File approval method and related device

Abstract

The invention belongs to the technical field of information processing, and discloses a file approval method and a related device, and the method comprises the steps: receiving a to-be-approved file submitted by an applicant, and extracting the preset file attribute information of the to-be-approved file; encrypting the to-be-approved file to obtain an encrypted to-be-approved file; generating an optimal approval path based on the approval path dynamic planning model in combination with preset file attribute information of the to-be-approved file; the approval path dynamic planning model is a pre-trained reinforcement learning model; transferring the encrypted to-be-approved file according to the optimal approval path, and distributing the to-be-approved file to the approval personnel of each approval level in the optimal approval path for approval until the approval personnel of all approval levels in the optimal approval path complete approval; according to the method, the reinforcement learning model is used as the dynamic approval path planning model, and the optimal approval path is accurately generated, so that the approval process can plan a reasonable circulation sequence according to the characteristics of the approval file, and the file approval efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of information processing, and particularly relates to a file examination and approval method and related device. BACKGROUND

[0002] With the acceleration of digital office, the traditional paper file examination and approval method cannot meet the efficient and safe office examination and approval demand; with the development of technology, the electronic examination and approval method has been widely applied; at present, in the electronic examination and approval process, the examination and approval flow table is sent to the examination and approval personnel one by one according to the examination and approval level, for example, Chinese patent application “a remote examination and approval system for important files” (publication number CN107169730A); however, the existing electronic examination and approval system cannot dynamically generate the optimal examination and approval path according to the file content, organization relationship and business rules, resulting in low examination and approval efficiency, especially when dealing with cross-department and multi-level examination and approval. SUMMARY

[0003] In view of the technical problems in the prior art, the present application provides a file examination and approval method and related device to solve the technical problem that the existing electronic examination and approval system cannot generate the optimal examination and approval path, resulting in low examination and approval efficiency.

[0004] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows: The present application provides a file examination and approval method, comprising: receiving a to-be-examined-and-approved file submitted by an applicant and extracting preset file attribute information of the to-be-examined-and-approved file; encrypting the to-be-examined-and-approved file to obtain an encrypted to-be-examined-and-approved file; generating an optimal examination and approval path based on an examination and approval path dynamic programming model in combination with the preset file attribute information of the to-be-examined-and-approved file; wherein the examination and approval path dynamic programming model is a pre-trained reinforcement learning model; circulating the encrypted to-be-examined-and-approved file according to the optimal examination and approval path and distributing it to the examination and approval personnel of each examination and approval level in the optimal examination and approval path for examination and approval until the examination and approval personnel of all examination and approval levels in the optimal examination and approval path complete the examination and approval.

[0005] Further, the preset file attribute information of the to-be-examined-and-approved file includes the file type, sensitive level and content keyword feature of the to-be-examined-and-approved file; wherein the sensitive level includes public, confidential and top secret.

[0006] Further, the process of generating the optimal examination and approval path based on the examination and approval path dynamic programming model in combination with the preset file attribute information of the to-be-examined-and-approved file is as follows: determining the required number of examination and approval levels of the to-be-examined-and-approved file according to the sensitive level of the to-be-examined-and-approved file; Based on the file type of the to-be-approved file, the required approval level number of the to-be-approved file, the content keyword characteristics of the to-be-approved file, the enterprise organizational structure relationship and the predetermined business rules, the approval path planning is carried out by using the approval path dynamic planning model, and a candidate approval path is obtained. The candidate approval path is checked for logical consistency and compliance, and an optimal approval path is obtained. For each approval level in the optimal approval path, a person who meets the authority requirement is assigned. The matching score of each person who meets the authority requirement is calculated, and the preset number of persons with the highest matching score are selected as the approval personnel of the corresponding approval level.

[0007] Further, the process of calculating the matching score of each person who meets the authority requirement is as follows:

[0008] Among them, is the matching score of the person who meets the authority requirement; is the matching degree of the authority level of the person and the current approval level; is the current task load of the person; is the association degree of the person and the department to which the file belongs; , , are weight coefficients.

[0009] Further, the encrypted to-be-approved file is forwarded according to the optimal approval path and distributed to the approval personnel of each approval level in the optimal approval path for approval, and after all the approval personnel in the optimal approval path complete the approval, it also includes: According to the risk level of the current business scene, a multi-factor authentication mechanism is adopted to perform electronic signature on the file after completing the approval; wherein the multi-factor mechanism includes biometric authentication, dynamic password authentication, fingerprint authentication and video live body detection authentication.

[0010] Further, before the to-be-approved file is encrypted to obtain the encrypted to-be-approved file, it also includes: The file format, desensitization state of sensitive information and file content of the to-be-approved file are checked for compliance.

[0011] The application also provides a file approval system, which comprises: A file preprocessing module is configured to receive a to-be-approved file submitted by an applicant and extract preset file attribute information of the to-be-approved file; encrypt the to-be-approved file to obtain an encrypted to-be-approved file. The approval path generation module is configured to generate an optimal approval path based on an approval path dynamic programming model and in combination with preset file attribute information of the to-be-approved file; and the approval path dynamic programming model is a pre-trained reinforcement learning model. The remote approval module is configured to circulate the encrypted to-be-approved file according to the optimal approval path, and distribute the to-be-approved file to the approval personnel of each approval level in the optimal approval path for approval until all the approval personnel of all the approval levels in the optimal approval path complete the approval.

[0012] The application further provides an electronic device, comprising: A processor adapted to execute a computer program; A computer readable storage medium having a computer program stored therein, wherein the computer program is executed by the processor to execute the file approval method.

[0013] The application further provides a computer readable storage medium having a computer program stored therein, wherein the computer program is executed by the processor to execute the file approval method.

[0014] The application further provides a computer program product comprising a computer program, wherein the computer program is executed by the processor to execute the file approval method.

[0015] Compared with the prior art, the application has the following beneficial effects: The file examination method provided by the application introduces a pre-trained reinforcement learning model as an examination path dynamic programming model, combines the extracted preset file attribute information of the to-be-examined file, and accurately generates an optimal examination path, so that the examination process can plan a reasonable flow sequence according to the characteristics of the to-be-examined file, breaking through the limitation that the traditional examination path is fixed or cannot be flexibly adjusted according to the actual situation, effectively improving the file examination efficiency, especially in the cross-department and multi-level examination scene, which can significantly shorten the examination period and improve the overall office efficiency. Specifically, the pre-trained reinforcement learning model is used as the examination path dynamic programming model, so that the model can adapt to different file attributes and examination scenes through continuous learning and optimization, automatically generate an optimal examination path, without manually setting a complex examination process, reducing manual intervention and reducing the possibility of human error, while improving the flexibility and adaptability of the examination process, so that the examination system can better cope with various complex business demands. Secondly, the to-be-examined file is encrypted to ensure the security of the file during transmission and flow, and to prevent the file content from being leaked or tampered without authorization. In addition, the encrypted to-be-examined file is transferred according to the generated optimal examination path and directly distributed to the examination personnel of each examination level, so that the file can quickly and accurately reach the hands of the appropriate examination personnel, avoiding the confusion and delay of the file during the examination process, and further improving the efficiency of the overall examination process.

[0016] Further, according to the risk level of the current business scene, a multi-factor authentication mechanism is used to electronically sign the file after examination, and the multi-factor authentication method has higher security than the single authentication method, which can more effectively prevent the electronic signature from being forged or used, ensure the authenticity and legality of the file examination, and enhance the security of the file during the entire examination process.

[0017] The file examination system, electronic device, computer readable storage medium and computer program product provided by the application have all the advantages of the above-mentioned file examination method. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 The flowchart of the file examination method provided for embodiment 1 is shown in the figure; Figure 2 The structural block diagram of the file examination system provided for embodiment 2 is shown in the figure; Figure 3 The structural block diagram of the electronic device provided for embodiment 3 is shown in the figure. DETAILED DESCRIPTION

[0019] In order to make the technical problems solved by the present application, technical solutions and beneficial effects clearer, the following specific embodiments are used to further explain the present application. It should be understood that the specific embodiments described herein are only used to explain the present application and not used to limit the present application.

[0020] The present application provides a file approval method, comprising the following steps: Step 100, receiving the to-be-approved file submitted by the applicant, and extracting the preset file attribute information of the to-be-approved file; encrypting the to-be-approved file to obtain an encrypted to-be-approved file.

[0021] Step 200, generating an optimal approval path based on an approval path dynamic programming model in combination with the preset file attribute information of the to-be-approved file; wherein the approval path dynamic programming model is a pre-trained reinforcement learning model.

[0022] Step 300, circulating the encrypted to-be-approved file according to the optimal approval path, and distributing it to the approval personnel of each approval level in the optimal approval path for approval until all the approval personnel of all the approval levels in the optimal approval path complete the approval.

[0023] Step 400, according to the risk level of the current business scenario, using a multi-factor authentication mechanism to perform electronic signature on the approved file; wherein the multi-factor mechanism includes biometric authentication, dynamic password authentication, fingerprint authentication and video liveness detection authentication.

[0024] In the above embodiment, the to-be-approved file is encrypted after being received and the preset attribute information is extracted, which ensures the security of the file; the pre-trained reinforcement learning model is used as the approval path dynamic programming model, and the optimal approval path is generated in combination with the file attribute information, which effectively solves the problem of low approval efficiency caused by the inability of the existing electronic approval system to dynamically generate an optimal path, especially when the approval is cross-department and multi-level; the file is circulated and distributed according to the optimal path, so that the approval process is smoother; according to the risk level of the business scenario, the multi-factor authentication mechanism is used for electronic signature, which further enhances the security and reliability of the file approval, and comprehensively improves the efficiency and security of the file approval.

[0025] The following is a further explanation of the file approval method provided by the present application in some specific embodiments: Embodiment 1 As shown in the accompanying Figure 1 Embodiment 1 provides a file approval method, comprising the following steps: Step 1, receiving the to-be-approved file submitted by the applicant, and extracting the preset file attribute information of the to-be-approved file; performing compliance verification on the file format, desensitization state of sensitive information, and file content of the to-be-approved file; encrypting the to-be-approved file to obtain an encrypted to-be-approved file.

[0026] Specifically, the steps are as follows: Step 11, receiving the to-be-approved file submitted by the applicant, and extracting the preset file attribute information of the to-be-approved file; wherein the preset file attribute information of the to-be-approved file includes the file type, sensitive level, and content keyword features of the to-be-approved file; the sensitive level includes public, confidential, and top secret.

[0027] Step 12, performing compliance verification on the file format, desensitization state of sensitive information, and file content of the to-be-approved file; wherein when performing compliance verification on the file format of the to-be-approved file, it includes confirming the matching state of the file extension and the actual content type, such as.pdf file should be binary PDF format, checking whether the file conforms to the characteristic structure, such as the syntax of XML / JSON, the column separator of CSV; when performing compliance verification on the desensitization state of sensitive information, using a preset custom dictionary for matching detection; when performing compliance verification on the file content, using regular expressions to verify the predetermined field range, format, and mandatory items in the file content.

[0028] Step 13, encrypting the to-be-approved file to obtain an encrypted to-be-approved file; specifically, generating a unique encryption key according to the identity information of the applicant; using the generated unique encryption key to encrypt the to-be-approved file to obtain an encrypted to-be-approved file; wherein the generated unique encryption key is bound to the identity of the applicant.

[0029] Step 2, based on the approval path dynamic planning model, combining the preset text attribute information of the to-be-approved file, generating an optimal approval path; wherein the approval path planning model is a pre-trained reinforcement learning model.

[0030] Specifically, the steps are as follows: Step 21, determining the required number of approval levels of the to-be-approved file according to the sensitive level of the to-be-approved file; wherein the required approval level of the to-be-approved file with a public sensitive level adopts one-level approval, i.e. the required number of approval levels of the to-be-approved file with a public sensitive level is 1; the required approval level of the to-be-approved file with a confidential sensitive level adopts two-level approval, i.e. the required number of approval levels of the to-be-approved file with a confidential sensitive level is 2; the required approval level of the to-be-approved file with a top secret sensitive level adopts three-level approval, i.e. the required number of approval levels of the to-be-approved file with a public sensitive level is 3.

[0031] Step 22, construct a reinforcement learning model and train the reinforcement learning model using pre-acquired historical file approval data to obtain a pre-trained reinforcement learning model as an approval path dynamic programming model; wherein the pre-acquired historical file approval data includes preset file attribute information of historical approval files, required approval level number, enterprise organizational structure relationship, business rules, approval path and approval result of historical approval files.

[0032] It should be noted that the approval path dynamic programming model is a reinforcement learning model trained by a large amount of data, which can generate a candidate approval path according to the required approval level number of the to-be-approved file, the file type of the to-be-approved file, the required approval level number of the to-be-approved file, the content keyword features of the to-be-approved file, the enterprise organizational structure relationship and the predetermined business rules; wherein the reinforcement learning model is, for example, a model based on deep Q network.

[0033] Step 23, based on the file type of the to-be-approved file, the required approval level number of the to-be-approved file, the content keyword features of the to-be-approved file, the enterprise organizational structure relationship and the predetermined business rules, using the approval path dynamic programming model to plan the approval path, and obtaining the candidate approval path; wherein the preset business rules include enterprise business reporting relationship, project collaboration network, enterprise compliance policy, approval authority matrix and special process exception rules; Specifically, the file type of the to-be-approved file, the required approval level number of the to-be-approved file, the content keyword features of the to-be-approved file, the enterprise organizational structure relationship and the predetermined business rules are input into the approval path dynamic programming model, and the approval path is planned by comprehensively evaluating the timeliness, compliance and resource consumption indicators, and the candidate approval path is generated and sorted in real time.

[0034] Step 24, checking the logical consistency and compliance of the candidate approval path to obtain the optimal approval path.

[0035] Step 25, assigning a person with authority to each approval level in the optimal approval path; wherein the first-level approval is performed by a person with junior or above authority, the second-level approval is performed by a person with intermediate or above authority, and the third-level approval is performed by a person with senior authority.

[0036] Step 26, calculating the matching score of each person with authority, and selecting a preset number of persons with the highest matching score as the approval personnel of the corresponding approval level; wherein the process of calculating the matching score of each person with authority is as follows:

[0037] wherein, a matching score of the authority level of the approver and the current approval level; a matching degree of the authority level of the approver and the current approval level; a current task load of the approver; a relevance of the approver and the department to which the file belongs; 、 、 are weight coefficients; preferably, 65%, 20%, 15%.

[0038] Step 3, the encrypted to-be-approved file is circulated according to the optimal approval path and distributed to the approvers at each approval level in the optimal approval path for approval until all the approvers at all the approval levels in the optimal approval path complete the approval.

[0039] Specifically, the encrypted to-be-approved file is circulated according to the optimal approval path and distributed to the approvers at the corresponding approval level in the optimal approval path for approval; after completing the approval operation at the current approval level, the approval operation at the next approval level is started until all the approval levels complete the approval.

[0040] In this embodiment 1, when the approvers at each approval level perform the approval, real-time coordinated decision-making of multiple approvers is supported; secondly, the operation authority of each approver is supported to be adjusted in real time according to the approval progress and the file state, and the main approver is supported to temporarily grant specific authority to other approvers; in addition, historical approval cases can be automatically matched based on the file content of the to-be-approved file, the potential compliance risks of the current approval decision are analyzed, and auxiliary decision support is provided for the approvers; and the whole process of collaborative decision-making can be recorded completely.

[0041] Step 4, according to the risk level of the current business scenario, a multi-factor authentication mechanism is adopted to perform electronic signature on the file that completes the approval. The multi-factor mechanism includes biometric authentication, dynamic password authentication, fingerprint authentication and video liveness detection authentication.

[0042] Specifically, the steps are as follows: Step 41, a behavior baseline model is established based on the historical operation behavior of the user; the risk level of the current business scenario is determined by using the behavior baseline model according to the real-time received external security threat data in combination with the sensitivity level and compliance requirements of the current business scenario; wherein, the risk level of the current business scenario includes a low-risk scenario, a medium-risk scenario and a high-risk scenario.

[0043] Step 42, according to the risk level of the current business scenario, determine the authentication mechanism; specifically, low-risk scenarios enable two-factor authentication mechanisms, which include biometric authentication and dynamic password authentication; medium-risk scenarios enable three-factor authentication mechanisms, which include biometric authentication, dynamic password authentication, and fingerprint authentication; high-risk scenarios enable four-factor authentication mechanisms, which include biometric authentication, dynamic password authentication, fingerprint authentication, and video liveness detection authentication.

[0044] Step 43, based on the determined authentication mechanism, digitally sign and apply electronic seals to the approved file.

[0045] Step 5, store the pre-determined approval key node information in the pre-set blockchain during the approval process, classify and store the approved file, and record the full-process operation log. The full-process operation log supports multi-dimensional query and analysis functions, including time, operation type, and user filtering queries.

[0046] The file approval method described in Embodiment 1 uses a pre-trained reinforcement learning model as an approval path dynamic programming model, automatically generates an approval path based on a reinforcement learning algorithm, effectively improves approval efficiency; checks the logical consistency and compliance of the approval path, detects and avoids conflicts in real time, and ensures the continuity and reliability of the approval process; dynamically adjusts subsequent nodes according to the execution situation, and flexibly responds to business changes; by introducing a multi-factor authentication mechanism, the authentication strength is automatically adjusted according to the risk level, and the file is encrypted and protected throughout its life cycle; the encryption key is strictly bound to the applicant's identity, ensuring that the approval process cannot be tampered with, and meeting the highest level of audit requirements.

[0047] In Embodiment 1, real-time multi-party collaborative annotation and discussion functions can be supported, decision-making efficiency is improved, historical similar cases are intelligently matched to provide accurate decision-making references for approval personnel, the decision-making process and basis are recorded completely, and a traceable approval knowledge base is formed; based on the multi-dimensional matching score calculation method, the matching score of each person with the authority is calculated, the matching accuracy of the approval personnel is effectively improved, the work load of each approval personnel is dynamically balanced, the uneven distribution of approval tasks is avoided, the high relevance of the approval personnel to the file content and department is ensured, and the approval quality is improved; through the full-process operation log record, multi-dimensional combination query and analysis are supported, key approval node information is stored on the chain, time proof is provided that cannot be tampered with, fine-grained permission change tracking is provided, and various compliance audit requirements are met.

[0048] Embodiment 2 As shown in the accompanying Figure 2 Embodiment 2 provides a file approval system, which includes a file preprocessing module, an approval path generation module, a remote approval module, an electronic visa module, and an archiving module.

[0049] The document preprocessing module is used to receive the documents submitted by the applicant for approval, extract the preset document attribute information of the documents for approval, encrypt the documents for approval, and obtain the encrypted documents for approval.

[0050] The approval path generation module is used to generate the optimal approval path based on the approval path dynamic programming model and the preset file attribute information of the documents to be approved; wherein, the approval path dynamic programming model is a pre-trained reinforcement learning model.

[0051] The remote approval module is used to distribute encrypted documents to approvers at each level of the optimal approval path, according to the optimal approval path, until all approvers at all levels of the optimal approval path have completed their approval.

[0052] The electronic visa module is used to electronically sign approved documents based on the risk level of the current business scenario and using a multi-factor authentication mechanism.

[0053] The archiving module is used to store the pre-determined key approval node information in a preset blockchain, classify and store the approved documents, and record the entire process operation log.

[0054] Example 3 As attached Figure 3 As shown, this embodiment 3 provides an electronic device, including: a memory for storing a computer program; a processor for executing the computer program to implement the steps of the document approval method; or, the processor executing the computer program to implement the functions of each module in the above-mentioned document approval system.

[0055] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a preset function, the instruction segments describing the execution process of the computer program in the electronic device.

[0056] The electronic device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above are examples of electronic devices and do not constitute a limitation on the electronic device. It may include more components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0057] The processor can be a central processing unit, and can also be other general-purpose processors, digital signal processors, application-specific integrated circuits, programmable logic devices, discrete gates or transistor logic components, discrete hardware components, or the like. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which is the control center of the electronic device and connects all parts of the electronic device through various interfaces and lines.

[0058] The memory can be used to store the computer programs and / or modules, and the processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory.

[0059] The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required for a function (such as a sound playing function, an image playing function, etc.), and the like; and the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), and the like. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, an intelligent memory card, a secure digital card, a flash memory card, at least one disk storage device, a flash memory device, or other volatile solid-state memory devices.

[0060] Embodiment 4 The embodiment 4 also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the file examination method.

[0061] The modules / units of the file examination system are stored in a computer readable storage medium if they are realized in the form of software function units and sold or used as independent products.

[0062] Based on such understanding, the present application realizes all or part of the processes of the above-mentioned file examination method, and can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer readable storage medium, and the computer program can realize the steps of the above-mentioned file examination method when executed by a processor. The computer program includes computer program codes, which can be in the form of source code, object code, executable files or preset intermediate forms, etc.

[0063] The computer readable storage medium can include any entity or apparatus, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal, software distribution medium and the like capable of carrying the computer program code.

[0064] Embodiment 5 Embodiment 5 provides a computer product, which includes a computer program stored in a computer readable storage medium; a processor of an electronic device reads the computer program from the computer readable storage medium, and the processor executes the computer program so that the electronic device can perform the file approval method described in embodiment 1, which will not be described here.

[0065] It should be noted that those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware, and the program can be stored in a computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments.

[0066] The above-mentioned embodiments are only one of the implementation manners of the technical solutions of the present application, and the scope of protection of the present application is not limited to the above-mentioned embodiments, but also includes any changes, substitutions and other implementation manners easily thought by those skilled in the art within the technical scope disclosed by the present application. It should be noted that the contents not described in detail in the specification of the present application are all the prior art known to those skilled in the art.

Claims

1. A method of file approval, characterized by, The method comprises the following steps: receiving a to-be-approved file submitted by an applicant, and extracting preset file attribute information of the to-be-approved file; encrypting the to-be-approved file to obtain an encrypted to-be-approved file; generating an optimal approval path based on an approval path dynamic programming model and in combination with the preset file attribute information of the to-be-approved file, wherein the approval path dynamic programming model is a pre-trained reinforcement learning model; circulating the encrypted to-be-approved file according to the optimal approval path, and distributing the to-be-approved file to an approval personnel of each approval level in the optimal approval path for approval until all the approval personnel of all the approval levels in the optimal approval path complete the approval.

2. The method of claim 1, wherein, The preset file attribute information of the to-be-approved file comprises a file type, a sensitive level and a content keyword feature of the to-be-approved file, wherein the sensitive level comprises public, confidential and top secret.

3. The method of claim 2, wherein, Based on the approval path dynamic programming model and in combination with the preset file attribute information of the to-be-approved file, the process of generating the optimal approval path is as follows: determining a required number of approval levels of the to-be-approved file according to the sensitive level of the to-be-approved file; planning an approval path by using the approval path dynamic programming model based on the file type of the to-be-approved file, the required number of approval levels of the to-be-approved file, the content keyword feature of the to-be-approved file, an enterprise organizational structure relationship and a pre-determined business rule, to obtain a candidate approval path; verifying the candidate approval path for logical consistency and compliance to obtain the optimal approval path; allocating a confirmation person meeting a permission requirement to each approval level in the optimal approval path; calculating a matching score of each confirmation person meeting the permission requirement, and selecting a preset number of confirmation persons with the highest matching scores as approval personnel of the corresponding approval levels.

4. The method of claim 3, wherein, The process of calculating the matching score of each confirmation person meeting the permission requirement is as follows: wherein, a matching score of the approver to the authority requirement; a matching degree of the authority level of the approver to the current approval level; a current task load of the approver; a relevance of the approver to the department to which the file belongs; , , are weight coefficients.

5. The method of claim 1, wherein, After circulating the encrypted to-be-approved file according to the optimal approval path, distributing the to-be-approved file to the approval personnel of each approval level in the optimal approval path for approval until all the approval personnel of all the approval levels in the optimal approval path complete the approval, the method further comprises the following steps: adopting a multi-factor authentication mechanism to perform electronic signature on the file after the approval according to a risk level of a current business scenario, wherein the multi-factor mechanism comprises biological feature authentication, dynamic password authentication, fingerprint authentication and video live body detection authentication.

6. The method of claim 1, wherein, Before encrypting the to-be-approved file to obtain the encrypted to-be-approved file, the method further comprises the following steps: verifying the compliance of a file format, a desensitization state of sensitive information and a file content of the to-be-approved file.

7. A document approval system characterized by, The method comprises the following steps: a file preprocessing module, configured to receive a to-be-approved file submitted by an applicant, and extract preset file attribute information of the to-be-approved file; encrypting the to-be-approved file to obtain an encrypted to-be-approved file; an approval path generation module, configured to generate an optimal approval path based on an approval path dynamic programming model and in combination with the preset file attribute information of the to-be-approved file, wherein the approval path dynamic programming model is a pre-trained reinforcement learning model; The remote examination and approval module is used for transferring the encrypted to-be-examined-and-approved file according to the optimal examination and approval path, distributing the file to the examination and approval personnel of each examination and approval level in the optimal examination and approval path for examination and approval, and until the examination and approval personnel of all the examination and approval levels in the optimal examination and approval path complete the examination and approval.

8. An electronic device, comprising: The computer program product comprises a computer program, and the computer program is executed by the processor to implement the file examination and approval method according to any one of claims 1-6. The computer program product comprises a computer program, and the computer program is executed by the processor to implement the file examination and approval method according to any one of claims 1-6. The computer program product comprises a computer program, and the computer program is executed by the processor to implement the file examination and approval method according to any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. The computer program product comprises a computer program, and the computer program is executed by the processor to implement the file examination and approval method according to any one of claims 1-6.

10. A computer program product, characterised in that, ​

Images