CN114445048A

CN114445048A - Contract processing method, device, equipment and storage medium based on RPA

Info

Publication number: CN114445048A
Application number: CN202210114385.9A
Authority: CN
Inventors: 唐剑锋; 王沛瑶; 石佳波; 陶建宇
Original assignee: Jiuxing Wuhan Information Technology Co ltd
Current assignee: Jiuxing Wuhan Information Technology Co ltd
Priority date: 2022-01-30
Filing date: 2022-01-30
Publication date: 2022-05-06

Abstract

The contract processing method, the contract processing device, the contract processing equipment and the contract processing storage medium based on the RPA technology realize contract processing, and the RPA technology has higher automation degree, so that the contract processing efficiency can be greatly improved, and the labor cost is reduced. Meanwhile, the first target information in the contract is identified by the neural network model, compared with a manual acquisition mode, the acquired first target information is more accurate, and when the contract is checked based on the first target information, the accuracy of a checking result can be improved, so that the standardization of the contract is improved, and the reasonable rights and interests of both parties of the contract are guaranteed.

Description

Contract processing method, device, equipment and storage medium based on RPA

Technical Field

The embodiment of the application relates to the technical field of finance, in particular to a contract processing method, a contract processing device, contract processing equipment and a contract processing storage medium based on RPA.

Background

In order to ensure the normative performance of daily work, contracts need to be established in many situations, such as buying and selling contracts, borrowing contracts, leasing contracts, entrusting contracts and the like, and the contracts generally involve items such as division of responsibility, settlement of expenses and the like. Therefore, the requirement for standardization is high, which requires that the relevant examination and verification of the contract be performed before the contract is established or signed.

In the related art, a professional auditor is usually required to perform manual auditing on the contract, but the efficiency of the manual auditing is low, and diversified contract auditing requirements are difficult to meet. Moreover, the standardization degree of the manual auditing process is low, so that the standardization of the contract is difficult to guarantee, and the two parties of the contract can be damaged in different degrees.

Disclosure of Invention

The embodiment of the application provides a contract processing method, a contract processing device, contract processing equipment and a storage medium based on RPA, and the method, the device and the storage medium are used for solving the technical problems that in the prior art, when the contract is checked and approved manually, the efficiency is low, the standardization degree is low, and the standardization of the contract is difficult to guarantee.

In a first aspect, the present application provides a contract processing method based on an RPA, applied to an RPA system, the contract processing method including: identifying first target information in the contract to be processed through a neural network model, wherein the first target information comprises type information and information to be checked of the contract to be processed, and the neural network model is obtained through training preset information of a sample contract; acquiring first verification information corresponding to the contract to be processed according to the type information; and auditing the information to be audited according to the first check information to obtain a target audit result corresponding to the contract to be audited.

Optionally, the auditing the information to be audited according to the first verification information to obtain a target audit result corresponding to the contract to be audited includes: auditing the information to be audited through the first check information to obtain a first audit result; responding to the first auditing result that the to-be-processed contract does not pass the auditing, and outputting prompt information, wherein the prompt information is used for indicating an auditor to audit the to-be-processed contract; acquiring second audit information, wherein the second audit information is obtained by auditing to-be-audited information by an auditor based on the verification information; and determining that the second audit information is a target audit result corresponding to the contract to be processed.

Optionally, the contract processing method further includes: and determining that the first auditing result is the target auditing result of the contract to be processed in response to the first auditing result being that the contract to be processed passes auditing.

Optionally, the contract processing method further includes: responding to the first auditing result that the to-be-processed contract does not pass the auditing and the second auditing result that the to-be-processed contract passes the auditing, and acquiring second target information of the to-be-processed contract; training the neural network model based on the second target information.

In a second aspect, an embodiment of the present application provides a neural network model training method based on an RPA, which is applied to an RPA system, and the neural network model training method includes: acquiring preset information corresponding to a sample contract, wherein the preset information comprises first target information; acquiring second check information corresponding to the sample contract, wherein the second check information is used for indicating the authenticity of the preset information; and training the initial neural network model based on the preset information and the second check information to obtain the neural network model.

Optionally, obtaining preset information corresponding to the sample contract includes: acquiring preset information corresponding to a sample contract by adopting at least one of an image Recognition technology, an Optical Character Recognition technology (OCR) and a Character Recognition technology; and/or; and acquiring preset information input by a user.

In a third aspect, the present application provides a contract processing apparatus based on RPA, applied to an RPA system, the contract processing apparatus including: the identification module is used for identifying first target information in the contract to be processed through a neural network model, the first target information comprises type information and information to be checked of the contract to be processed, and the neural network model is obtained through training of preset information of a sample contract; the acquisition module is used for acquiring first check information corresponding to the contract to be processed according to the type information; and the auditing module is used for auditing the information to be audited according to the first check information to obtain a target auditing result corresponding to the contract to be audited.

Optionally, the auditing module is specifically configured to: auditing the information to be audited through the first verification information to obtain a first audit result; responding to the first auditing result that the to-be-processed contract does not pass the auditing, and outputting prompt information, wherein the prompt information is used for indicating an auditor to audit the to-be-processed contract; acquiring second audit information, wherein the second audit information is obtained by auditing to-be-audited information by an auditor based on the verification information; and determining that the second audit information is a target audit result corresponding to the contract to be processed.

Optionally, the contract processing apparatus further includes: and the determining module is used for responding to the first auditing result that the to-be-processed contract passes the auditing and determining that the first auditing result is the target auditing result of the to-be-processed contract.

Optionally, the contract processing apparatus further includes: the training module is used for responding to the first auditing result that the to-be-processed contract does not pass the auditing and responding to the second auditing result that the to-be-processed contract passes the auditing, and acquiring second target information of the to-be-processed contract; training the neural network model based on the second target information.

In a fourth aspect, an embodiment of the present application provides an RPA-based neural network model training apparatus, including: the acquisition module is used for acquiring preset information corresponding to the sample contract and second check information corresponding to the sample contract, wherein the preset information comprises first target information, and the second check information is used for indicating the authenticity of the preset information; and the training module is used for training the initial neural network model based on the preset information and the second check information to obtain the neural network model.

Optionally, the obtaining module is specifically configured to collect preset information corresponding to the sample contract by using at least one of an image recognition technology, an optical character recognition technology OCR and a character recognition technology; and/or; and acquiring preset information input by a user.

In a fifth aspect, the present application provides an electronic device, comprising:

a memory, a processor, and a computer program stored on the memory and executable on the processor; the computer program, when executed by the processor, implements the steps of the contract processing method as defined in any one of the first aspects, and/or implements the steps of the neural network model training method as defined in any one of the second aspects.

In a sixth aspect, the present application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the contract processing method according to any one of the first aspect, and/or performs the steps of the neural network model training method according to any one of the second aspect.

In a seventh aspect, the present application provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of the contract processing method according to any one of the first aspect and/or performs the steps of the neural network model training method according to any one of the second aspect.

The contract processing method, the contract processing device, the contract processing equipment and the storage medium are applied to an RPA system, and the contract processing method comprises the following steps: and identifying first target information in the contract to be processed through the neural network model, acquiring first verification information corresponding to the contract to be processed according to the type information, and finally verifying the information to be verified according to the first verification information to obtain a target verification result corresponding to the contract to be processed. In the scheme provided by the embodiment of the application, the contract is audited based on the RPA technology, the automation degree is high, the processing efficiency of the contract can be greatly improved, the first target information in the contract is identified by adopting the neural network model, the accuracy of the identification result can be improved, the standardization of the contract is further improved, and the reasonable rights and interests of both parties of the contract are guaranteed.

Drawings

The above and other objects, features and advantages of exemplary embodiments of the present application will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the present application are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:

fig. 1 is a schematic view of an application scenario of a contract processing method provided in an embodiment of the present application;

FIG. 2 is a first schematic flowchart of a contract processing method according to an embodiment of the present application;

FIG. 3 is a second schematic flowchart of a contract processing method according to an embodiment of the present application;

fig. 4 is a schematic flowchart of a neural network model training method according to an embodiment of the present disclosure;

FIG. 5 is a schematic structural diagram of a contract processing apparatus provided in an embodiment of the present application;

fig. 6 is a schematic structural diagram of a neural network model training apparatus according to an embodiment of the present disclosure;

fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the application.

In the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The terms "first," "second," and the like in the description of the present application and in the above-described figures are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein.

It should be understood that, in the various embodiments of the present application, the size of the serial number of each process does not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

It should be understood that, in this application, "comprises" and "comprising," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Robot Process Automation (RPA) is a novel artificial intelligence virtual Process Automation robot, and with the development of economic globalization, the RPA is widely applied to various fields requiring Process Automation, especially in the field of finance and tax Automation.

In the related art, a professional auditor is usually required to manually audit the contract, but the efficiency of manual audit is low, and diversified contract audit requirements are difficult to meet. Moreover, the standardization degree in the manual auditing process is low, so that the standardization of the contract is difficult to guarantee, and the two parties for establishing the contract are lost in different degrees.

In view of this, the present application provides a contract processing method, apparatus, device and storage medium based on RPA, which implement contract processing based on RPA technology, and because the degree of automation of RPA technology is high, the contract processing efficiency can be greatly improved, and the labor cost can be reduced. Meanwhile, the first target information in the contract is identified by adopting the neural network model, the acquired first target information is more accurate, and when the contract is checked based on the first target information, the accuracy of the checking result can be improved, the normalization of the contract can be improved, and the reasonable rights and interests of both parties of the contract can be guaranteed.

The technical solution of the present application will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.

Fig. 1 is a schematic view of an application scenario of a contract processing method according to an embodiment of the present application. As shown in fig. 1, the applicable scenarios of the embodiment of the present application include: the terminal device 101.

In some embodiments, the terminal device 101 may be any device such as a desktop computer, a notebook, a Personal Digital Assistant (PDA), a smart phone, a tablet computer, and the like, and it should be understood that the terminal device 101 in fig. 1 is an example of a desktop computer, but is not limited thereto.

At least one RPA system is operated on the terminal device 101.

It should be noted that the specific form of the RPA system in the embodiment of the present invention is not limited, for example, the RPA system may be integrated in the terminal device 101, may be embedded in the terminal device 101 in the form of a plug-in, or the like, or may be independent from the terminal device 101 as long as the RPA system can automatically access the related functions of the terminal device 101.

In an application of the present application, an RPA program may be configured in the terminal device 101, so that the terminal device 101 may simulate a mouse, a keyboard, and other operations of a user to automatically access a related function of the terminal device 101 according to a rule set in the RPA program, thereby triggering the terminal device 101 to perform corresponding processing on the same.

Specifically, in the process of processing the contract, the terminal device 101 identifies the relevant information on the contract through the neural network model based on the RPA system, and audits the relevant information through a preset audit rule, thereby implementing the audit on the contract. Next, the above-described embodiment will be described with reference to specific examples.

Fig. 2 is a first flowchart illustrating a contract processing method according to an embodiment of the present application. The main body of the method related to the embodiment of the present application may be the terminal device 101 referred to above. As shown in fig. 2, the contract processing method provided in the embodiment of the present application specifically includes the following steps:

s201, identifying first target information in the contract to be processed through a neural network model.

The first target information comprises type information and information to be checked of the contract to be processed, and the neural network model is obtained by training through preset information of the sample contract.

It should be understood that the type information is used to indicate a contract type to which the pending contract belongs, and the embodiments of the present application are not particularly limited to the contract type of the pending contract, for example, the pending contract includes, but is not limited to, any one or more of the following: the buying and selling contract, the borrowing contract, the leasing contract and the entrusting contract are the same.

The information to be audited is the content to be audited in the contract to be processed, the corresponding information to be audited is different for different types of contracts, and the embodiment of the application is not specifically limited for the specific type of the information to be audited. In some embodiments, the information to be audited includes, but is not limited to, any one or more of the following: the information about the party (e.g., the party's name, address, contact, etc.), date information (e.g., the date of contract, date of fulfillment, expiration date, etc.), contract-related product information (e.g., the product's name, model, specification, variety, grade, color, quantity, unit price, etc.), price or consideration information (e.g., the cost of the buy-sell contract, the rent of the lease contract, the deposit, the principal and interest paid by the borrower to the borrower in the borrowing contract, etc.), etc.

In one implementation of the present application, the pending contract may be an electronic contract, and the electronic contract may be in a variety of formats, for example, may be in a text format, an image format, and the like.

In another implementation of the present application, the pending contract can also be a paper contract for the entity.

It should be understood that, for the to-be-processed contracts of different formats, the way of acquiring the first target information is also different, and when the neural network model is trained, the above-mentioned contracts of multiple formats and multiple types can be used as training samples, so that the neural network model can realize the recognition function of the contracts of multiple types, and the applied scheme can be flexibly applied to multiple contract recognition scenarios, for example, a remote contract recognition scenario, and the like.

Specifically, in the first aspect, if the to-be-processed contract is an electronic contract in a text format, the first target information may be acquired by using a text recognition technology based on a neural network model;

in a second aspect, if the to-be-processed contract is an electronic contract in an image format, the first target information may be acquired by using an image recognition technology based on a neural network model;

in a third aspect, if the pending contract is a paper contract of an entity, an image acquisition device of the terminal device may be used to perform image acquisition or text recognition on the paper contract, so as to obtain an electronic contract corresponding to the paper contract, and then based on the neural network model, the first target information in the electronic contract is obtained by using the method of the first aspect or the second aspect.

The inventors have discovered that different types of contracts correspond to different keywords, and that the keywords are typically located in fixed locations, e.g., for a rental contract, there are typically "lease, rent, lease" and the like keywords on the top page (or other designated page) of the pending contract; for a trade contract, there are keywords such as "buy and sell" on the first page of the pending contract. In view of this, in an optional implementation manner, keyword identification may be performed on a specific page of the contract to be processed, so as to determine the type to which the contract to be processed belongs according to the correspondence between the keyword and the contract type, thereby obtaining the type information of the contract to be processed.

In an optional implementation manner, for various types of contracts, unified preset audit information may be set, where the preset audit information may include key information that may appear in the various contracts as much as possible, and when acquiring the to-be-audited information of the to-be-processed contract, the to-be-audited information corresponding to the preset audit information is acquired by performing a comprehensive search in the contract according to the preset audit information.

By the method, important information in the to-be-processed contract can be avoided from being omitted, the condition that the to-be-processed contract is not checked in place due to the missing information is avoided, and the strictness of the to-be-processed contract checking process is improved.

In another optional implementation manner, the information to be audited may also be different for different types of contracts, and in this application embodiment, the type information of the contract to be audited may be determined first, then the information to be audited corresponding to the type information may be determined based on the type information, and the information to be audited may be identified in the contract to be audited.

Still taking the above as an example, for a lease contract, the corresponding information to be audited may include, but is not limited to, any one or more of the following: the related information of the renting object, the information of the party, the rent, the deposit, the renting date and the like; for a sales contract, the corresponding information to be audited may include, but is not limited to, any one or more of the following: product information, party information, contract fulfillment date, and the like.

In the embodiment of the application, different information to be checked is set based on different types of contracts, and in the identification process, only the information to be checked is required to be identified, and the content of the whole contract is not required to be identified, so that the identification efficiency can be improved, and the processing efficiency of the contract is further improved.

In addition, the trained neural network model is adopted to complete the recognition process, so that the accuracy of the recognition result can be guaranteed, and the recognition efficiency is improved.

S202, acquiring first verification information corresponding to the contract to be processed according to the type information.

And S203, auditing the information to be audited according to the first check information to obtain a target audit result corresponding to the contract to be audited.

It should be noted that the first verification information is used for verifying the information to be verified, and because the contents to be verified in the different types of contracts are different, in the embodiment of the present application, the first verification information corresponding to the contract to be processed may be determined according to the type information.

For the lease contract, the first verification information may include, but is not limited to, any one or more of the following items corresponding to the information to be audited: the method comprises the following steps of (1) verifying information of a rented object, and verifying information of a party, rent, deposit, rental date and the like; for a sales contract, the corresponding information to be audited may include, but is not limited to, any one or more of the following: product information, party information, contract fulfillment date, etc.

As an alternative implementation manner, the first verification information may be preset by the relevant person, for example, before the pending contract is processed, the relevant person may set real verification information for each audit item in the pending contract, for example, taking the pending contract as a lease contract as an example, the verification information of the lease object, the verification information and the lease of the party, the specific amount of the deposit, the specific lease date, and the like may be set in advance.

Specifically, when the information to be checked is checked based on the first check information, the target check result of the contract to be processed can be obtained according to the matching degree between the information to be checked and the corresponding first check information.

As another alternative, the first verification information may be obtained from other documents. For example, for a sales contract, there may be associated transaction documents, such as invoices, receipts, etc., which typically contain the same information as in the pending contract, such as party information, payment amount, transaction date, and product type.

In the embodiment of the application, as the reliability of the related information on the transaction document is higher, the information on the document is used as the verification information to verify the agreement, the reliability of the verification result can be further ensured, the verification information does not need to be manually input in the process, the human resources can be liberated, and the contract verification efficiency is improved.

In the scheme provided by the embodiment of the application, the contract is processed based on the RPA technology, and the automation degree of the RPA technology is higher, so that the processing efficiency of the contract can be greatly improved, and the labor cost is reduced. Meanwhile, the first target information in the contract is identified by adopting the neural network model, the acquired first target information is more accurate, and in the process of auditing the contract based on the first target information, the accuracy of an auditing result can be improved, so that the normalization of the contract is improved, and the reasonable rights and interests of both parties of contract establishment are guaranteed.

Next, the scheme of the above embodiment will be described in more detail with reference to fig. 3. Fig. 3 is a schematic flow chart diagram of a contract processing method according to an embodiment of the present application. As shown in fig. 3, the contract processing method provided in the embodiment of the present application specifically includes the following steps:

s301, identifying first target information in the contract to be processed through the neural network model.

The first target information comprises type information and information to be checked of the contract to be processed, and the neural network model is obtained by training through preset information of the sample contract;

s302, acquiring first verification information corresponding to the contract to be processed according to the type information.

And S303, auditing the information to be audited through the first verification information to obtain a first audit result.

It should be noted that, in the embodiment of the present application, implementation principles and advantageous effects of steps S301 to S303 are similar to those of steps S201 to S203 in the embodiment shown in fig. 2, and a manner of obtaining the first audit result may refer to a manner of obtaining the target audit result in the embodiment, and specifically, refer to the embodiment, and details are not repeated here.

S304, determining whether the first checking result is that the to-be-processed contract passes the check.

It should be understood that, since the to-be-checked information of the contract has a great influence on both parties of the contract, in the embodiment of the present application, when all the to-be-checked information of the to-be-processed contract passes the check, it may be determined that the first checking result is that the to-be-processed contract passes the check, so as to guarantee the normalization of the contract to the greatest extent and guarantee the reasonable rights of both parties.

Optionally, verification items may also be preset, and at least when the verification items pass verification, the first review result is determined as that the pending contract passes review.

S305, responding to the first auditing result that the contract to be processed passes the auditing, and determining that the first auditing result is the target auditing result of the contract to be processed.

S306, responding to the first examination result that the to-be-processed contract does not pass the examination, and outputting prompt information.

The prompt information is used for instructing an auditor to audit the contract to be processed.

The embodiment of the present application is not particularly limited as to the manner of outputting the prompt information. For example, the prompt information may be sent to an electronic device such as a mobile phone or a computer of the associated user, or may also be sent to a preset account.

Optionally, the related information that fails to pass the audit may be indicated in the prompt information, for example, the items that fail to pass the audit, the reason why the check corresponding to each item fails, the position of each item in the contract, and the like. By the scheme, the user can quickly determine the related information which does not pass the audit, so that the response is quickly realized, and the contract processing efficiency is improved.

And S307, acquiring second review information.

And the second audit information is obtained by auditing the information to be audited by the auditor based on the verification information.

Specifically, when receiving the prompt information, the related user can perform manual review on the to-be-processed contract according to the related information in the prompt information to obtain a second review result.

On one hand, the user can directly perform the auditing operation on the device receiving the prompt information, and after the auditing is completed, the second auditing result is sent to the terminal device in the embodiment of the application. By the scheme, the situation that the user cannot directly use the terminal equipment to perform auditing operation can be solved, and the method and the device can be flexibly applied to a remote auditing scene of a contract.

On the other hand, after receiving the prompt message, the user can directly perform auditing on the terminal device provided in the embodiment of the application, and output a second auditing result through the terminal device.

And S308, determining whether the second checking result is that the to-be-processed contract passes the check.

S309, responding to the second auditing result that the to-be-processed contract passes the auditing, and determining that the second auditing information is a target auditing result corresponding to the to-be-processed contract.

And S310, acquiring second target information of the contract to be processed.

It should be noted that, if the pending contract does not pass the audit when the RPA system is used for the audit, and the pending contract may pass the audit when the manual audit is used, the first target information identified by the neural network model may be incorrect, so that the first audit result is inaccurate.

In view of this, in the embodiment of the present application, the neural network model may be trained by using accurate information corresponding to the contract to be processed, so as to further optimize the neural network model.

In one aspect, the second target information may be obtained from the pending contract by the user and manually entered into the terminal device.

On the other hand, the second target information may also be obtained from other documents corresponding to the contract to be processed by the terminal device, where the documents are, for example, receipts, invoices, and the like, and this embodiment of the present application is not specifically limited, and as for the manner of obtaining the second target information obtained from the documents, the manner of obtaining the second target information is similar to the manner of obtaining the verification information in the embodiment shown in fig. 2, and details are not repeated here.

And S311, training the neural network model based on the second target information.

It should be noted that, as for the training mode of the neural network model, it is shown in the embodiment shown in fig. 4, and details are not described here.

And S312, in response to the second auditing result that the to-be-processed contract fails to be audited, determining that the target auditing result is that the to-be-processed contract fails to be audited.

In the embodiment of the application, when the RPA system is adopted for auditing, the to-be-processed contract does not pass auditing, and when manual auditing is adopted, the to-be-processed contract passes auditing, the accurate information of the contract can be adopted for carrying out optimization training on the neural network model, the reliability of the neural network model can be further improved, the relevant information of the contract can be accurately identified, and therefore the normalization of the contract is improved.

Fig. 4 is a schematic flowchart of a neural network model training method according to an embodiment of the present application. The main body of the method related to the embodiment of the present application may be the terminal device 101 referred to above. As shown in fig. 4, the neural network model training method provided in the embodiment of the present application specifically includes the following steps:

s401, obtaining preset information corresponding to the sample contract.

It should be noted that, for the type of the sample contract, reference may be made to the above description about the type of the contract to be processed, and details are not described herein.

The preset information includes first target information, that is, when the neural network model is trained, the target information to be identified needs to be adopted to train the neural network model, so that the neural network model has the capability of identifying the target information.

Optionally, the preset information may further include information of other categories in the contract, for example, interference information, and the like, and the embodiment of the present application is not specifically limited.

S402, second check-up information corresponding to the sample contract is obtained.

The second check information is used for indicating authenticity corresponding to each preset information, and the second check information can be obtained through manual marking. Specifically, when the preset information is real information, the preset information is marked as a positive sample, and when the preset information is error information such as interference information, the preset information is marked as a negative sample.

And S403, training the initial neural network model based on the preset information and the second check information to obtain the neural network model.

Specifically, preset information and second check information are input into the initial neural network model, iterative training is performed based on the preset information and the second check information, a loss function corresponding to each training is obtained, when the loss function corresponding to any training round meets a preset requirement, the training is finished, the neural network model corresponding to the model parameter of the current training round is determined to be a final model, and details are not repeated in a specific training mode in the embodiment of the application.

In a first alternative implementation, at least one of an image recognition technology, an optical character recognition technology, an OCR technology, and a character recognition technology may be adopted to collect the preset information corresponding to the sample contract.

Specifically, the agreement may be recognized by using at least one of an image recognition technique, an optical character recognition technique OCR and a character recognition technique, the sample agreement may be converted into a text format based on the recognition result, and then the preset information corresponding to the preset keyword in the text format may be acquired based on the preset keyword by using the character recognition technique.

Illustratively, in recognition using OCR technology, the shape of each character in the sample contract is first determined by detecting dark and light patterns, and then the shape is translated into computer text using character recognition methods. Specifically, the characters in the paper document can be converted into an image file of a black-and-white dot matrix based on an optical mode, and the characters in the image are converted into a text format through recognition software, so that the preset information is obtained based on the image in the text format.

In a second alternative embodiment, the preset information may also be obtained manually. The embodiments of the present application are not described in detail.

In the embodiment of the application, the acquisition mode of multiple preset information is provided, the acquisition requirements of different terminal devices can be met, meanwhile, the efficiency and the accuracy of the acquisition mode can be guaranteed, the accuracy of acquiring target information by a neural network model is improved, the accuracy of a contract auditing result is improved, and the normalization of a contract is further guaranteed.

Fig. 5 is a schematic structural diagram of a contract processing apparatus according to an embodiment of the present application. As shown in fig. 5, the contract processing apparatus 500 provided in the embodiment of the present application is applied to an RPA system, and the contract processing apparatus 500 specifically includes:

the identification module 501 is configured to identify first target information in the to-be-processed contract through a neural network model, where the first target information includes type information and to-be-checked information of the to-be-processed contract, and the neural network model is obtained by training preset information of a sample contract; an obtaining module 502, configured to obtain, according to the type information, first verification information corresponding to the contract to be processed; the auditing module 503 is configured to audit the information to be audited according to the first check information, and obtain a target auditing result corresponding to the contract to be audited.

Optionally, the auditing module 503 is specifically configured to: auditing the information to be audited through the first check information to obtain a first audit result; responding to the first examination result that the contract to be processed does not pass the examination, and outputting prompt information, wherein the prompt information is used for indicating an examiner to examine the contract to be processed; acquiring second audit information, wherein the second audit information is obtained by auditing to-be-audited information by an auditor based on the verification information; and determining that the second audit information is a target audit result corresponding to the contract to be processed.

Optionally, the contract processing apparatus 500 further includes: the determining module 504 is configured to determine that the first review result is a target review result of the to-be-processed contract in response to that the first review result is that the to-be-processed contract passes the review.

Optionally, the contract processing apparatus 500 further includes: the training module 505 is configured to, in response to that the first audit result indicates that the to-be-processed contract does not pass the audit, and that the second audit result indicates that the to-be-processed contract passes the audit, obtain second target information of the to-be-processed contract; training the neural network model based on the second target information.

It should be noted that the contract processing apparatus provided in this embodiment may be configured to execute the steps of the contract processing method provided in any one of the foregoing method embodiments, and the implementation principle and the technical effect are similar, which are not described herein again.

Fig. 6 is a schematic structural diagram of a neural network model training device according to an embodiment of the present application. As shown in fig. 6, the neural network model training apparatus 600 provided in the embodiment of the present application is applied to an RPA system, and the neural network model training apparatus 600 may include:

the acquiring module 601 is configured to acquire preset information corresponding to a sample contract and second check information corresponding to the sample contract, where the preset information includes first target information; the training module 602 is configured to train the initial neural network model based on preset information and second check information, so as to obtain a neural network model.

Optionally, the obtaining module 601 is specifically configured to collect preset information corresponding to the sample contract by using at least one of an image recognition technology, an optical character recognition technology OCR and a character recognition technology; and/or; and acquiring preset information input by a user.

It should be noted that the neural network model training device provided in this embodiment may be used to perform the steps of the neural network model training method provided in any of the foregoing method embodiments, and the implementation principle and the technical effect are similar, which are not described herein again.

Fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 7, the electronic device 700 provided by the present application may include:

as shown in fig. 7, the electronic device 700 is embodied in the form of a general-purpose electronic device. The components of the electronic device 700 may include, but are not limited to: the at least one processing unit 701 and the at least one memory unit 702 are connected to a bus 703 that connects different system components (including the processing unit 701 and the memory unit 702).

The bus 703 includes a data bus, a control bus, and an address bus. The storage unit 702 may include readable media in the form of volatile memory, such as Random Access Memory (RAM)712 and/or cache memory 722, and may further include readable media in the form of non-volatile memory, such as Read Only Memory (ROM) 732.

Storage unit 702 can also include a program/utility 752 having a set (at least one) of program modules 742, such program modules 742 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.

The electronic device 700 may also communicate with one or more external devices 704 (e.g., a keyboard, a pointing device, etc.). Such communication may occur via an input/output (I/O) interface 707. Also, the electronic device 700 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet) via the network adapter 706. As shown in FIG. 7, the network adapter 706 communicates with the other modules of the electronic device 700 over the bus 703.

It should be understood that although not shown in the figures, other hardware and/or software modules may be used in conjunction with electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others. For the implementation principle and the technical effect of the electronic device provided by this embodiment, reference may be made to the foregoing embodiments, which are not described herein again.

The embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the contract processing method according to any one of the foregoing embodiments.

Embodiments of the present application further provide a computer program product, which includes a computer program, and when being executed by a processor, the computer program implements the steps of the contract processing method provided in any of the foregoing embodiments, and/or implements the steps of the neural network model training method provided in any of the foregoing embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the division of modules is only one logical division, and other divisions may be realized in practice, for example, a plurality of modules may be combined or integrated into another system, or some features may be omitted, or not executed.

The integrated module implemented in the form of a software functional module may be stored in a computer-readable storage medium. The software functional module is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor to execute some steps of the methods according to the embodiments of the present application.

It should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of the hardware and software modules within the processor.

The memory may comprise a high-speed RAM memory, and may further comprise a non-volatile storage NVM, such as at least one disk memory, and may also be a usb disk, a removable hard disk, a read-only memory, a magnetic or optical disk, etc.

The storage medium may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.

An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the storage medium may reside as discrete components in an electronic device or host device.

It should be noted that, in this document, 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 an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are included in the scope of the present application.

Claims

1. A contract processing method based on RPA is characterized in that the contract processing method is applied to a Robot Process Automation (RPA) system, and comprises the following steps:

identifying first target information in a contract to be processed through a neural network model, wherein the first target information comprises type information and information to be checked of the contract to be processed, and the neural network model is obtained through training preset information of a sample contract;

acquiring first verification information corresponding to the contract to be processed according to the type information;

and auditing the information to be audited according to the first verification information to obtain a target auditing result corresponding to the contract to be processed.

2. The contract processing method according to claim 1, wherein the auditing the information to be audited according to the first verification information to obtain a target auditing result corresponding to the contract to be processed includes:

auditing the information to be audited through the first verification information to obtain a first audit result;

responding to the first examination result that the to-be-processed contract does not pass examination, and outputting prompt information, wherein the prompt information is used for indicating an examiner to examine the to-be-processed contract;

acquiring second audit information, wherein the second audit information is obtained by auditors auditing the information to be audited based on the verification information;

and determining that the second auditing information is a target auditing result corresponding to the contract to be processed.

3. The contract processing method according to claim 2, further comprising:

and determining that the first auditing result is a target auditing result of the to-be-processed contract in response to the first auditing result being that the to-be-processed contract passes auditing.

4. The contract processing method according to claim 3, further comprising:

responding to the first auditing result that the contract to be processed does not pass the auditing and the second auditing result that the contract to be processed passes the auditing, and acquiring second target information of the contract to be processed;

training the neural network model based on the second target information.

5. A neural network model training method based on RPA is characterized in that the neural network model training method is applied to an RPA system and comprises the following steps:

acquiring preset information corresponding to a sample contract, wherein the preset information comprises first target information;

acquiring second check-up information corresponding to the sample contract, wherein the second check-up information is used for indicating the authenticity of the preset information;

and training an initial neural network model based on the preset information and the second check information to obtain the neural network model.

6. The neural network model training method of claim 5, wherein the obtaining of the preset information corresponding to the sample contract comprises:

acquiring preset information corresponding to the sample contract by adopting at least one of an image recognition technology, an optical character recognition technology (OCR) and a character recognition technology; and/or; and acquiring the preset information input by the user.

7. A contract processing apparatus based on RPA is characterized in that, applied to the RPA system, the contract processing apparatus includes:

the identification module is used for identifying first target information in the contract to be processed through a neural network model, wherein the first target information comprises type information and information to be checked of the contract to be processed, and the neural network model is obtained through training of preset information of a sample contract;

the acquisition module acquires first verification information corresponding to the contract to be processed according to the type information;

and the auditing module is used for auditing the information to be audited according to the first verification information to obtain a target auditing result corresponding to the contract to be processed.

8. A neural network model training device based on RPA is characterized in that, when applied to an RPA system, the neural network model training device comprises:

the system comprises an acquisition module, a verification module and a verification module, wherein the acquisition module is used for acquiring preset information corresponding to a sample contract and second verification information corresponding to the sample contract, the preset information comprises first target information, and the second verification information is used for indicating the authenticity of the preset information;

and the training module is used for training the initial neural network model based on the preset information and the second check information to obtain the neural network model.

9. An electronic device, characterized in that the electronic device comprises:

a memory, a processor, and a computer program stored on the memory and executable on the processor;

the computer program, when being executed by the processor, realizes the steps of the contract processing method according to any one of claims 1-4 and/or realizes the steps of the neural network model training method according to claim 5 or 6.

10. A computer-readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, carries out the steps of the contract processing method according to one of claims 1 to 4 and/or the steps of the neural network model training method according to one of claims 5 or 6.