A slip data prediction method, device and medium

Abstract

The application discloses a kind of bill data prediction method, equipment and medium, method includes: in preset period, obtain the historical bill data of user in ERP system;Effective bill data is obtained by filtering historical bill data;According to a plurality of dimensions, feature classification is carried out to effective bill data, and grouping bill data is obtained;Fill single behavior information of user is predicted by grouping bill data and pre-constructed neural network model;According to fill single behavior information, bill data that user will fill is predicted.By obtaining the historical bill data of user in ERP system, based on historical bill data and pre-constructed neural network model, the fill single behavior information of user can be predicted, so that the bill data that will be filled predicted can be shown to user when user fills new bill, help user to fill quickly, to improve fill efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, and particularly relates to a bill data prediction method, device and medium. BACKGROUND

[0002] With the development of computer application technology, people have more and more demands for various applications, and various application systems need to store more and more data, such as bill applications. For bill applications, in the field of ERP (Enterprise Resourse Planning), filling in bills is a very common operation.

[0003] In an ERP system, as the complexity of various bill businesses increases, a user fills in a bill in a process that carries a large number of business fields, which is tedious and increases the user filling time, thereby causing the user to have a low filling efficiency when filling in a bill. SUMMARY

[0004] Embodiments of the present application provide a bill data prediction method, device and medium, which are used to solve the problem of low filling efficiency of a user when filling in a bill.

[0005] Embodiments of the present application adopt the following technical solutions:

[0006] In an aspect, the present application provides a bill data prediction method, which includes: obtaining historical bill data of a user in an ERP system in a preset period; filtering the historical bill data to obtain effective bill data; classifying features of the effective bill data according to a plurality of preset dimensions to obtain grouped bill data; predicting filling behavior information of the user by using the grouped bill data and a pre-constructed neural network model; and predicting bill data to be filled in by the user according to the filling behavior information.

[0007] In one example, the predicting filling behavior information of the user by using the grouped bill data and the pre-constructed neural network model specifically includes: performing discrete sampling processing on the grouped bill data to generate discrete bill data; performing permutation and combination on the discrete bill data to obtain permutation and combination bill data; inputting the permutation and combination bill data into the pre-constructed neural network model to generate the filling behavior information of the user.

[0008] In one example, before the predicting the filling behavior information of the user by the effective bill data and the pre-constructed neural network model, the method further comprises: obtaining sample bill data of a sample user; filtering the sample bill data to obtain effective sample bill data; classifying features of the effective sample bill data according to a plurality of preset dimensions to obtain grouped sample bill data; performing discrete sampling processing on the grouped sample bill data to generate discrete sample bill data; performing permutation and combination on the discrete sample bill data to obtain permutation and combination sample bill data; training an initial neural network model by taking the permutation and combination sample bill data as input and taking filling behavior information of the sample user as output until a training stop condition is reached to obtain the neural network model.

[0009] In one example, the performing discrete sampling processing on the grouped sample bill data to generate discrete sample bill data specifically comprises: determining a minimum feature value of a data dimension and a maximum feature value of the data dimension in each grouped sample bill data; determining average discrete sample bill data between the minimum feature value of the data dimension and the maximum feature value of the data dimension; and generating discrete sample bill data according to the average discrete sample bill data.

[0010] In one example, the filtering the historical bill data to obtain effective bill data specifically comprises: determining bill data that has been deleted by the user in the historical bill data according to a pre-set deletion flag; determining bill data with incomplete filling in the historical bill data; and filtering the bill data that has been deleted by the user and the bill data with incomplete filling in the historical bill data to obtain effective bill data.

[0011] In one example, the classifying features of the effective bill data according to a plurality of preset dimensions to obtain grouped bill data specifically comprises: determining that the plurality of preset dimensions include at least one of a time dimension, a location dimension, a business matter dimension, and a dimension of whether to exceed a standard; generating feature values corresponding to the plurality of dimensions respectively for the effective bill data; comparing feature values of each effective bill data under the same dimension in a plurality of effective bills to generate similarity between each effective bill data; and dividing effective bill data with a similarity lower than a preset similarity threshold into the same group of bill data.

[0012] In an example, the predicting the form data to be filled by the user according to the form filling behavior information specifically comprises: converting the form filling behavior information according to a preset user permission to generate a form filling behavior analysis report to be displayed to the user; receiving a report display request of the user and displaying the form filling behavior analysis report to the user; receiving a form filling request of the user, predicting the form data to be filled by the user according to the form filling behavior analysis report; and displaying the predicted form data to be filled by the user to the user.

[0013] In an example, after the predicting the form filling behavior information of the user by the grouped form data and the pre-constructed neural network model, the method further comprises: determining a prediction accuracy of the form filling behavior information based on feedback of the user; if the prediction accuracy is lower than a preset accuracy threshold, training the neural network model by using the historical form data; and updating the neural network model according to the trained neural network model.

[0014] In another aspect, an embodiment of the present application provides a form data prediction device, comprising: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: acquire historical form data of a user in an ERP system in a preset period; filter the historical form data to obtain effective form data; classify features of the effective form data according to a plurality of preset dimensions to obtain grouped form data; predict form filling behavior information of the user by using the grouped form data and a pre-constructed neural network model; and predict form data to be filled by the user according to the form filling behavior information.

[0015] In another aspect, an embodiment of the present application provides a form data prediction nonvolatile computer storage medium, which stores computer executable instructions, and the computer executable instructions are configured to: acquire historical form data of a user in an ERP system in a preset period; filter the historical form data to obtain effective form data; classify features of the effective form data according to a plurality of preset dimensions to obtain grouped form data; predict form filling behavior information of the user by using the grouped form data and a pre-constructed neural network model; and predict form data to be filled by the user according to the form filling behavior information.

[0016] The above at least one technical solution adopted by the embodiments of the present application can achieve the following beneficial effects:

[0017] By acquiring historical bill data of a user in an ERP system, based on the historical bill data and a pre-constructed neural network model, the filling behavior information of the user can be predicted, so that the user can be shown the predicted bill data to be filled in when the user fills a new bill, helping the user to quickly fill in the bill, so as to improve the filling efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the present application, some embodiments of the present application will be described in detail below with reference to the accompanying drawings, in which:

[0019] Figure 1 A flowchart of a bill data prediction method provided by an embodiment of the present application;

[0020] Figure 2 A structural diagram of a bill data prediction device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely below with reference to specific embodiments and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0022] Some embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0023] Figure 1 A flowchart of a bill data prediction method provided by an embodiment of the present application. The method can be applied to different business fields, such as the Internet financial business field, the e-commerce business field, the instant messaging business field, the game business field, the public service business field, etc. Some input parameters or intermediate results in the flow allow manual intervention to adjust to help improve accuracy.

[0024] The implementation of the analysis method related to the embodiments of the present application can be a terminal device or a server, and the present application does not make special limitations thereon. For the convenience of understanding and description, the following embodiments are described in detail with the server as an example.

[0025] It should be noted that the server can be a single device, or a system composed of multiple devices, i.e., a distributed server, and the present application does not make specific limitations thereon.

[0026] Figure 1 The flow in the above embodiment can include the following steps:

[0027] S101: Obtain historical bill data of a user in an ERP system in a preset period.

[0028] In some embodiments of the present application, before obtaining the historical bill data, the type of the historical bill data of the user needs to be determined. The types include private bills and public bills, such as reimbursement, application, loan, repayment, and the like for private bills, and salary, materials, assets, and the like for public bills.

[0029] Then, after obtaining the historical bill data, the bill type, business trip time, business trip location, whether it is over-standard, and the bill information such as the most business trip city, the longest business trip duration, and business trip during holidays need to be determined.

[0030] S102: Filter the historical bill data to obtain valid bill data.

[0031] In some embodiments of the present application, first, according to a preset deletion flag, such as a bill with a deletion flag of "yes". In the historical bill data, the bill data that has been deleted by the user is determined, thereby reducing the noise interference of the deleted data.

[0032] Then, in the historical bill data, the bill data with incomplete filling is determined. For example, for a travel type reimbursement bill, a complete trip bill can be set, that is, a bill with non-empty departure city, arrival city, departure time, and arrival time of the trip details.

[0033] Finally, in the historical bill data, the bill data that has been deleted by the user and the bill data with incomplete filling are filtered to obtain valid bill data, ensuring the validity of the bill data.

[0034] S103: According to a plurality of preset dimensions, the valid bill data is classified by features to obtain grouped bill data.

[0035] In some embodiments of the present application, a plurality of dimensions need to be determined first, such as time dimension, location dimension, business transaction dimension, and over-standard dimension, that is, user time, user location, user frequency, and user over-standard feature dimensions.

[0036] Then, the feature values corresponding to the valid bill data in the plurality of dimensions are generated.

[0037] Then, in the plurality of valid bills, the feature values of each valid bill data in the same dimension are compared to generate the similarity between each valid bill data.

[0038] Finally, the valid bill data with a similarity lower than a preset similarity threshold is divided into the same group of bill data, thereby predicting the filling behavior of the user in the plurality of dimensions through the plurality of dimension feature subdivisions.

[0039] S104: Predict the filling behavior information of the user by the grouped bill data and the pre-constructed neural network model.

[0040] In some embodiments of the present application, the grouped bill data needs to be first discretely sampled to generate discrete bill data. During the discrete sampling process, the minimum feature value of the data dimension and the maximum feature value of the data dimension are determined in each grouped bill data. Then, the average discrete bill data between the minimum feature value of the data dimension and the maximum feature value of the data dimension is determined, and the discrete bill data is generated according to the average discrete bill data.

[0041] Then, the discrete bill data is arranged and combined to obtain arranged and combined bill data.

[0042] Finally, the arranged and combined bill data is input into the pre-constructed neural network model to generate the filling behavior information of the user, realizing the prediction of the user's filling behavior in each dimension.

[0043] During the construction of the neural network model, the sample bill data of the sample user is first obtained, and then the sample bill data is filtered to obtain effective sample bill data.

[0044] Then, according to the preset multiple dimensions, the effective sample bill data is classified by features to obtain grouped sample bill data.

[0045] Then, the grouped sample bill data is discretely sampled to generate discrete sample bill data. During the discrete sampling process, the minimum feature value of the data dimension and the maximum feature value of the data dimension are determined in each grouped sample bill data. Then, the average discrete sample bill data between the minimum feature value of the data dimension and the maximum feature value of the data dimension is determined, and the discrete sample bill data is generated according to the average discrete sample bill data.

[0046] Then, the discrete sample bill data is arranged and combined to obtain arranged and combined sample bill data.

[0047] Finally, the arranged and combined sample bill data is input, and the filling behavior information of the sample user is output. The initial neural network model is trained until the training stopping condition is reached, and the neural network model is obtained.

[0048] The arranged and combined data is trained by machine, and the machine learning algorithm is used to enhance the bill statistical information and reduce noise interference, greatly improving the accuracy of the prediction of the user's filling behavior habit.

[0049] S105: According to the filling behavior information, predict the bill data to be filled by the user.

[0050] In some embodiments of the present application, the filling behavior information is converted according to the preset user rights to generate a filling behavior analysis report for the user, the report display request of the user is received, and the filling behavior analysis report is displayed to the user, so that the behavior analysis report of the related personal data dynamically generated for the machine training result is realized.

[0051] It should be noted that, for the user rights, the data related to the analysis report that the user can see is dynamically displayed by setting the function, and for the report display mode, the report background picture is set to adjust the display style of the related data, so that the analysis data of the user behavior habit is more intuitively displayed.

[0052] In addition, when the user fills out the form, the filling request of the user is received, the form data to be filled out by the user is predicted according to the filling behavior analysis report, so that the form data to be filled out by the user is displayed to the user, and after the generated user filling behavior analysis report, the user fills out the form again, and the content information of the form to be filled out is inversely predicted according to the report, which is more quickly and intuitively displayed to the user, and the user can quickly fill out the form.

[0053] In an embodiment of the present application, the prediction accuracy of the filling behavior information is determined based on the feedback of the user.

[0054] If the prediction accuracy is lower than the preset accuracy threshold, the neural network model is trained by the historical form data.

[0055] Finally, the neural network model is updated according to the trained neural network model, so that with the increase of the historical forms, the prediction analysis report will be more accurate to feedback the common behavior of the user in the filling process after the update and training of the neural network model.

[0056] It should be noted that, although the embodiments of the present application are introduced and described in sequence with reference to steps S101 to S105, this does not mean that steps S101 to S105 must be executed in strict sequence. The embodiments of the present application introduce and describe steps S101 to S105 in sequence as shown in the method 1000, in order to facilitate the understanding of the technical solution of the embodiments of the present application by those skilled in the art. In other words, in the embodiments of the present application, the sequence between steps S101 to S105 can be appropriately adjusted according to actual needs. Figure 1 Figure 1 The sequence between steps S101 to S105 can be appropriately adjusted according to actual needs.

[0057] The sequence between steps S101 to S105 can be appropriately adjusted according to actual needs. Figure 1 ​The method comprises the following steps: obtaining historical document data of a user in an ERP system, and predicting filling behavior information of the user based on the historical document data and a pre-constructed neural network model, so as to predict the user's common behavior habits, display predicted document data to be filled by the user to the user when the user fills a new document, help the user fill the document quickly, improve the filling efficiency, and generate a filling behavior analysis report to enable the user to intuitively and clearly understand the filling habits, confirm whether the behavior habits need to be adjusted according to company standards, and improve the user experience.

[0058] Based on the same idea, some embodiments of the present application also provide a device and a non-volatile computer storage medium corresponding to the above method.

[0059] Figure 2 A structural schematic diagram of a document data prediction device provided by an embodiment of the present application comprises the following:

[0060] at least one processor; and

[0061] a memory in communication connection with the at least one processor; wherein

[0062] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0063] obtain historical document data of a user in an ERP system within a preset period;

[0064] filter the historical document data to obtain valid document data;

[0065] classify the valid document data according to a plurality of preset dimensions to obtain grouped document data;

[0066] predict filling behavior information of the user by using the grouped document data and a pre-constructed neural network model;

[0067] predict document data to be filled by the user according to the filling behavior information.

[0068] A non-volatile computer storage medium provided by some embodiments of the present application stores computer executable instructions, and the computer executable instructions are set to:

[0069] obtain historical document data of a user in an ERP system within a preset period;

[0070] filter the historical document data to obtain valid document data;

[0071] According to a plurality of preset dimensions, the effective bill data is classified by features to obtain grouped bill data;

[0072] By the grouped bill data and a pre-constructed neural network model, the filling behavior information of the user is predicted.

[0073] According to the filling behavior information, bill data to be filled by the user is predicted.

[0074] Each of the embodiments in the present application is described in a progressive manner, and the same and similar parts of each of the embodiments can be referred to each other. Each of the embodiments mainly describes the difference from other embodiments. In particular, for the device and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the description of the method embodiments.

[0075] The device and medium provided by the embodiments of the present application are one-to-one corresponding to the method, and therefore, the device and medium also have the similar beneficial technical effects as the method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the device and medium will not be described here.

[0076] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0077] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The function of one flow or multiple flows and / or blocks Figure 1 The device that implements the function specified in one block or multiple blocks.

[0078] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0079] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0080] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0081] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, such as read-only memory (ROM), EPROM, and / or flash memory, etc. The memory is an example of computer readable media.

[0082] Computer readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to computing devices. According to the definition herein, computer readable media does not include transitory media, such as modulated data signals and carrier waves.

[0083] It should also be noted that the terms "comprising", "comprises", "including", "includes" or any other variation thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a... " does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0084] The above description is only the preferred embodiment of the present application, and is not intended to limit the present application. The present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the technical principles of the present application shall fall within the protection scope of the present application.

Claims

1. A document data prediction method, characterized in that: The method comprises: Obtain the user's historical document data in the ERP system within a preset period; Filtering the historical document data to obtain valid document data; According to a plurality of preset dimensions, the valid document data is subjected to feature classification to obtain grouped document data; Predicting the user's form filling behavior information through the grouped document data and a pre-built neural network model; Predicting the document data that the user will fill out based on the form filling behavior information; The method of predicting the user's form filling behavior information by using the grouped form data and a pre-built neural network model specifically includes: performing discrete sampling processing on the grouped document data to generate discrete document data; Performing permutation and combination on the discrete document data to obtain permutation and combination document data; Inputting the permutation and combination document data into the pre-built neural network model to generate the user's form filling behavior information; Before predicting the user's form filling behavior information using the valid document data and the pre-built neural network model, the method further includes: Get sample document data of sample users; Filtering the sample document data to obtain valid sample document data; According to a plurality of preset dimensions, the valid sample document data is subjected to feature classification to obtain grouped sample document data; Performing discrete sampling processing on the grouped sample document data to generate discrete sample document data; Performing permutations and combinations on the discrete sample document data to obtain permutation and combination sample document data; The permutation and combination sample document data is used as input, and the sample user's form filling behavior information is used as output, and the initial neural network model is trained until the training stop condition is reached, thereby obtaining the neural network model; The feature classification of the valid document data according to the preset multiple dimensions to obtain the grouped document data specifically includes: Determine at least one of the preset multiple dimensions including a time dimension, a location dimension, a business matter dimension, and whether a dimension exceeds a standard; Generating characteristic values ​​corresponding to the valid document data in the multiple dimensions respectively; Among multiple valid documents, the feature values ​​of each valid document data under the same dimension are compared to generate the similarity between each valid document data; Classify valid document data with similarity lower than a preset similarity threshold into the same group of document data; The step of predicting the document data to be filled out by the user based on the form filling behavior information specifically includes: According to the preset user permissions, the form filling behavior information is converted to generate a form filling behavior analysis report displayed to the user; receiving a report display request from the user, and displaying the form filling behavior analysis report to the user; receiving a form filling request from the user, and predicting the form data to be filled in by the user based on the form filling behavior analysis report; The document data that is predicted to be filled out by the user is displayed to the user.

2. The method according to claim 1, characterized in that The discrete sampling process is performed on the grouped sample document data to generate discrete sample document data, specifically comprising: In each grouped sample document data, determine the minimum eigenvalue and the maximum eigenvalue of the data dimension; Determine the average discrete sample document data between the minimum eigenvalue of the data dimension and the maximum eigenvalue of the data dimension; Discrete sample document data is generated according to the average discrete sample document data.

3. The method according to claim 1, characterized in that The filtering of the historical document data to obtain valid document data specifically includes: According to a preset deletion flag, determining the document data that has been deleted by the user in the historical document data; Determining incomplete document data from the historical document data; In the historical document data, the document data that has been deleted by the user and the incomplete document data are filtered to obtain valid document data.

4. The method according to claim 1, wherein After predicting the user's form filling behavior information by using the grouped document data and the pre-built neural network model, the method further includes: Determining the prediction accuracy of the form filling behavior information based on user feedback; If the prediction accuracy is lower than a preset accuracy threshold, the neural network model is trained using the historical document data; The neural network model is updated according to the trained neural network model.

5. A document data prediction device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: execute a document data prediction method as described in any one of claims 1 to 4 above.

6. A document data prediction non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured to execute a document data prediction method as described in any one of claims 1 to 4 above.

Images