CN116029852B - Method and device for confirming reimbursement bill accounting subjects - Google Patents

Abstract

The embodiment of the application relates to a method and a device for confirming an account of a reimbursement bill, which comprise the steps of obtaining historical reimbursement bill data and corresponding account accounting data; training to obtain a reimbursement bill accounting subject classification model according to all or part of historical reimbursement bill data and corresponding accounting subject accounting data; and processing the target reimbursement bill data by using the reimbursement bill accounting bill classification model to obtain accounting bill accounting data corresponding to the target reimbursement bill data. The application can realize automatic accounting and entry of accounting contents in the reimbursement application form, and the generated reimbursement form accounting subject classification model is a machine learning model formed by using big data of which the accounting is completed by the units, so that the application can be more flexibly suitable for accounting rules of different scales and different types of units, lighten the workload of financial staff, and effectively avoid the compliance risk possibly caused by misselection or misfilling of the accounting by the financial staff of the units.

Description

Method and device for confirming reimbursement bill accounting subjects

Technical Field

The application relates to the technical field of electronic reimbursement certificate management, in particular to a reimbursement bill accounting subject confirmation method and device.

Background

It is self-evident that, with the continuous development of computer technology, a variety of financial accounting software has appeared, which can simplify the work content of financial management to a certain extent and lighten the workload of financial staff.

In the prior art, because different units conduct different activities, corresponding accounting subjects and accounting rules are different, when accounting is performed by financial staff of a certain unit, accounting subjects can be recorded into financial accounting software only after accounting subjects are confirmed according to specified accounting rules by manpower. With the continuous increase of unit operation activities, the phenomenon that personnel change occurs in the unit production process can also occur frequently, more and more reimbursement sheets can occur, so that the work content of financial management can be more and more complex and tedious, and the workload is also continuously increased. Therefore, in the prior art, due to the fact that the purpose of accounting needs to be confirmed manually, the problem that the accounting efficiency and the accounting accuracy are possibly reduced is caused.

Disclosure of Invention

In view of the above, one of the technical problems to be solved by the embodiments of the present application is to provide a method and a device for confirming account of reimbursement bill, which are used for solving the problem that in the prior art, due to the need of manually confirming account, the financial account-making efficiency and the financial accounting accuracy are likely to be reduced.

The first aspect of the embodiment of the application discloses a reimbursement bill accounting subject confirmation method, which comprises the following steps: acquiring historical reimbursement bill data and corresponding accounting subject accounting data; the historical reimbursement single data is used for recording accounting contents of at least one completed accounting in the reimbursement application form; the accounting subject billing data is used for recording a billing entry corresponding to the billing content after billing is completed;

training to obtain a reimbursement bill accounting subject classification model according to all or part of the historical reimbursement bill data and the corresponding accounting subject accounting data;

processing target reimbursement bill data by using the reimbursement bill accounting bill classification model to obtain accounting bill data corresponding to the target reimbursement bill data; the target reimbursement data are used for recording accounting contents of at least one incomplete accounting in the reimbursement application form.

The second aspect of the embodiment of the application discloses a reimbursement sheet accounting subject confirmation device, which comprises:

the acquisition module is used for acquiring historical reimbursement bill data and corresponding accounting subject accounting data; the historical reimbursement single data is used for recording accounting contents of at least one completed accounting in the reimbursement application form; the accounting subject billing data is used for recording a billing entry corresponding to the billing content after billing is completed;

The model training module is used for training to obtain a reimbursement bill accounting subject classification model according to all or part of the historical reimbursement bill data and the corresponding accounting subject accounting data;

the accounting module is used for processing the target reimbursement bill data by utilizing the reimbursement bill accounting subject classification model to obtain accounting subject accounting data corresponding to the target reimbursement bill data; the target reimbursement data are used for recording accounting contents of at least one incomplete accounting in the reimbursement application form.

Compared with the prior art, after the embodiment of the application trains to obtain the reimbursement bill accounting classification model according to the historical reimbursement bill data and the corresponding accounting bill accounting data, the reimbursement bill accounting data can be processed by using the reimbursement bill accounting bill classification model to obtain the accounting bill accounting data corresponding to the target reimbursement bill data, so that the automatic accounting and recording of the accounting contents in the reimbursement application bill is realized. Because the generated reimbursement bill accounting subject classification model is a machine learning model formed by using big data of the completed accounting of the units, the accounting rules of different scales and different types of units can be flexibly adapted, the workload of financial staff is lightened, and the compliance risk possibly caused by the wrong selection or the wrong filling of the accounting staff of the units can be effectively avoided.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for validating a reimbursement sheet accounting subject in accordance with an embodiment of the present application;

FIG. 2 is a flow chart of a method for validating a reimbursement bill accounting subject disclosed in the second embodiment of the application;

fig. 3 is a schematic block diagram of a confirmation device for reimbursement sheets accounting subjects according to a third embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

It should be noted that the terms "first," "second," "third," and "fourth," etc. in the description and claims of the present application are used for distinguishing between different objects and not for describing a particular sequential order. The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, apparatus, article, or device that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or device.

Example 1

Referring to fig. 1, fig. 1 is a schematic flowchart of a method for confirming a reimbursement bill accounting subject according to an embodiment of the present application, the method for confirming a reimbursement bill accounting subject includes:

step S101, a plurality of historical reimbursement bill data and corresponding accounting data are obtained.

In this embodiment, the historical reimbursement data is used to record the accounting contents of at least one completed accounting in the reimbursement bill. The reimbursement application form is recorded by a manager through manual or electronic automation, the content is not limited, reasonable setting can be carried out according to the requirements of units in practical application, but the required accounting content for accounting by the units is needed, namely, the accounting content in the reimbursement application form can be all the content in the reimbursement application form or part of the content in the reimbursement application form.

For example, the content identified by the reimbursement application form may include reimbursement person, department to which the reimbursement person belongs, reimbursement event, and time frame, reimbursement amount, type of voucher, settlement means, and source of expense to which the reimbursement event corresponds. If the reimbursement application form needs to be examined and approved for billing, the content identified by the reimbursement application form can also comprise an auditor, the job title of the auditor, the audit opinion, the audit time and the like. When the accounting is performed, the accounting entry corresponding to the accounting content can be determined according to the department of the reimbursement person, the reimbursement principal, the reimbursement amount, the voucher type, the settlement mode and the expense source, namely, the department of the reimbursement person, the reimbursement principal, the reimbursement amount, the voucher type, the settlement mode and the expense source belong to the accounting content in the reimbursement application form.

In this embodiment, the accounting data is used for recording an accounting entry corresponding to the accounting content after accounting is completed, one or more levels may be included in the accounting entry, one or more categories may be included in each level, the specific structure of the accounting entry is not limited, and in practical application, the accounting entry may be reasonably set according to the needs of the unit. In addition, a reimbursement application form may correspond to a plurality of reimbursement vouchers, which may be recorded in different categories when accounting for a reimbursement application form.

In this embodiment, the data size of the historical reimbursement bill data is not limited, but in order to improve the accuracy of data processing by using the reimbursement bill accounting subject classification model, in this step, as much historical reimbursement bill data as possible can be selected, that is, as many reimbursement bill applications with accounting completed are obtained as much as possible, and the accounting contents in different reimbursement bill with accounting completed are different.

Alternatively, in order to facilitate the extraction and processing of the relevant information, the historical reimbursement sheet data may be preferably structured data. For example, the accounting contents in the reimbursement application form may be recorded in an XML format.

Step S102, training to obtain a reimbursement bill accounting subject classification model according to all or part of historical reimbursement bill data and corresponding accounting subject accounting data.

In this embodiment, the reimbursement bill accounting subject classification model is a machine learning model, and the input of the model is historical reimbursement bill data or information used for recording accounting contents in reimbursement bill data, and the output is accounting records corresponding to the accounting contents. After the historical reimbursement bill data and the corresponding accounting data are obtained, a reimbursement bill accounting classification model can be obtained by training through a preset machine learning algorithm according to all or part of the historical reimbursement bill data and the corresponding accounting data. The type of the preset machine learning algorithm is not limited. For example, at least one of a non-linear support vector machine algorithm, a Bayesian classifier algorithm, a K-nearest neighbor algorithm, etc., or any other machine-learned classification algorithm model may be used.

Optionally, in order to improve accuracy of data processing by using the reimbursement bill accounting object classification model, only a part of historical reimbursement bill data and corresponding accounting object accounting data can be selected as a training data set to be respectively used as input and output of the reimbursement bill accounting object classification model, and training is performed to obtain the reimbursement bill accounting object classification model; and selecting a part of historical reimbursement bill data and corresponding accounting account numbers as a test data set, and carrying out test verification on the reimbursement bill accounting account classification model obtained through training so as to evaluate the output accuracy of the reimbursement bill accounting account classification model obtained through training. When the output accuracy of the reimbursement bill accounting object classification model does not reach the preset standard, new historical reimbursement bill data and corresponding accounting object accounting data are added, and training and accuracy assessment are continuously carried out on the reimbursement bill accounting object classification model until the preset standard is reached.

And step S103, processing the target reimbursement bill data by using the reimbursement bill accounting bill classification model to obtain accounting bill data corresponding to the target reimbursement bill data.

In this embodiment, the target reimbursement data is used for recording accounting contents of at least one incomplete accounting in the reimbursement application form, and the accounting data of the accounting subjects corresponding to the accounting contents in the reimbursement application form which is not yet accounted can be automatically determined by using the reimbursement bill accounting subject classification model obtained by training, so that the accounting subject accounting data corresponding to the target reimbursement application form data is obtained.

In this embodiment, the processing manner of the accounting data corresponding to the target reimbursement single data is not limited, and can be reasonably selected according to the actual application requirement. For example, accounting can be automatically completed according to accounting data corresponding to the target reimbursement data; accounting data corresponding to the target reimbursement data can also be provided for financial staff, and accounting can be automatically completed after the accounting data is confirmed by the financial staff, or manually recorded by the financial staff for accounting.

Optionally, it is contemplated that automated validation of accounting records using a reimbursement bill accounting category model may be subject to a certain amount of error, particularly when the amount of historical reimbursement bill data and corresponding accounting data trained thereon is less, the likelihood of error may be higher. Thus, in order to ensure the accuracy of the billing result, a step of manual confirmation may be added. Specifically, step S103 may further include:

and E1, processing the target reimbursement bill data by using a reimbursement bill accounting subject classification model to obtain machine classification data corresponding to the target reimbursement bill data.

The machine classification data is data obtained after automatic processing by using a reimbursement bill accounting subject classification model, namely, the machine classification data is used for recording accounting records corresponding to accounting contents obtained by processing the reimbursement bill accounting subject classification model. The machine classification data corresponding to all or part of the target reimbursement data can be displayed to the user by utilizing a preset man-machine interaction interface, and the user confirms all or part of the content and sends corresponding confirmation instructions.

And E2, obtaining accounting data corresponding to the target reimbursement data according to a confirmation instruction of the user to the machine classification data corresponding to the target reimbursement data.

According to the confirmation instruction of the machine classification data corresponding to the target reimbursement single data sent by the user, whether the machine classification data corresponding to the target reimbursement single data obtained after the automatic processing is required to be adjusted or not is determined, if so, according to the confirmation instruction of the machine classification data corresponding to the target reimbursement single data sent by the user, the machine classification data corresponding to the target reimbursement single data is adjusted, and accounting subject accounting data corresponding to the target reimbursement single data is obtained; if not, the machine classification data corresponding to the target reimbursement bill data is used as accounting subject accounting data corresponding to the target reimbursement bill data.

Further, considering that in practical application, when the data amount for training the reimbursement bill accounting subject classification model is enough, the automatic confirmation accuracy of the reimbursement bill accounting subject classification model for accounting and recording can be improved to be very high, and at the stage, if the reimbursement bill accounting subject classification model is fully checked manually, the efficiency is lower. Therefore, in order to improve the billing efficiency, the billing content of the billing entry or the billing content with lower accuracy judgment can not be determined only for the billing account classification model of the reimbursement bill, and the user can confirm again.

Specifically, sub-step E1 includes: and processing the target reimbursement bill data by using the reimbursement bill accounting subject classification model to obtain machine classification data and machine unclassified data corresponding to the target reimbursement bill data. Wherein the machine unclassified data is used to record billing content for which the reimbursement sheet accounting subject classification model is indeterminate of classification.

Correspondingly, sub-step E2 comprises: and providing the unclassified data of the machine corresponding to the target reimbursement single data for the user to confirm. And according to a confirmation instruction of the user to the unclassified data of the machine corresponding to the target reimbursement single data, acquiring accounting subject accounting data corresponding to the target reimbursement single data.

According to a confirmation instruction of a user on unclassified data of a machine corresponding to target reimbursement bill data, determining an accounting entry corresponding to accounting content which cannot be classified by an accounting classification model, obtaining manual classification data corresponding to the target reimbursement bill data, and taking the manual classification data and the machine classification data corresponding to the target reimbursement bill data together as corresponding accounting account data.

In addition, when the target reimbursement data does not have corresponding machine unclassified data, the machine classified data corresponding to the target reimbursement data can be directly used as corresponding accounting account data.

Further, in order to avoid the risk of compliance due to classification errors of the reimbursement bill accounting object classification model, billing content with lower classification accuracy of the reimbursement bill accounting object classification model may be determined as machine unclassified data. Specifically, sub-step E1 may further comprise:

and classifying the target reimbursement single data by using a reimbursement single accounting subject classification model to obtain preliminary classification data and classification accuracy evaluation data corresponding to the target reimbursement single data.

The primary classification data is used for recording an accounting entry of which the accounting content is divided by the account classification model of the reimbursement bill; the classification accuracy evaluation data is used for recording accuracy evaluation values of each account content classified by the reimbursement bill accounting subject classification model.

When the accuracy evaluation value of the accounting content is larger than or equal to a preset evaluation threshold value, determining preliminary classification data corresponding to the accounting content as machine classification data; when the accuracy evaluation value of the billing content is smaller than a preset evaluation threshold value, the billing content is determined as machine unclassified data.

The evaluation threshold can be reasonably set according to actual demands of units, so that the demand of different types of units can be better met by training to obtain the reimbursement bill accounting subject classification model.

Optionally, after step S103, the method may further include: and updating the account classification model of the reimbursement bill according to the account accounting data corresponding to the target reimbursement bill data.

Along with the increasing number of reimbursement application forms after accounting, namely, the continuously accumulation of historical reimbursement form data and corresponding accounting subject accounting data, the classification accuracy of accounting records corresponding to the accounting contents determined by using the reimbursement form accounting subject classification model can be improved continuously.

According to the embodiment of the application, after the account sorting model of the reimbursement bill is obtained through training according to the historical reimbursement bill data and the corresponding account accounting data, the target reimbursement bill data can be processed by using the account sorting model of the reimbursement bill to obtain the account accounting data corresponding to the target reimbursement bill data, so that the automatic accounting and entry of the accounting contents in the reimbursement application bill is realized. Because the generated reimbursement bill accounting subject classification model is a machine learning model formed by using big data of the completed accounting of the units, the accounting rules of different scales and different types of units can be flexibly adapted, the workload of financial staff is lightened, and the compliance risk possibly caused by the wrong selection or the wrong filling of the accounting staff of the units can be effectively avoided.

Example two

As shown in fig. 2, fig. 2 is a schematic flowchart of a reimbursement sheet accounting subject confirmation method according to a second embodiment of the application, where the reimbursement sheet accounting subject confirmation method includes:

step S201, the first public key is utilized to verify the first signature information in the first reimbursement application form data sent by the first user side.

In this embodiment, the first user terminal may be an intelligent terminal used by a reimbursement manager or an intelligent terminal used by an approver, which is not limited herein. However, in order to facilitate understanding of the technical solution of the present embodiment, the first user terminal may be considered as an intelligent terminal used by the reimbursement sponsor.

In this embodiment, the first reimbursement application form data may further include reimbursement information in addition to the first signature information. Because different units may have different requirements on the information required by reimbursement, the specific content and the generation mode included in the reimbursement information are not limited, and the reimbursement information can be reasonably selected according to the actual application requirements. However, in general, reimbursement information is used at least for recording reimbursement persons, approvers, reimbursement amounts, and reimbursement issues.

In this embodiment, the first signature information is generated by signing the reimbursement information with the first private key by the first user side, and in order to prevent the reimbursement information from being tampered in the transmission and storage processes, the first user side may generate the reimbursement information by signing with the first private key.

In this embodiment, at least two key pairs may be generated in advance, where the first public key and the first private key are a private key pair, and the second public key and the second private key are a private key pair. The first public key and the second public key are both stored in the server, the first private key is stored in the first user, and the second private key is stored in the second user. The specific type, generation mode and electronic signature mode of the key pair are not limited, and can be reasonably selected according to actual application requirements.

In this embodiment, since the first public key and the first private key are a private key pair, after receiving the first reimbursement application form data, the server may verify the first signature information therein by using the first public key to determine whether the first reimbursement application form data is legal.

Optionally, considering that the electronic reimbursement certificate corresponding to the reimbursement information is used for financial accounting after the reimbursement approval is completed, the electronic reimbursement certificate can be signed in order to avoid that the related information of the electronic reimbursement certificate corresponding to the reimbursement information is tampered in the approval or storage stage. Specifically, the reimbursement information is further used for recording an electronic reimbursement certificate, and correspondingly, the first reimbursement application form data further comprises certificate identification information and certificate message digest information. The certificate identification information is used for recording the unique identification of the electronic reimbursement certificate, and the certificate message digest information is generated by carrying out hash calculation on the unique identification identified on the electronic reimbursement certificate and the consumption amount.

Further, in order to prevent the private key from being copied randomly, it may be preferable that the key pairs are both generated by the user side and the public key is sent to the server side, that is, the first user side generates the first public key and the first private key and sends the first public key to the server side for storage; the second user side generates a second public key and a second private key, and sends the second public key to the server side for storage.

Further, in order to avoid that other people impersonate the actual user of the first user terminal, the step S201 may further include: after the personal biological characteristics of the first user pass verification, the first user side can sign the reimbursement information by using the first private key and generate first signature information. The first user is a user of the set first user terminal.

Step S202, when the first signature information passes verification, the first reimbursement application data is sent to the second user side, so that the second user side can audit the reimbursement information and generate first audit information, and the second private key is utilized to sign the first audit information and generate second signature information.

In this embodiment, when the server verifies that the first signature information passes, it indicates that the first reimbursement application form data is legal, so that the first reimbursement application form data can be sent to the second user side. The second user end and the server end can realize network communication through a local area network or the Internet, and the second user end can be an intelligent terminal used by an approver. After the second user receives the first reimbursement application form data, the reimbursement information in the first reimbursement application form data can be checked and audited, and first audit information is generated.

In this embodiment, the server side stores the correspondence between each user identifier and each user identifier in advance, so that after receiving the first reimbursement form application data, the server side can determine the corresponding second user identifier according to the approver identifier recorded in the reimbursement information.

In this embodiment, in order to prevent the content audited by the second user side from being tampered, the second user side may further sign the first audit information by using the second private key and generate second signature information, and then generate second reimbursement application form data at least according to the second signature information and the first audit information, and send the second reimbursement application form data to the server side. The specific information included in the second reimbursement application form data is not limited in kind, and can be reasonably selected according to actual application requirements. For example, to reduce the amount of data transmission, the second reimbursement application form data may include only the second signature information and the first trial batch information; in order to ensure the integrity and tamper resistance of the data, the second reimbursement application form data can also comprise first reimbursement application form data besides the second signature information and the first trial batch information.

In this embodiment, when the verification of the first signature information fails, it indicates that the first reimbursement application form data is illegal, so that the server may not need to perform subsequent other processing.

In this embodiment, in order to record the whole electronic approval process, when the server side verifies that the first signature information passes, the first reimbursement application form data may also be stored.

Further, to facilitate extraction and presentation of the data, it may be preferable that both the reimbursement information and the first trial information be structured data. The specific format of the structured data is not limited, and can be reasonably selected according to actual application requirements. For example, in view of the versatility of the data, it may be preferable that the format data is XML format data.

Further, in order to avoid that other people impersonate the actual user of the second user terminal to audit the reimbursement information by using the second user terminal, it may be preferable that the second user terminal signs the first approval information by using the second private key and generates the second signature information after the second user terminal performs the personal biological feature verification on the second user terminal. The second user is a user of the set second user terminal.

Further, in order to obtain a better tamper-proof effect, it may be preferable that the second signature information is generated by signing the first trial batch information, the reimbursement information and the first signature information by the second user side using the second private key.

Step S203, receiving second reimbursement application form data sent by the second user terminal.

And step S204, verifying the second signature information by using the second public key, and obtaining target reimbursement bill data according to the second reimbursement bill data when the second signature information is verified.

In this embodiment, the second public key is used to verify the second signature information, so as to determine whether the second reimbursement application form data is legal, and the subsequent processing steps are not limited, so that reasonable selection can be performed according to the actual application requirements. For example, when the second public key is used to verify that the second signature information fails, it indicates that the second reimbursement application form data is illegal, and the server side may not need to perform subsequent other processing, or may send the second reimbursement application form data to other user sides for approval processing, or may send a corresponding prompt to the first user side. When the second public key is used for verifying the second signature information, the second reimbursement application form data can be stored, or the first reimbursement application form data and/or the second reimbursement application form data can be sent to other user terminals to continue approval processing, and corresponding prompts can be sent to the first user terminal.

Optionally, when there are multiple approvers, i.e. at least three clients, the embodiment may further include steps S205-S207 after step S204, specifically:

Step S205, when the second signature information passes the verification, the second reimbursement application data is sent to the third user terminal, so that the third user terminal can audit the reimbursement information and the first audit information and generate second audit information, and the third private key is utilized to sign the second audit information and generate third signature information.

When the server verifies the second signature information, the second reimbursement application form data are indicated to be legal, so that the second reimbursement application form data can be sent to the third user side. The third user end and the server end can realize network communication through a local area network or the Internet, the third user end can be an intelligent terminal used by another approver, and the approval is performed after the approval of the approver using the second user end is completed. After receiving the second reimbursement application form data, the third user terminal can check and audit reimbursement information in the second reimbursement application form data and generate third approved information.

Similarly to the description in the foregoing steps, in order to prevent tampering of the content audited by the third user side, the third user side may further sign the second audit information by using the third private key and generate third signature information, and then generate third reimbursement application form data at least according to the third signature information and the second audit information, and send the third reimbursement application form data to the server side. The specific information included in the third reimbursement application form data is not limited in kind, and can be reasonably selected according to actual application requirements. For example, to reduce the amount of data transmission, the third reimbursement application form data may include only the third signature information and the second trial batch information; in order to ensure the integrity and tamper resistance of the data, the third reimbursement application form data can also comprise second reimbursement application form data besides the third signature information and the second trial batch information.

The third public key and the third private key of the private key pair can be pre-generated, wherein the third public key is also stored in the server side, and the third private key is stored in the third user side.

Step S206, receiving third reimbursement application form data sent by a third user side.

Step S207, the third signature information is verified by using the third public key.

The third public key is utilized to verify the third signature information, so that whether the third reimbursement application form data are legal or not can be determined, the subsequent processing steps are not limited, and reasonable selection can be performed according to actual application requirements. For example, when the third public key is used to verify that the third signature information fails, it indicates that the third reimbursement application form data is illegal, and the server side may not need to perform subsequent other processing, or may send the third reimbursement application form data to other user sides for approval processing, or may send corresponding prompts to the first user side and/or the second user side. When the third public key is used for verifying the third signature information, the third reimbursement application form data can be stored, or the first reimbursement application form data and/or the third reimbursement application form data can be sent to other user terminals to continue approval processing, and corresponding prompts can be sent to the third user terminal and/or the second user terminal.

Further, in order to improve the reliability of the data storage at the server side, step S207 may further include:

and when the third signature information passes verification, signing the first reimbursement application form data, the second reimbursement application form data and the third reimbursement application form data to obtain target reimbursement application form data.

The method for signing the first reimbursement application form data, the second reimbursement application form data and the third reimbursement application form data by the server side is not limited, and reasonable selection can be performed according to actual application requirements.

Further, when there are multiple approvers, i.e. at least three clients, the embodiment may further include steps S208-S211 after step S201, specifically:

step S208, when the verification of the first signature information is passed, the first reimbursement application data is sent to a second user side and a third user side, so that the second user side can audit the reimbursement information and generate first audit information, and the second private key is used for signing the first audit information and generating second signature information; and the third user terminal can audit the reimbursement information and generate second audit information, and signs the second audit information by using a third private key and generates second signature information.

The first reimbursement application form data are sent to the second user side and the third user side, so that the second user side and the third user side can respectively audit reimbursement information in the first reimbursement application form data.

The method for generating the second signature information and the second signature information, and the method for generating the second reimbursement application form data and the third reimbursement application form data by the second user side and the third user side are substantially the same as or similar to the method in the second embodiment, and are not described herein.

Step S209, receiving second reimbursement application form data sent by the second user side and third reimbursement application form data sent by the third user side.

The receiving order of the second reimbursement application form data and the third reimbursement application form data is not limited because different approvers may approve at different times.

Step S210, the second signature information is verified by the second public key, and the third signature information is verified by the third public key.

The verification method for the second signature information and the third signature information is basically the same as or similar to the verification method described above, and will not be described in detail herein.

Step S211, when the second signature information and the third signature information pass verification, the first reimbursement application form data, the second reimbursement application form data and the third reimbursement application form data are stored.

When the second signature information and the third signature information are verified, the two approvers are informed of finishing approval of the reimbursement information, so that the first reimbursement application form data, the second reimbursement application form data and the third reimbursement application form data can be stored for recording the electronic approval process.

Further, in order to improve the reliability of the data storage at the server side, step S211 may further include:

and when the second signature information and the third signature information pass verification, signing the first reimbursement application form data, the second reimbursement application form data and the third reimbursement application form data to obtain target reimbursement application form data.

In addition, it should be noted that, the embodiment is only illustrated by taking two or three approvers as an example, and actually may include more than three approvers, that is, further include a plurality of clients, and each client and the server have an exclusive key pair therebetween. When more approvers exist, the signature can be carried out after all the approvers complete all the audits and send corresponding reimbursement application form data so as to obtain target reimbursement form data.

And step S214, processing the target reimbursement bill data by using the reimbursement bill accounting bill classification model to obtain accounting bill data corresponding to the target reimbursement bill data.

Step S214 in this embodiment is substantially the same as or similar to step S103 in the first embodiment, and will not be described herein.

In addition, step S212 and step S213 may be further included before step S214, specifically:

step S212, a plurality of historical reimbursement bill data and corresponding accounting data are obtained.

Step S212 in this embodiment is substantially the same as or similar to step S101 in the first embodiment, and will not be described herein.

Step S213, training to obtain a reimbursement bill accounting subject classification model according to all or part of the historical reimbursement bill data and the corresponding accounting subject accounting data.

Step S213 in this embodiment is substantially the same as or similar to step S102 in the first embodiment, and will not be described herein.

Optionally, in order to automatically generate the first reimbursement application form data to perform subsequent reimbursement approval and financial accounting on the target business trip, the business trip reimbursement efficiency is improved, and before step S201, the method may further include:

step S001, receiving business trip application data.

In this embodiment, the business trip application data is used to record at least a business trip person, a business trip start point, a business trip destination, and a business trip end point of the target business trip. The business trip personnel or the sponsor assisting in processing business trip administrative matters can input business trip related information in the electronic system before or after business trip of the business trip personnel so as to generate business trip application data.

In this embodiment, the business trip application data includes first application sub-information, second application sub-information, third application sub-information and fourth application sub-information, where the first application sub-information is used for recording a business trip person of a target business trip, the second application sub-information is used for recording a business trip start point, the third application sub-information is used for recording a business trip destination, and the fourth application sub-information is used for recording a business trip end point.

Step S002, screening all stored electronic reimbursement credential data according to the business trip application data to obtain traffic credential data corresponding to the target business trip.

In this embodiment, the electronic reimbursement document data includes at least traffic document data, where the traffic document data is used to record a consumption amount, a trip start point, and a trip end point identified on at least one traffic reimbursement document. Other content identified by the traffic voucher data is not limited, and can be reasonably selected according to actual application requirements. For example, in addition to the amount of consumption, the start of journey, and the end of journey, the traffic voucher data can be used to record journey times and journey personnel identified on the traffic reimbursement voucher.

To facilitate understanding of the solution of the present embodiment, a train ticket is further illustrated. For a train ticket, the fare amount marked on the train ticket is the consumption amount, the marked start station is the journey start point, the marked end station is the journey end point, the marked riding date is the journey time, and the marked rider is the journey personnel.

In this embodiment, the internet server stores the electronic reimbursement certificate data in advance, and generates index data corresponding to the electronic reimbursement certificate data, where the index data is at least used to record an association relationship between each electronic reimbursement certificate and the user, a type of each electronic reimbursement certificate, and a consumption amount identified on each electronic reimbursement certificate. When the electronic reimbursement vouchers are traffic reimbursement vouchers, the index data is also used for recording the travel starting point and the travel ending point identified on each traffic reimbursement voucher. For the present embodiment, the user corresponds to a business trip person, that is, an association relationship between the business trip person and the electronic reimbursement credential data is pre-established at the internet server.

In this embodiment, the method for establishing the index data is not limited, and may be reasonably selected according to the actual application requirements. For example, the association of the traffic reimbursement credential with the user may be established by the trip person identified on the traffic reimbursement credential; or the user can reserve a train ticket, an airplane ticket or a taxi through a designated channel, and establish an association relationship between the reimbursement certificate and the user based on data generated in a reservation process.

Optionally, in order to obtain the association relationship between the traffic reimbursement credential and the user more accurately, step S001 may further include: and screening the electronic reimbursement voucher data according to the business trip application data and the business trip preset data to obtain the traffic voucher data corresponding to the target business trip. The travel preset data are used for recording the association relation between the target travel and at least one traffic reimbursement certificate.

Further, in order to obtain the association relation between the food sink reimbursement certificates and the user more accurately, the travel reservation data is also used for recording the association relation between the target travel and at least one food sink reimbursement certificate.

In this embodiment, the target business trip is a trip in which a reimbursement application form needs to be generated and reimbursement approval is performed. Because the index data corresponding to the electronic reimbursement voucher data is generated in advance, after the business trip application data is obtained, the electronic reimbursement voucher data stored in the server side can be screened according to a business trip person, a business trip start point, a business trip destination and a business trip end point of a target business trip recorded in the business trip application data, so that the traffic voucher data corresponding to the target business trip is obtained.

Optionally, step S002 may include sub-step S002a and sub-step S002b, wherein:

and step S002a, screening the electronic reimbursement voucher data according to the first application sub-information to obtain electronic reimbursement voucher data corresponding to the business person.

The index data can be used for recording the association relation between each electronic reimbursement voucher and the user, so that after the first application sub-information is obtained, the electronic reimbursement voucher data can be screened by utilizing the user identification of the business trip person to obtain the electronic reimbursement voucher data corresponding to the business trip person.

And step S002b, screening the electronic reimbursement voucher data corresponding to the business trip according to the second application sub-information, the third application sub-information and the fourth application sub-information to obtain the traffic voucher data corresponding to the target business trip.

In this embodiment, the index data is further used to record the type of each electronic reimbursement certificate, the consumption amount identified on each electronic reimbursement certificate, and the trip start point and the trip end point identified on each traffic reimbursement certificate, so after obtaining the electronic reimbursement certificate data corresponding to the business trip person, the traffic certificate data corresponding to the target business trip can be further obtained by screening.

Compared with the method for screening the electronic reimbursement voucher data by utilizing the first application sub-information, the second application sub-information, the third application sub-information and the fourth application sub-information, the method has the advantages that the electronic reimbursement voucher data corresponding to a business trip person is screened and obtained by utilizing the first application sub-information, and the traffic voucher data corresponding to a target business trip is screened and obtained by utilizing the second application sub-information, the third application sub-information and the fourth application sub-information, so that the processing pressure of parallel processing of the data can be reduced.

Optionally, considering that in practical application, the prestored electronic reimbursement certificate data may not be complete, that is, all reimbursement certificates of the target business trip may not be recorded and stored, in order to more comprehensively obtain the traffic certificate data corresponding to the target business trip, whether the traffic reimbursement certificates are collected completely or not may be judged according to the business trip application data and the traffic certificate data corresponding to the target business trip. In particular, sub-step S002B may further include the following sub-steps B1-B5:

And a sub-step B1, wherein the first journey closed-loop data is obtained according to the second application sub-information, the third application sub-information and the fourth application sub-information corresponding to the target journey.

In general, a trip route may form a complete route, so that the second application sub-information, the third application sub-information and the fourth application sub-information corresponding to the target trip route may be used to obtain the first route closed-loop data for recording a trip route.

For example, if the start of the trip, the destination of the trip, and the end of the trip are beijing, shanghai, and beijing, respectively, the first line Cheng Bihuan of data is for the recorded trip route to be "beijing-shanghai-beijing".

And B2, obtaining second journey closed-loop data according to the traffic evidence data corresponding to the target journey.

The traffic voucher data is used for recording the travel starting point and the travel ending point which are identified on at least one traffic reimbursement voucher, so that first travel closed-loop data for recording one or more travel routes can be obtained according to the traffic voucher data corresponding to the target business trip.

For example, if the traffic voucher data records a total of three traffic reimbursement vouchers with the identified travel start and travel end points, the first traffic reimbursement voucher with the identified travel start and travel end points being Beijing and Shanghai, the second traffic reimbursement voucher with the identified travel start and travel end points being Nanjing and Tianjin, and the third traffic reimbursement voucher with the identified travel start and travel end points being Tianjin and Beijing, two travel routes "Beijing-Shanghai" and "Nanjing-Tianjin-Beijing" may be formed after all traffic reimbursement vouchers with the identified travel start and travel end points being concatenated, the second travel closed loop data for recording travel routes being "Beijing-Shanghai" and "Nanjing-Tianjin-Beijing".

And a sub-step B3, judging whether the traffic reimbursement evidence of the target business trip is completely collected according to the first line Cheng Bihuan data and the second trip closed-loop data.

And B4, if so, generating first reimbursement application form data according to the traffic evidence data corresponding to the target business trip.

When the second travel closed-loop data is used for recording a travel route, and the travel starting point and the travel ending point of the travel route recorded by the first line Cheng Bihuan data and the second travel closed-loop data are the same, the traffic reimbursement evidence of the target business trip is considered to be completely collected, so that the first reimbursement application form data can be generated according to the traffic evidence data corresponding to the target business trip.

And B5, if not, prompting that manual confirmation is needed, and generating first reimbursement application form data according to the received complete confirmation instruction.

When the second trip closed-loop data is used for recording a plurality of trip routes, or the trip starting point and the trip ending point of the trip routes recorded by the first line Cheng Bihuan data and the second trip closed-loop data are different, the traffic reimbursement evidence of the target trip is considered to be not completely collected. And prompting to manually confirm or supplement the traffic evidence data, and generating first reimbursement application form data according to the received complete confirmation instruction.

Optionally, considering that in practical application, the prestored electronic reimbursement certificate data may not be complete, that is, all reimbursement certificates of the target business trip may not be recorded and stored, in order to more comprehensively obtain traffic certificate data corresponding to the target business trip, whether the traffic reimbursement certificates are collected completely or not may be judged according to the traffic certificate data corresponding to the target business trip. In particular, sub-step S002b may further include the following sub-steps C1-C4:

and C1, acquiring primary screening data of the journey certificate according to the second application sub-information, the third application sub-information and the fourth application sub-information.

The journey credential prescreening data is traffic credential data associated with at least one of a journey start point, a journey destination and a journey end point of a target journey in electronic reimbursement credential data corresponding to a journey.

And C2, judging whether the journey starting point and the journey ending point marked by all traffic reimbursement vouchers recorded by the journey voucher primary screening data can form a location-related closed loop.

The related closed loop of the place means that after the travel starting point and the travel end point marked by all traffic reimbursement certificates are connected in series, a complete travel route can be formed, and the starting point and the end point of the route are consistent with the travel starting point and the travel end point of the target travel.

For example, assuming that the starting point of the target business trip is Beijing, the business trip destination is Shanghai, the ending point of the business trip is Beijing, the business trip voucher primary screening data totally records three traffic reimbursement vouchers, the starting point and the ending point of the business trip marked on the first traffic reimbursement voucher are Beijing and Shanghai respectively, the starting point and the ending point of the business trip marked on the second traffic reimbursement voucher are Shanghai and Tianjin respectively, the starting point and the ending point of the business trip marked on the third traffic reimbursement voucher are Tianjin and Beijing respectively, and after the starting points and the ending points of the business trips marked on all the traffic reimbursement vouchers are connected in series, the formed business trip route is 'Beijing-Shanghai-Tianjin-Beijing' which coincides with the business trip starting point and the business trip ending point of the target business trip, so that a closed loop of association of places can be formed by determining all the traffic reimbursement vouchers marked on the primary screening data.

For another example, assuming that the starting point of the target business trip is Beijing, the business trip destination is Shanghai, the ending point of the business trip is Beijing, three traffic reimbursement certificates are recorded in total in the business trip certificate primary screening data, the starting point and the ending point of the business trip identified on the first traffic reimbursement certificate are Beijing and Shanghai respectively, the starting point and the ending point of the business trip identified on the second traffic reimbursement certificate are Shanghai and Nanjing respectively, the starting point and the ending point of the business trip identified on the third traffic reimbursement certificate are Nanjing and Tianjin respectively, and after the starting points and the ending points of the business trips identified on all the traffic reimbursement certificates are connected in series, the formed business trip route is 'Beijing-Shanghai-Tianjin' which is inconsistent with the business trip ending point of the target business trip, so that a closed loop is not formed between the starting point and the business trip ending point identified on the business trip primary screening data can be determined.

For another example, assuming that the starting point of the target business trip is Beijing, the destination of the business trip is Shanghai, the end point of the business trip is Beijing, three traffic reimbursement certificates are recorded in total in the business trip primary screening data, the starting point and the end point of the business trip marked on the first traffic reimbursement certificate are Beijing and Shanghai respectively, the starting point and the end point of the business trip marked on the second traffic reimbursement certificate are Nanjing and Tianjin respectively, and the starting point and the end point of the business trip marked on the third traffic reimbursement certificate are Tianjin and Beijing respectively, after the starting point and the end point of the business trip marked on all the traffic reimbursement certificates are connected in series, two business trip routes of Beijing-Shanghai and Nanjing-Tianjin-Beijing are formed, and a complete business trip route is not formed, so that the starting point and the end point of the business trip marked on the business trip primary screening data can be determined.

And C3, if so, determining the journey certificate primary screening data as the traffic certificate data corresponding to the target business trip.

When the travel starting point and the travel ending point, which are marked by all the traffic reimbursement certificates and recorded by the travel certificate primary screening data, are determined to form a relevant closed loop of the place, the traffic reimbursement certificates in the target business trip are indicated to be likely to be complete, so that the travel certificate primary screening data can be determined as the traffic certificate data corresponding to the target business trip.

And C4, if not, acquiring journey credential supplementary data according to journey credential primary screening data, and determining the journey credential primary screening data and the journey credential supplementary data as traffic credential data corresponding to a target business trip.

When it is determined that the trip start point and the trip end point identified by all traffic reimbursement vouchers recorded by the trip voucher primary screening data cannot form a relevant closed loop of a place, it is indicated that the traffic reimbursement vouchers in the target business trip are likely to be incomplete, and the traffic reimbursement vouchers need to be supplemented, namely, the trip voucher supplementary data need to be obtained.

The journey credential supplementary data is traffic credential data associated with journey starting points or journey ending points identified by all traffic reimbursement credentials recorded by journey credential prescreening data in electronic reimbursement credential data corresponding to a business person. The collection mode of the journey certificate supplementary data is not limited, and reasonable setting can be carried out according to actual application requirements. For example, the sponsor may be prompted to supplement and upload corresponding traffic credential data, or the electronic reimbursement credential data may be rescreened using a preset algorithm.

Further, considering that the same business trip person is unlikely to be simultaneously present in two different trips in the actual application process, in order to improve the accuracy of the traffic credential data collection corresponding to the target business trip, when the traffic credential data is further used to record the trip start time point identified on at least one traffic reimbursement credential, the substep C4 may further include:

and judging whether an overlapped journey exists or not according to the journey starting time point, the journey starting point and the journey ending point marked by all traffic reimbursement vouchers recorded by the journey voucher primary screening data and the journey voucher supplemental data. Wherein, overlapping journey refers to journey with conflict in time, and if the situation is not expected, the same business trip person cannot obtain two different traffic reimbursement certificates at the same time.

If yes, prompting that manual confirmation is needed, and determining the traffic credential data corresponding to the target business trip according to the received trip confirmation instruction.

If not, determining the journey credential prescreening data and journey credential supplementary data as traffic credential data corresponding to the target business trip.

Optionally, in order to more comprehensively obtain the traffic credential data corresponding to the target business trip, the electronic reimbursement credential data may be further screened according to a time range of the target business trip. Specifically, the business trip application data further includes fifth application sub-information for recording a time range of the target business trip; the traffic voucher data is further for recording a trip start time point identified on at least one traffic reimbursement voucher.

Correspondingly, sub-step S002b comprises: and screening the electronic reimbursement voucher data corresponding to the business trip according to the second application sub-information, the third application sub-information, the fourth application sub-information and the fifth application sub-information to obtain the traffic voucher data corresponding to the target business trip.

Step S003, first reimbursement application form data are generated according to traffic credential data corresponding to the target business trip.

In this embodiment, the first reimbursement application form data is at least used for recording the business trip person of the target business trip, and the consumption amount, the trip start point and the trip end point corresponding to all traffic reimbursement certificates to be reimbursed. In addition to these information, other information recorded by the first reimbursement application form data is not limited in kind and can be reasonably selected according to actual application requirements. For example, for some units, other information such as the time of business trip, the amount of subsidy, etc. needs to be identified on the reimbursement application before reimbursement approval is performed, and therefore the first reimbursement application data can also be used to identify these information.

In this embodiment, the consumption amount, the trip start point and the trip end point identified on the traffic reimbursement voucher are recorded on the traffic reimbursement voucher data, and the business trip person, the business trip start point, the business trip destination and the business trip end point of the target business trip are recorded in the business trip application data, so after the traffic voucher data corresponding to the target business trip is obtained, the first reimbursement application form data can be generated according to the traffic voucher data corresponding to the target business trip.

Therefore, after the business trip person, the business trip starting point, the business trip destination and the business trip finishing point of the target business trip are determined, the first reimbursement application form data can be automatically generated, so that subsequent reimbursement approval and financial account making can be performed on the target business trip, and the business trip reimbursement efficiency is improved.

Optionally, considering that in addition to the traffic reimbursement credential, the sink reimbursement credential may be obtained during the business trip, in order to comprehensively collect various reimbursement credentials generated during the business trip and generate reimbursement application forms, the substep S002b may further include:

and D1, screening the electronic reimbursement voucher data corresponding to the business trip according to the traffic voucher data corresponding to the target business trip to obtain the food destination voucher data corresponding to the target business trip.

Wherein the sink voucher data is used for recording the amount of consumption identified on the at least one sink type reimbursement voucher, as well as the time of consumption and/or the place of consumption.

Correspondingly, step S003 may further include a substep D2: and generating first reimbursement application form data according to the traffic credential data and the food destination credential data corresponding to the target business trip.

Further, in order to improve the accuracy and the integrity of the collection of the food sink reimbursement vouchers, the food sink vouchers data corresponding to the target business trip can be determined together according to business trip application data and the traffic vouchers data corresponding to the target business trip. Specifically, sub-step D2 may further comprise:

And screening the electronic reimbursement voucher data corresponding to the business trip according to the third application sub-information, the fifth application sub-information and the traffic voucher data corresponding to the target business trip to obtain the food destination voucher data corresponding to the target business trip.

Further, in order to perform compliance verification on whether the food-class reimbursement vouchers belong to the reimbursement vouchers of the target business trip, the food-class reimbursement vouchers data are also used for recording food-class time and/or food-class places of at least one food-class reimbursement voucher locking mark; correspondingly, sub-step D2 may further comprise: judging whether the destination credential data corresponding to the target business trip is compliant according to the business trip application data and/or the traffic credential data corresponding to the target business trip; if so, generating first reimbursement application form data according to traffic evidence data and food destination evidence data corresponding to the target business trip; if not, prompting that manual confirmation is needed, and generating first reimbursement application form data according to the received compliance confirmation instruction.

Further, considering that in practical application, the business trip standards enjoyed by business trip persons of different job classes are different, in order to simplify the subsequent approval content of the reimbursement bill, compliance verification can be performed in the process of generating reimbursement bill data. Specifically, the traffic credential data is further used to record the identified vehicle seat on the at least one traffic reimbursement credential: the business trip application data further includes sixth sub-information for recording the business trip person position level. Correspondingly, sub-step D2 may further comprise:

And judging whether the total consumption amount exceeds the standard according to the sixth sub-information, the traffic evidence data and the food destination evidence data corresponding to the target business trip.

If not, generating first reimbursement application form data directly according to the traffic evidence data and the food destination evidence data corresponding to the target business trip.

If yes, prompting that manual confirmation is needed, and generating first reimbursement application form data according to the received budget confirmation instruction.

Further, considering that in practical application, a business trip person may have a corresponding patch, after determining the time range of the target business trip, the amount of the patch may be obtained by calculation according to the time range of the target business trip. Specifically, the reimbursement application form data is also used for recording the subsidy amount corresponding to the target business trip.

Optionally, considering that the electronic reimbursement certificate has replicability, if a traffic reimbursement certificate is repeatedly checked out multiple times, the financial compliance risk is brought, so in order to avoid that the traffic reimbursement certificate is repeatedly checked out multiple times, corresponding verification can be performed on the electronic reimbursement certificate data for generating the first reimbursement application form data. Specifically, step S003 may include:

And obtaining reimburseable credential data corresponding to the target business trip according to the traffic credential data corresponding to the target business trip. And generating first reimbursement application form data according to reimburseable credential data corresponding to the target business trip.

Wherein the reimburseable document data is used for recording the consumption amount, the travel starting point and the travel ending point which are identified on at least one traffic reimbursement document which is not reimbursed.

Optionally, considering that some units of the financial system are deployed in an intranet, the intranet is isolated from the internet, in order to transmit the electronic reimbursement credential data from the external network to the intranet, if the implementation subject of step S001-step S003 is a terminal located in the intranet, step S000a and step S000b may be further included before step S001. Specifically:

step S000a, encrypting the electronic reimbursement voucher data corresponding to the target electronic reimbursement voucher by using a preset encryption algorithm to obtain voucher encrypted data.

In consideration of that some units of financial systems are deployed in an intranet, the intranet is isolated from the internet, so that in order to realize safe transmission of electronic reimbursement credential data from an external network to the intranet, the internet server can further encrypt the electronic reimbursement credential data.

The target electronic reimbursement certificate is an electronic reimbursement certificate which needs to be transmitted from an internet server (external network) to an internal network terminal, and the electronic reimbursement certificate is recorded by using corresponding electronic reimbursement certificate data.

Step S000b, generating two-dimensional code image data according to the credential encryption data, so that the credential encryption data can be obtained after scanning and identifying the two-dimensional code image data.

The two-dimensional code image data can be scanned and identified by a terminal located in an intranet to obtain the credential encryption data. The specific type of the terminal of the intranet is not limited. For example, the terminal located in the intranet can be a desktop computer, a notebook computer or a mobile phone, and the two-dimensional code image data can be scanned and identified through a camera or a two-dimensional code scanning gun.

After the terminal of the intranet obtains the certificate encrypted data, the terminal can decrypt the certificate encrypted data to obtain electronic reimbursement certificate decrypted data; and regenerating the electronic reimbursement voucher data corresponding to the target electronic reimbursement voucher according to the electronic reimbursement voucher decryption data.

Because the certificate encryption data is generated by encrypting by using a preset encryption algorithm, the intelligent terminal in the intranet can decrypt the certificate encryption data according to a decryption algorithm corresponding to the intelligent terminal in the intranet, so as to obtain the reimbursed electronic certificate decryption data. After the terminal in the intranet obtains the electronic reimbursement voucher decryption data, the terminal can directly or indirectly obtain the electronic reimbursement voucher data corresponding to the target electronic reimbursement voucher according to the electronic reimbursement voucher decryption data, so that the electronic reimbursement voucher data corresponding to the target electronic reimbursement voucher is transmitted from the external network to the intranet.

Further, in order to improve the security of data transmission, before step S000a, the method may further include: and generating a pair of asymmetric public keys and private keys by using a preset encryption algorithm, wherein the private keys are stored in a terminal of an intranet, and the public keys are stored in an Internet server, so that the Internet server can encrypt electronic reimbursement credential data corresponding to the target electronic reimbursement credential by using the public keys.

Further, considering that the layout structure of the electronic reimbursement certificate is limited in practical application, and one type of electronic reimbursement certificate generally has a unified layout standard, in order to reduce the data transmission amount, it may be preferable to transmit only the content sub-data corresponding to the target electronic reimbursement certificate when transmitting the data from the external network to the internal network, that is, the internet server encrypts the content sub-data corresponding to the target electronic reimbursement certificate by using a preset encryption algorithm and generates the two-dimensional code image data.

Correspondingly, the intranet terminal identifies the two-dimensional code image data, decrypts the credential encrypted data after obtaining the credential encrypted data, and obtains electronic reimbursement credential decrypted data; decrypting the data according to the electronic reimbursement certificate to obtain content sub-data corresponding to the target electronic reimbursement certificate; and then, according to the content sub-data corresponding to the target electronic reimbursement voucher, regenerating the electronic reimbursement voucher data corresponding to the target electronic reimbursement voucher by utilizing a preset electronic reimbursement voucher construction algorithm.

In order to obtain complete electronic reimbursement voucher data corresponding to the target electronic reimbursement voucher, the intranet terminal also needs to process the content sub-data corresponding to the target electronic reimbursement voucher by using a preset electronic reimbursement voucher construction algorithm so as to regenerate the electronic reimbursement voucher data corresponding to the target electronic reimbursement voucher.

Because only the content sub-data corresponding to the target electronic reimbursement certificate is encrypted and the two-dimensional code image data is generated, and the format sub-data corresponding to the target electronic reimbursement certificate is not encrypted and the two-dimensional code image data is generated, the data transmission quantity can be reduced.

Further, in order to facilitate extraction of the content sub-data, it may be preferable that the content sub-data is structured data.

Further, considering that the formats of the electronic reimbursement credential data at present are generally two types, namely PDF and OFD, for the OFD format, the structured data can be directly extracted; for the PDF format, a corresponding process is also required to obtain structured data. Therefore, in order to be compatible with different types of electronic reimbursement document data, when the format of the electronic reimbursement document data corresponding to the target electronic reimbursement document is PDF format, before step S000a, the internet server may further process the electronic reimbursement document data corresponding to the target electronic reimbursement document by using a preset data extraction algorithm, so as to obtain content sub-data corresponding to the target electronic reimbursement document.

Further, considering that the content identified on one electronic reimbursement certificate may be relatively more, and the data volume that can be recorded by one two-dimensional code is limited, when the data volume that needs to be transmitted from the external network to the internal network is larger, the problem can be solved by generating a plurality of two-dimensional codes.

Specifically, the two-dimensional code image data corresponding to the target electronic reimbursement certificate comprises one or more first image sub-data, each first image sub-data is used for representing one two-dimensional code, and the two-dimensional code represented by each first image sub-data is at least used for recording unique identification information corresponding to the target electronic reimbursement certificate, at least part of encrypted content sub-data and serial number information corresponding to each first image sub-data.

Each electronic reimbursement certificate is provided with unique identification information for carrying out unique identification on the electronic reimbursement certificate, so that different electronic reimbursement certificate data cannot be confused in the data transmission and processing processes.

When the data size of the content sub-data corresponding to the target electronic reimbursement voucher is larger, after the content sub-data is encrypted, the encrypted content sub-data can be split into a plurality of parts, and the first image sub-data is generated together with the unique identification information corresponding to the target electronic reimbursement voucher.

Correspondingly, the step of decrypting the credential encrypted data by the terminal of the intranet to obtain the electronic reimbursement credential decrypted data may include: and identifying all the first image sub-data, and obtaining the certificate encryption data according to the serial number information corresponding to each first image sub-data and the unique identification information corresponding to the target electronic reimbursement certificate.

In addition, after the internet server encrypts the content sub-data corresponding to the target electronic reimbursement certificate and generates a plurality of first image sub-data, in order to ensure the integrity of the data obtained by the intranet terminal and the electronic reimbursement certificate data corresponding to the target electronic reimbursement certificate can be obtained by subsequent accurate decryption, when the first image sub-data is generated, serial number information corresponding to each first image sub-data is recorded in the two-dimension code represented by each first image sub-data, wherein the serial number information is determined according to the position of the content sub-data recorded in the two-dimension code in the complete content sub-data.

For example, when the internet server encrypts the content sub-data corresponding to the target electronic reimbursement voucher with the unique identification information of "1234", the encrypted content sub-data is obtained as "010111110000"; the encrypted content sub-data may be sequentially split into three parts of "0101", "1111" and "0000", and three first image sub-data are generated, where the first image sub-data is at least used to record the unique identification information "1234", the encrypted partial content sub-data "0101" and the serial number "1", the second first image sub-data is at least used to record the unique identification information "1234", the encrypted partial content sub-data "1111" and the serial number "2", and the third first image sub-data is at least used to record the unique identification information "1234", the encrypted partial content sub-data "0000" and the serial number "3".

Further, considering that the electronic reimbursement certificates generally have unique identifiers, in order to conveniently obtain the unique identifier information corresponding to the target electronic reimbursement certificate, the unique identifier information corresponding to the target electronic reimbursement certificate can be determined according to the content sub-data corresponding to the target electronic reimbursement certificate. The specific determination method is not limited, and can be reasonably selected according to actual application requirements.

For example, considering that the electronic reimbursement certificates generally have unique number identifications, in order to conveniently obtain the unique identification information corresponding to the target electronic reimbursement certificate, the unique identification information corresponding to the target electronic reimbursement certificate is determined according to the electronic reimbursement certificate number information corresponding to the target electronic reimbursement certificate.

For another example, to ensure reliability of data transmission, the internet server and the intranet terminal may have a private key and a public key of a private key pair, respectively, and the internet server may generate a signature value, such as a hash value, for the content sub-data corresponding to the target electronic reimbursement certificate by using the private key, and then generate one or more first image sub-data by using the hash value as the unique identification information corresponding to the target electronic reimbursement certificate. Therefore, after the intranet terminal obtains the certificate encrypted data, the hash value can be checked by the public key, and the subsequent decryption operation can be executed only after the verification is passed.

Further, in order to ensure the integrity of data transmission, a two-dimensional code for recording the required transmission data amount can be separately generated. Specifically, the two-dimensional code image data corresponding to the target electronic reimbursement certificate comprises second image sub-data, the second image sub-data is used for representing a two-dimensional code, and the two-dimensional code represented by the second image sub-data is at least used for recording unique identification information corresponding to the target electronic reimbursement certificate and the quantity information of the first image sub-data.

Further, the second image sub-data is at least used for recording unique identification information, electronic reimbursement voucher number information, purchaser information, seller information, verification information, billing operation information and quantity information of the first image sub-data corresponding to the target electronic reimbursement voucher; the first image sub-data is at least used for recording partial project content information corresponding to the target electronic reimbursement certificate.

After the target user sets the receiving mailbox, the internet server can automatically analyze the target mail data in the receiving mailbox, automatically acquire the electronic reimbursement voucher data corresponding to the target mail data and establish the association relationship between the target user and the electronic reimbursement voucher data corresponding to the target mail data, thereby facilitating the target user to check and manage the individual electronic reimbursement voucher data and facilitating the subsequent reimbursement application form generation, reimbursement application form approval, financial account making and other processes to realize electronic and automatic processing.

Further, in order to facilitate the target user to search and manage the electronic reimbursement credential data of the individual, and facilitate the subsequent reimbursement application form generation, reimbursement application form approval, financial billing and other processes to realize electronic and automatic processing, step S000 c-step S000e may be further included before step S001. Specifically:

and step S000c, analyzing the target mail data received by the mailbox of the target user, and judging whether the target mail data is used for extracting the electronic reimbursement credential data.

The target user is a user who needs to perform electronic reimbursement credential management through an internet server, and the execution subject of the embodiment may be preferably the internet server. The internet server may be provided with a dedicated receiving mailbox for each user, or each user may set other receiving mailboxes by himself, so that the internet server may acquire mail data of each user from the receiving mailboxes.

The target mail data is all or part of mail data in a mailbox of a target user, namely, the target mail data is used for recording one or more mails, the range of the target mail data is not limited, and reasonable selection can be performed according to actual application requirements. For example, when the internet server acquires mail data in the mailbox of the target user for the first time, the target mail data may be all mail data in the mailbox of the target user; when the internet server does not acquire the mail data in the mailbox of the target user for the first time, the target mail data may be mail data received by the mailbox of the target user by the internet server between the current acquisition operation and the last acquisition operation time.

In view of the fact that mail such as work mail, advertisement mail, etc. may exist in the receiving mailbox of the target user, these mail cannot be used for extracting the electronic reimbursement voucher data, it is necessary to parse the target mail data after it is obtained to determine whether or not the target mail data is used for extracting the electronic reimbursement voucher data. The specific method for analyzing the target mail data is not limited, and can be reasonably selected according to actual application requirements. For example, the analysis rule of the mail data may be determined in advance by analyzing the transmission manner and transmission content of the electronic reimbursement voucher data by the issuing end of the plurality of electronic reimbursement vouchers, so that it may be determined whether the target mail data is used for extracting the electronic reimbursement voucher data in step S000 c.

The electronic reimbursement voucher data comprises content sub-data and format sub-data, wherein the content sub-data is used for representing the content of the electronic reimbursement voucher, and the format sub-data is used for representing the format structure of the electronic reimbursement voucher. For example, for an electronic invoice, the format sub-data may be used to represent the form format of the electronic invoice, and the format and position of standardized words such as "machine number", "invoice code", "invoice number", "billing date", "check code", "purchaser name", "purchaser tax payer identifier", "goods or tax administration, service name", "unit price", "amount", etc., that is, the format sub-data corresponding to the electronic invoice of this type is the same; the content sub-data may be used to represent specific information corresponding to "machine number", "invoice code", "invoice number", "billing date", "check code", "purchaser name", "purchaser tax payer identification number", "goods or tax administration, service name", "unit price", "amount" identified on the electronic invoice, where the information of different electronic invoices may be different, i.e. the content sub-data corresponding to different electronic invoices may be different.

The electronic reimbursement voucher data corresponding to the electronic reimbursement voucher can be content sub-data and format sub-data, or can be only content sub-data, and the embodiment is not limited herein, and can be reasonably selected according to actual application requirements.

Further, in order to better protect the personal privacy of the user and reduce the data processing amount of other types of mails, it may be preferable that the internet server sets up a dedicated receiving mailbox for each user, instead of the user hosting the private mailbox to the internet of things server.

And step S000d, when the target mail data is determined to be used for extracting the electronic reimbursement credential data, acquiring the corresponding electronic reimbursement credential data according to the target mail data.

When it is determined that all or part of the target mail data is available for extracting the electronic reimbursement credential data, the corresponding electronic reimbursement credential data can be obtained according to the part of the target mail data available for extracting the electronic reimbursement credential data, that is, according to the data corresponding to the mail from which the electronic reimbursement credential data can be obtained. The method comprises the steps that a plurality of issuing terminals are arranged on a mail service platform, and the issuing terminals are connected with the issuing terminals, wherein the issuing terminals are connected with the issuing terminals, and the issuing terminals are connected with the issuing terminals.

For example, the part of the issuing end can directly send the electronic reimbursement certificate data as a mail attachment to a receiving mailbox of the target user, and for the sending mode, an extraction rule can be set as downloading the mail attachment; the download link of the electronic reimbursement voucher data which is sent by the partial issuing end is connected to the receiving mailbox of the target user, and aiming at the sending mode, an extraction rule can be set to directly download the electronic reimbursement voucher data from the download link; the display link of the electronic reimbursement credential data sent by the partially opened end is connected to the receiving mailbox of the target user, and aiming at the sending mode, the extraction rule can be set to enter the interface corresponding to the display link, and the electronic reimbursement credential data can be downloaded or intercepted from the interface.

When it is determined that all or part of the target mail data cannot be used for extracting the electronic reimbursement certificate data, it is indicated that the mails corresponding to the data may be other types of mails, and the data may not be processed any more.

Further, the judging efficiency of whether the target mail data is used for extracting the electronic reimbursement credential data is improved, and the storing and processing of the abnormal data is reduced, step S000C may include sub-step a and sub-step B, and step S000d may include sub-step C. Specifically:

And a sub-step A, judging whether the target mail data comprises the credential extraction information according to a preset credential extraction rule.

The method comprises the step A, wherein a credential extraction rule can be preset before the step A, and credential extraction information for judging whether the mail can be used for extracting electronic reimbursement credential data and a credential extraction algorithm corresponding to each type of credential extraction information can be set in the credential extraction rule. The specific content represented by the credential extraction information is not limited, and can be reasonably set according to actual application requirements.

For example, the voucher extraction information can be preset to represent keywords such as "electronic invoice", "electronic ticket", "reimbursement voucher", and if the mail recorded by the target mail data includes at least one of "electronic invoice", "electronic ticket", "reimbursement voucher", it can be determined that the target mail data includes the voucher extraction information, that is, it can be determined that the target mail data can be used to extract the electronic reimbursement voucher data.

And B, if so, determining a credential extraction algorithm corresponding to the credential extraction information according to the credential extraction information and the credential extraction rule.

When it is determined that the target mail data includes at least one type of credential extraction information, a credential extraction algorithm corresponding to all credential extraction information may be further determined, so that electronic reimbursement credential data may be obtained from the mail attachment or through a network link by using the credential extraction algorithm corresponding to the credential extraction information.

And C, downloading the electronic reimbursement voucher data corresponding to the target mail data by utilizing a voucher extraction algorithm corresponding to the voucher extraction information.

Further, in order to improve the data processing efficiency, invalid voucher information for judging that the target mail data cannot be used for extracting the electronic reimbursement voucher data may also be set in the voucher extraction rule. Specifically, step S000c may include: and judging whether the target mail data comprises the credential extraction information or the invalid credential information according to the credential extraction rule. Correspondingly, the embodiment may further include: when it is determined that invalid credential information is included in the target mail data, the processing of the target mail data is terminated.

For example, the invalid credential information may be preset to be used to represent keywords such as "commodity discount", "work report", "business trip application", and if the mail recorded in the target mail data includes at least one of "commodity discount", "work report", "business trip application", it may be determined that the invalid credential information is included in the target mail data, that is, it may be determined that the target mail data is not available for extracting the electronic reimbursement credential data.

Further, considering that with the continuous progress of technology, the issuing end of the electronic reimbursement voucher may adjust the extraction mode of the electronic reimbursement voucher data, so that in order to better adapt to the change of the extraction mode of different electronic reimbursement voucher data, the voucher extraction rule can be updated. Specifically, the present embodiment may further include the following steps A1-A3:

And step A1, when the fact that the target mail data does not comprise the credential extraction information and the invalid credential information is determined according to the credential extraction rule, the target mail data is identified as abnormal data and stored.

And step A2, classifying all the stored abnormal data by using a preset machine learning algorithm.

And step A3, updating the credential extraction rule when the plurality of abnormal data are divided into one category.

If the credential extraction rule needs to be updated based on the abnormal data, the efficiency is low, and whether the credential extraction rule belongs to high-frequency abnormality or low-frequency abnormality is difficult to judge, so that all the abnormal data can be classified automatically through a preset machine learning algorithm, when a plurality of abnormal data are classified into one class, the abnormal data with high frequency can be indicated to appear, the abnormal data can be distinguished independently, if the abnormal data classified into one class are found after research, the credential extraction rule set in the credential extraction rule and the invalid credential information can be adjusted, the credential extraction rule, namely the credential extraction information and the corresponding credential extraction algorithm can be updated, and/or the invalid credential information can be updated accordingly.

And step S000e, generating user associated data and storing the electronic reimbursement credential data corresponding to the user associated data and the target mail data.

The user association data is at least used for identifying the association relationship between the target user and the electronic reimbursement credential data corresponding to the target mail data. After the internet server stores the electronic reimbursement credential data corresponding to the user association data and the target mail data, the electronic reimbursement credential data associated with the target user can be searched from all the electronic reimbursement credential data stored in the internet server based on the user association data, so that the target user can conveniently search and manage the personal electronic reimbursement credential data, and other processes such as generation of a subsequent reimbursement application form, approval of the reimbursement application form, financial account making and the like can be conveniently realized.

For example, when the target user is a business trip person, the server side may establish an association relationship between the target user and the electronic reimbursement credential data when the server side obtains the electronic reimbursement credential data for recording a food-class reimbursement credential through the mailbox of the target user. When the target user needs to generate a reimbursement application form for the business trip, the electronic reimbursement credential data for recording the food-class reimbursement credential can be searched from all electronic reimbursement credential data stored in the internet server through the user associated data, and first reimbursement application form data is generated.

Example III

An embodiment of the present application provides a reimbursement sheet accounting subject confirmation device, and fig. 3 is a schematic structural diagram of the reimbursement sheet accounting subject confirmation device disclosed in the embodiment of the present application, where the device includes:

the acquisition module is used for acquiring historical reimbursement bill data and corresponding accounting subject accounting data; the historical reimbursement bill data is used for recording accounting contents of at least one completed accounting in the reimbursement application bill; the accounting data is used for recording the accounting entry corresponding to the accounting content of which the accounting is completed.

The model training module is used for training and obtaining the reimbursement bill accounting object classification model according to all or part of the historical reimbursement bill data and the corresponding accounting object accounting data.

The accounting module is used for processing the target reimbursement bill data by using the reimbursement bill accounting bill classification model to obtain accounting bill accounting data corresponding to the target reimbursement bill data; the target reimbursement bill data is used for recording at least one accounting content which is not completed in the reimbursement bill.

Optionally, the accounting module is further configured to process the target reimbursement bill data by using a reimbursement bill accounting subject classification model to obtain machine classification data corresponding to the target reimbursement bill data; the machine classification data are at least used for recording accounting records corresponding to accounting contents obtained by processing the reimbursement bill accounting subject classification model;

And obtaining accounting subject accounting data corresponding to the target reimbursement data according to the confirmation instruction of the user on the machine classification data corresponding to the target reimbursement data.

Optionally, the accounting module is further configured to process the target reimbursement bill data by using a reimbursement bill accounting subject classification model to obtain machine classification data and machine unclassified data corresponding to the target reimbursement bill data; wherein the machine unclassified data is used to record billing content for which the reimbursement sheet accounting subject classification model is indeterminate of classification.

And providing the unclassified data of the machine corresponding to the target reimbursement single data for the user to confirm; and according to a confirmation instruction of the user to the unclassified data of the machine corresponding to the target reimbursement single data, acquiring accounting subject accounting data corresponding to the target reimbursement single data.

Optionally, the accounting module is further configured to perform classification processing on the target reimbursement data by using a reimbursement bill accounting subject classification model to obtain preliminary classification data and classification accuracy evaluation data corresponding to the target reimbursement data; the primary classification data is used for recording an accounting entry of which the accounting content is divided by the account classification model of the reimbursement bill; the classification accuracy evaluation data is used for recording accuracy evaluation values of each account content classified by the reimbursement bill accounting subject classification model;

When the accuracy evaluation value of the accounting content is larger than or equal to a preset evaluation threshold value, determining preliminary classification data corresponding to the accounting content as machine classification data;

when the accuracy evaluation value of the billing content is smaller than a preset evaluation threshold value, the billing content is determined as machine unclassified data.

Optionally, the device further comprises an updating module, which is used for updating the reimbursement bill accounting object classification model according to accounting object accounting data corresponding to the target reimbursement bill data.

Optionally, the device further includes a verification module, configured to verify, with a first public key, first signature information in first reimbursement application form data sent by the first user side; the first reimbursement application form data also comprises reimbursement information, wherein the reimbursement information is at least used for recording reimbursement persons, auditors, reimbursement amount and reimbursement matters; the first signature information is generated after the first user signs the reimbursement information by using a first private key; the first public key and the first private key are a private key pair;

when the first signature information passes verification, the first reimbursement application data is sent to a second user side, so that the second user side can audit reimbursement information and generate first audit information, and the second private key is utilized to sign the first audit information and generate second signature information;

Receiving second reimbursement application form data sent by a second user side; the second reimbursement application form data comprises first trial batch information and second signature information;

verifying the second signature information by using the second public key, wherein the second public key and the second private key are a private key pair;

and when the second signature information passes the verification, obtaining target reimbursement bill data according to the second reimbursement bill data.

Optionally, the second signature information is generated after the second user side signs the first trial batch information, the reimbursement information and the first signature information by using the second private key.

Optionally, the device further comprises an application form generating module, which is used for screening the electronic reimbursement voucher data according to the business trip application data to obtain traffic voucher data corresponding to the target business trip; the business trip application data are at least used for recording a business trip person, a business trip starting point, a business trip destination and a business trip end point of a target business trip; the electronic reimbursement voucher data is used for recording the content identified on at least one reimbursement voucher; the traffic voucher data is used for recording the consumption amount, the journey starting point and the journey ending point marked on at least one traffic reimbursement voucher;

Generating first reimbursement application form data according to the traffic evidence data corresponding to the target business trip; the first reimbursement application form data are at least used for recording the business trip person of the target business trip and the consumption amount, the trip starting point and the trip ending point corresponding to all traffic reimbursement certificates needing reimbursement.

Optionally, the business trip application data includes first application sub-information, second application sub-information, third application sub-information and fourth application sub-information for recording a business trip person, a business trip start point, a business trip destination and a business trip end point of the target business trip, respectively;

the application form generation module is also used for screening the electronic reimbursement voucher data according to the first application sub-information to obtain electronic reimbursement voucher data corresponding to the business person; and screening the electronic reimbursement voucher data corresponding to the business trip according to the second application sub-information, the third application sub-information and the fourth application sub-information to obtain the traffic voucher data corresponding to the target business trip.

The reimbursement bill accounting subject confirmation device of the embodiment can realize the corresponding reimbursement bill accounting subject confirmation method in the method embodiments, has the beneficial effects of the corresponding method embodiments, and is not repeated herein.

Thus, specific embodiments of the present application have been described. Other embodiments are within the scope of the following claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may be advantageous.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It should also be noted that the terms "comprises," "comprising," 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 may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.

It will be apparent to those skilled in the art that embodiments of the present application may be provided as a method, apparatus, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments in part.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.

Claims (7)

1. A method of validating a reimbursement sheet accounting subject, the method comprising:

acquiring historical reimbursement bill data and corresponding accounting subject accounting data; the historical reimbursement single data is used for recording accounting contents of at least one completed accounting in the reimbursement application form; the accounting subject billing data is used for recording a billing entry corresponding to the billing content after billing is completed;

training to obtain a reimbursement bill accounting subject classification model according to all or part of the historical reimbursement bill data and the corresponding accounting subject accounting data;

classifying the target reimbursement single data by using the reimbursement single accounting subject classification model to obtain preliminary classification data and classification accuracy evaluation data corresponding to the target reimbursement single data; the target reimbursement data are used for recording accounting contents of at least one incomplete accounting in the reimbursement application form; the preliminary classification data is used for recording an accounting entry of the accounting content divided by the reimbursement bill accounting object classification model; the classification accuracy evaluation data is used for recording accuracy evaluation values of the reimbursement bill accounting subject classification model for classifying each billing content;

When the accuracy evaluation value of the billing content is greater than or equal to a preset evaluation threshold value, determining the preliminary classification data corresponding to the billing content as machine classification data; obtaining accounting data corresponding to the target reimbursement data according to a confirmation instruction of a user to the machine classification data corresponding to the target reimbursement data; the machine classification data are at least used for recording accounting records corresponding to the accounting contents obtained by processing the reimbursement bill accounting subject classification model;

when the accuracy evaluation value of the billing content is smaller than a preset evaluation threshold value, determining the billing content as machine unclassified data; providing the unclassified data of the machine corresponding to the target reimbursement single data for a user to confirm; according to the confirmation instruction of the user on the unclassified data of the machine corresponding to the target reimbursement single data, acquiring the accounting account data corresponding to the target reimbursement single data; wherein the machine unclassified data is used to record the billing content for which the reimbursement sheet accounting subject classification model is indeterminate classification.

2. The method according to claim 1, wherein the method further comprises:

and updating the reimbursement bill accounting subject classification model according to the accounting subject accounting data corresponding to the target reimbursement bill data.

3. The method according to claim 1, wherein the method further comprises:

verifying the first signature information in the first reimbursement application form data sent by the first user terminal by using the first public key; the first reimbursement application form data also comprises reimbursement information, wherein the reimbursement information is at least used for recording reimbursement persons, auditors, reimbursement amount and reimbursement event; the first signature information is generated after the first user signs the reimbursement information by using a first private key; the first public key and the first private key are a private key pair;

when the first signature information passes verification, the first reimbursement application data is sent to a second user side, so that the second user side can audit the reimbursement information and generate first audit information, and a second private key is utilized to sign the first audit information and generate second signature information;

Receiving second reimbursement application form data sent by the second user side; wherein the second reimbursement application form data includes the first trial information and the second signature information;

verifying the second signature information by using a second public key, wherein the second public key and the second private key are a private key pair;

and when the second signature information passes the verification, acquiring the target reimbursement bill data according to the second reimbursement bill data.

4. The method of claim 3, wherein the second signature information is generated by the second client signing the first audit information, the reimbursement information, and the first signature information with the second private key.

5. A method according to claim 3, characterized in that the method further comprises:

screening the electronic reimbursement evidence data according to the business trip application data to obtain traffic evidence data corresponding to a target business trip; the business trip application data are at least used for recording a business trip person, a business trip starting point, a business trip destination and a business trip end point of a target business trip; the electronic reimbursement voucher data is used for recording the content identified on at least one reimbursement voucher; the traffic evidence data is used for recording the consumption amount, the journey starting point and the journey ending point marked on at least one traffic reimbursement evidence;

Generating the first reimbursement application form data according to the traffic evidence data corresponding to the target business trip; the first reimbursement application form data are at least used for recording the business trip person of the target business trip and the consumption amount, the trip starting point and the trip ending point corresponding to all traffic reimbursement certificates needing reimbursement.

6. The method of claim 5, wherein the business trip application data includes first, second, third, and fourth application sub-information for recording the business trip person, the business trip start point, the business trip destination, and the business trip end point, respectively, of the target business trip;

screening the electronic reimbursement evidence data according to the business trip application data, and obtaining the traffic evidence data corresponding to the target business trip comprises the following steps:

screening the electronic reimbursement voucher data according to the first application sub-information to obtain the electronic reimbursement voucher data corresponding to the business person;

and screening the electronic reimbursement voucher data corresponding to the business trip according to the second application sub-information, the third application sub-information and the fourth application sub-information to obtain the traffic voucher data corresponding to the target business trip.

7. A reimbursement sheet accounting subject verification device, comprising:

the acquisition module is used for acquiring historical reimbursement bill data and corresponding accounting subject accounting data; the historical reimbursement single data is used for recording accounting contents of at least one completed accounting in the reimbursement application form; the accounting subject billing data is used for recording a billing entry corresponding to the billing content after billing is completed;

the model training module is used for training to obtain a reimbursement bill accounting subject classification model according to all or part of the historical reimbursement bill data and the corresponding accounting subject accounting data;

the billing module is used for classifying the target reimbursement single data by using the reimbursement single accounting subject classification model to obtain preliminary classification data and classification accuracy evaluation data corresponding to the target reimbursement single data; the target reimbursement data are used for recording accounting contents of at least one incomplete accounting in the reimbursement application form; the preliminary classification data is used for recording an accounting entry of the accounting content divided by the reimbursement bill accounting object classification model; the classification accuracy evaluation data is used for recording accuracy evaluation values of the reimbursement bill accounting subject classification model for classifying each billing content;

When the accuracy evaluation value of the billing content is greater than or equal to a preset evaluation threshold value, determining the preliminary classification data corresponding to the billing content as machine classification data; obtaining accounting data corresponding to the target reimbursement data according to a confirmation instruction of a user to the machine classification data corresponding to the target reimbursement data; the machine classification data are at least used for recording accounting records corresponding to the accounting contents obtained by processing the reimbursement bill accounting subject classification model;

when the accuracy evaluation value of the billing content is smaller than a preset evaluation threshold value, determining the billing content as machine unclassified data; providing the unclassified data of the machine corresponding to the target reimbursement single data for a user to confirm; according to the confirmation instruction of the user on the unclassified data of the machine corresponding to the target reimbursement single data, acquiring the accounting account data corresponding to the target reimbursement single data; wherein the machine unclassified data is used to record the billing content for which the reimbursement sheet accounting subject classification model is indeterminate classification.