification in the first target classification result; the user is specifically a financial staff of the enterprise;

[0130] Step 7: According to the modification instruction sent by the client, the wrong classification in the first target classification result is modified to obtain the second target classification result; a resource data bill is generated according to the second target classification result; and the resource data bill can be further pushed to the corresponding employee;

[0131] Step 8: Use the second target classification results to expand the training data set and retrain the first target classification model to improve the classification accuracy of the first target classification model.

[0132] For a specific scenario example, see Figure 3 As shown, based on the above-mentioned method for generating resource data bills, the present application also proposes a system for generating resource data bills, which includes: a big data acquisition module, a data storage module, a machine learning module, a classification verification module, an online editing module, and a payroll generation module. The big data acquisition module can read data from the database to obtain an original data table; the original data table includes original data values ​​and original classification items corresponding to the original data values. The big data acquisition module sends the original data table to the machine learning module. The data storage module can store a training data set, and the training data set includes historical classification items and historical classification results corresponding to the historical classification items. The machine learning module uses the training data set to train the initial machine learning model to obtain a first target classification model. The machine learning module uses the first target classification model to classify the original data table to obtain a first target classification result, and pushes the first target classification result to the classification verification module. The classification verification module verifies whether the first target push result is correctly classified: the format of the original data value corresponding to the identity identifier is verified to obtain the first verification result; the original data value corresponding to the income sub-item, the original data value corresponding to the expenditure sub-item, and the original data value corresponding to the resource data amount are numerically verified to obtain the second verification result; the original data value corresponding to the basic information, the original data value corresponding to the income sub-item, and the original data value corresponding to the expenditure sub-item are feature-verified to obtain the third verification result; when the first verification result, the second verification result, and the third verification result are all passed, the target verification result is determined to be passed. The payroll generation module fills in the preset template of the target software according to the first target classification result, and further uses the target software to generate a resource data bill. The payroll generation module can also push the resource data bill to the corresponding employee. See Figure 4As shown, when the target verification result is not passed, the first target classification result is pushed to the online editing module. The online editing module sends an error prompt message to the client, reminding the user holding the client to manually modify the erroneous classification in the first target classification result. The online editing module modifies the erroneous classification in the first target classification result according to the modification instruction sent by the client, obtains the second target classification result, and sends the second target classification result to the payroll generation module and the data storage module respectively. The data storage module uses the second target classification result to expand the training data set. The machine learning module can use the expanded training data set to train the first target classification model again. The payroll generation module fills in the preset template of the target software according to the second target classification result, and further uses the target software to generate a resource data bill. The payroll generation module can also push the resource data bill to the corresponding employees.

[0133] Based on the above-mentioned method for generating resource data bills, this specification also proposes an embodiment of a device for generating resource data bills, see Figure 5 As shown, the device for generating the resource data bill specifically includes the following modules: an acquisition module 501 , a classification module 502 , a verification module 503 , a filling module 504 , and a generation module 505 .

[0134] The acquisition module 501 is used to acquire an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values;

[0135] The classification module 502 is used to process the original data table using a first target classification model to obtain a first target classification result; wherein the first target classification result matches a preset template of the target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software;

[0136] A verification module 503 is used to verify the first target classification result to obtain a target verification result;

[0137] A filling module 504 is used to fill the first target classification result into the preset template of the target software according to the target verification result, if it is determined that the target verification result is passed;

[0138] The generating module 505 is used to generate a resource data bill using the target software according to the preset template of the filled target software.

[0139] In some embodiments, the classification module 502 is specifically used to use a first target classification model to divide the original classification items into target classification items according to the semantic information of the original classification items; wherein the target classification items include: identity identification, basic information, income sub-items, expenditure sub-items, and resource data amounts; the target classification items are combined with the original data values ​​corresponding to the target classification items to obtain a first target classification result; wherein the first target classification result includes: identity identification and its corresponding original data value, basic information and its corresponding original data value, income sub-items and its corresponding original data value, expenditure sub-items and its corresponding original data value, resource data amounts and its corresponding original data value.

[0140] In some embodiments, the verification module 503 is specifically used to perform a format check on the original data value corresponding to the identity identifier to obtain a first verification result; when it is determined that the first verification result is passed, a numerical check is performed on the original data value corresponding to the income sub-item, the original data value corresponding to the expenditure sub-item, and the original data value corresponding to the resource data amount to obtain a second verification result; when it is determined that the second verification result is passed, a feature check is performed on the original data value corresponding to the basic information, the original data value corresponding to the income sub-item, and the original data value corresponding to the expenditure sub-item to obtain a third verification result; when it is determined that the third verification result is passed, the target verification result is determined to be passed.

[0141] In some embodiments, the device is also used to obtain a first target classification model in the following manner: obtain a training data set; wherein the training data set includes historical classification items and historical classification results corresponding to the historical classification items; and use the training data set to train an initial machine learning model to obtain a first target classification model.

[0142] In some embodiments, the device is also used to: send a prompt message to the client when it is determined that the first verification result is failed, and / or when it is determined that the second verification result is failed, and / or when it is determined that the third verification result is failed; receive and execute modification instructions sent by the client, re-divide the original classification items and the original data values ​​to obtain a second target classification result; fill the second target classification result into the preset template of the target software; and use the target software to generate a resource data bill according to the preset template of the filled target software.

[0143] In some embodiments, the device is further used to: add the second target classification result to the training data set to obtain an expanded training data set; and use the expanded training data set to train the first target classification model to obtain a second target classification model.

[0144] It should be noted that the units, devices or modules described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. For the convenience of description, the above devices are described separately by functions divided into various modules. Of course, when implementing this specification, the functions of each module can be implemented in the same or more software and / or hardware, or the modules that implement the same function can be implemented by a combination of multiple sub-modules or sub-units. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0145] The embodiment of the present specification also provides a computer storage medium for a method for generating a resource data bill, wherein the computer storage medium stores computer program instructions, which, when executed by a processor, implement the following steps: obtaining an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values; processing the original data table using a first target classification model to obtain a first target classification result; wherein the first target classification result matches a preset template of a target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software; verifying the first target classification result to obtain a target verification result; based on the target verification result, when it is determined that the target verification result is passed, filling the first target classification result into the preset template of the target software; and using the target software to generate a resource data bill according to the preset template of the filled target software.

[0146] In this embodiment, the storage medium includes, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a cache, a hard disk (HDD), or a memory card. The memory may be used to store computer program instructions. The network communication unit may be an interface for network connection communication set in accordance with the standard specified by the communication protocol.

[0147] In this embodiment, the functions and effects specifically implemented by the program instructions stored in the computer storage medium can be explained in comparison with other implementations and will not be described in detail here.

[0148] The present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements: obtaining an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values; processing the original data table using a first target classification model to obtain a first target classification result; wherein the first target classification result matches a preset template of a target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software; verifying the first target classification result to obtain a target verification result; based on the target verification result, when it is determined that the target verification result is passed, filling the first target classification result into the preset template of the target software; using the target software, generating a resource data bill according to the preset template of the filled target software.

[0149] The present specification also provides a server, including a processor and a memory for storing processor executable instructions, and the processor can perform the following steps according to the instructions when implemented: obtaining an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values; processing the original data table using a first target classification model to obtain a first target classification result; wherein the first target classification result matches a preset template of a target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software; verifying the first target classification result to obtain a target verification result; according to the target verification result, when it is determined that the target verification result is passed, filling the first target classification result into the preset template of the target software; using the target software, generating a resource data bill according to the preset template of the filled target software.

[0150] In order to complete the above instructions more accurately, refer to Figure 6 As shown, the embodiment of this specification also provides another specific server, wherein the server includes a network communication port 601, a processor 602 and a memory 603, and the above structures are connected through internal cables so that each structure can perform specific data interaction.

[0151] The network communication port 601 can be specifically used to obtain an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values.

[0152] The processor 602 can be specifically used to process the original data table using a first target classification model to obtain a first target classification result; wherein the first target classification result matches the preset template of the target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software; the first target classification result is verified to obtain a target verification result; based on the target verification result, when it is determined that the target verification result is passed, the first target classification result is filled into the preset template of the target software; and the target software is used to generate a resource data bill according to the preset template of the filled target software.

[0153] The memory 603 may be specifically used to store corresponding instruction programs.

[0154] In this embodiment, the network communication port 601 can be a virtual port that is bound to different communication protocols so that different data can be sent or received. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. In addition, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM, CDMA, etc.; it can also be a Wifi chip; it can also be a Bluetooth chip.

[0155] In this embodiment, the processor 602 may be implemented in any appropriate manner. For example, the processor may take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) executable by the (micro)processor, a logic gate, a switch, an application specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, etc. This specification does not limit this.

[0156] In this embodiment, the memory 603 may include multiple levels. In a digital system, anything that can store binary data can be a memory; in an integrated circuit, a circuit with a storage function but no physical form is also called a memory, such as RAM, FIFO, etc.; in a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0157] Although the present specification provides method operation steps as described in the embodiments or flow charts, more or less operation steps may be included based on conventional or non-creative means. The order of steps listed in the embodiments is only one way of executing the order of many steps, and does not represent a unique execution order. When the device or client product in practice is executed, it can be executed in sequence or in parallel according to the method shown in the embodiments or the drawings (for example, a parallel processor or a multi-threaded processing environment, or even a distributed data processing environment). The term "include", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, product or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such a process, method, product or device. In the absence of more restrictions, it is not excluded that there are other identical or equivalent elements in the process, method, product or device including the elements. The first, second, etc. words are used to represent the name, and do not represent any particular order.

[0158] Those skilled in the art also know that, in addition to implementing the controller in a purely computer-readable program code, the controller can be made to implement the same function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, such a controller can be considered as a hardware component, and the devices for implementing various functions included therein can also be considered as structures within the hardware component. Or even, the devices for implementing various functions can be considered as both software modules for implementing the method and structures within the hardware component.

[0159] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media including storage devices.

[0160] Through the description of the above embodiments, it can be known that those skilled in the art can clearly understand that the present specification can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present specification can essentially be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a mobile terminal, a server, or a network device, etc.) to execute the methods described in each embodiment of the present specification or some parts of the embodiments.

[0161] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. This specification can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc.

[0162] Although the present specification is described through embodiments, those skilled in the art will appreciate that there are many modifications and changes to the present specification without departing from the spirit of the present specification, and it is intended that the appended claims include these modifications and changes without departing from the spirit of the present specification.

Claims

1. A method for generating a resource data bill, characterized in that: The method comprises: Obtaining an original data table; wherein the original data table includes original data values ​​and original classification items corresponding to the original data values; The original data table is processed using a first target classification model to obtain a first target classification result; wherein the first target classification result matches a preset template of the target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software; Verifying the first target classification result to obtain a target verification result; According to the target verification result, if it is determined that the target verification result is passed, filling the first target classification result into a preset template of the target software; The target software is used to generate a resource data bill according to the preset template of the populated target software.

2. The method according to claim 1, characterized in that Processing the original data table using the first target classification model to obtain a first target classification result includes: Using the first target classification model, the original classification items are divided into target classification items according to the semantic information of the original classification items; wherein the target classification items include: identity identification, basic information, income sub-items, expenditure sub-items, and resource data amount; Combine the target classification item with the original data value corresponding to the target classification item to obtain a first target classification result; wherein the first target classification result includes: identity identification and its corresponding original data value, basic information and its corresponding original data value, income sub-items and its corresponding original data value, expenditure sub-items and its corresponding original data value, resource data amount and its corresponding original data value.

3. The method according to claim 1, characterized in that The first target classification model is obtained by: Acquire a training data set; wherein the training data set includes historical classification items and historical classification results corresponding to the historical classification items; The training data set is used to train an initial machine learning model to obtain a first target classification model.

4. The method according to claim 3, characterized in that The training data set also includes historical data values; the historical data values ​​have a corresponding relationship with the historical classification items; The method further comprises: The semantic features of the historical classification items and the data features of the historical data values ​​are extracted.

5. The method according to claim 2, characterized in that: Verifying the first target classification result to obtain a target verification result includes: Performing format verification on the original data value corresponding to the identity identifier to obtain a first verification result; When it is determined that the first verification result is passed, numerical verification is performed on the original data value corresponding to the income sub-item, the original data value corresponding to the expenditure sub-item, and the original data value corresponding to the resource data amount to obtain a second verification result; When it is determined that the second verification result is passed, feature verification is performed on the original data value corresponding to the basic information, the original data value corresponding to the income sub-item, and the original data value corresponding to the expenditure sub-item to obtain a third verification result; When it is determined that the third verification result is passed, the target verification result is determined to be passed.

6. The method according to claim 5, characterized in that Performing numerical verification on the original data values ​​corresponding to the income sub-items, the original data values ​​corresponding to the expenditure sub-items, and the original data values ​​corresponding to the resource data amounts, and obtaining a second verification result, including: The original data values ​​corresponding to the income sub-items and the original data values ​​corresponding to the expenditure sub-items are summed up respectively to obtain the total income data value and the total expenditure data value; The subtraction result of the total income item data value and the total expenditure item data value is taken as the first result; Detecting whether the first result is equal to the original data value corresponding to the resource data amount; When it is determined that the first result is equal to the original data value corresponding to the resource data amount, the second verification result is determined to be passed.

7. The method according to claim 4 or 5, characterized in that: Perform feature verification on the original data values ​​corresponding to the basic information, the original data values ​​corresponding to the income sub-items, and the original data values ​​corresponding to the expenditure sub-items to obtain a third verification result, including: Extracting semantic features of original classification items and data features of original data values; Calculate the semantic similarity between the semantic features of the original classification items and the semantic features of the historical classification items; Determining a unique matching historical classification item for the original classification item according to the semantic similarity; Check whether the data characteristics of the original data value corresponding to the original classification item and the data characteristics of the historical data value corresponding to the uniquely matched historical classification item are consistent; When it is determined that the data feature of the original data value corresponding to the original classification item is consistent with the data feature of the historical data value corresponding to the uniquely matched historical classification item, the third verification result is determined to be passed.

8. The method according to claim 5, characterized in that The method further comprises: When it is determined that the first verification result is not passed, and / or when it is determined that the second verification result is not passed, and / or when it is determined that the third verification result is not passed, sending a prompt message to the client; receiving and executing the modification instruction sent by the client, reclassifying the original classification items and the original data values, and obtaining a second target classification result; Filling the second target classification result into the preset template of the target software; The target software is used to generate a resource data bill according to the preset template of the populated target software.

9. The method according to claim 3 or 8, characterized in that: The method further comprises: Adding the second target classification result to the training data set to obtain an expanded training data set; The first target classification model is trained using the expanded training data set to obtain a second target classification model.

10. A device for generating a resource data bill, characterized in that: The device comprises: An acquisition module, configured to acquire an original data table, wherein the original data table includes original data values ​​and original classification items corresponding to the original data values; A classification module, used for processing the original data table using a first target classification model to obtain a first target classification result; wherein the first target classification result matches a preset template of a target software; the target software is used to generate a resource data bill according to the preset template; the first target classification result includes: the original data value and a target classification item corresponding to the original data value; the target classification item is determined according to the preset template of the target software; A verification module, used to verify the first target classification result to obtain a target verification result; A filling module, configured to fill the first target classification result into a preset template of the target software according to the target verification result, if it is determined that the target verification result is passed; A generation module is used to generate a resource data bill using the target software according to a preset template of the filled target software.

11. A server, characterized in that: The invention comprises a processor and a memory for storing processor executable instructions, wherein the processor implements the steps of the method according to any one of claims 1 to 9 when executing the instructions.

12. A computer-readable storage medium, characterized in that: Computer instructions are stored thereon, and when the instructions are executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images