RPA and AI-based project scheme approval method, device, equipment and medium

Abstract

Embodiments of the present application disclose a project scheme approval method and device based on RPA and AI, equipment and medium. The method comprises: S1, identifying the declaration file of the project scheme to obtain the content to be approved, wherein the content to be approved comprises an expert information table and a content of a declarer commitment; S2, respectively matching each item of the content to be approved with the corresponding standard template; S3, if each item of the content to be approved is matched successfully, it is determined that the declaration file is approved. By adopting the above technical scheme, the problems of low efficiency and accuracy of artificial in the process of auditing the project scheme are solved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of process automation technology, and more specifically, to a project scheme approval method, apparatus, equipment and medium based on RPA and AI. Background Technology

[0002] RPA (Robotic Process Automation) uses specific "robot software" to simulate human operations on a computer and automatically execute process tasks according to rules.

[0003] AI (Artificial Intelligence) is a new technical science that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence.

[0004] RPA boasts unique advantages: low-code and non-intrusive. Low-code means RPA can be operated without requiring high IT skills; even business personnel without programming knowledge can develop processes. Non-intrusive means RPA can simulate human operations without requiring software systems to expose interfaces. However, traditional RPA has certain limitations: it can only be based on fixed rules, and its application scenarios are restricted. With the continuous development of AI technology, the deep integration of RPA and AI overcomes the limitations of traditional RPA. RPA + AI = Handwork + Headwork, significantly changing the value of labor.

[0005] Currently, the company's project approval process, such as applications for soil and water conservation projects, typically involves manual review. After submission, the application materials must be reviewed manually. Generally, application materials range from dozens to hundreds of pages, and the large number of applications leads to lengthy approval times; companies often need to wait around three days for a result. If problems arise during the approval process, the waiting time will be even longer. Furthermore, manual review is prone to irregularities or errors due to individual incompetence or negligence, resulting in low efficiency and accuracy. Summary of the Invention

[0006] This invention provides a project proposal approval method, apparatus, equipment, and medium to overcome the problems of low efficiency and accuracy in the process of manually reviewing project proposals.

[0007] In a first aspect, this invention provides a project proposal approval method based on RPA and AI, including:

[0008] S1. Identify the application documents for the project proposal to obtain the content to be approved, which includes the expert information form and the content of the applicant's commitment letter;

[0009] S2. Match each item pending approval with its corresponding standard template;

[0010] S3. If all pending approval items are successfully matched, the application document is deemed approved.

[0011] Optionally, after determining that the application documents have been approved, the method further includes:

[0012] S4. Fill in the approval information in the electronic approval template and submit the completed electronic approval template. The approval information indicates that the application document has been approved. The approval information includes the project name and the applicant's information.

[0013] Optionally, S2 specifically includes:

[0014] S21. Match the content of the applicant's commitment letter with the content of the standard commitment letter;

[0015] S22. If the applicant's seal is found in the content of the applicant's commitment letter, the applicant's seal is matched with the applicant's name.

[0016] S23. Extract the expert's identity information from the expert information and match the expert's identity information with a preset expert database.

[0017] Optionally, S3 specifically includes:

[0018] If the content of the applicant's commitment letter is consistent with the content of the standard commitment letter, and,

[0019] The content of the applicant's seal is consistent with the applicant's name, and,

[0020] If the expert's identity information exists in the preset expert database, then the application document is deemed approved.

[0021] Optionally, S21 specifically includes:

[0022] The Natural Language Processing (NLP) component is invoked to match the content of the applicant's commitment letter with the content of the standard commitment letter.

[0023] Optionally, S23 specifically includes:

[0024] The Optical Character Recognition (OCR) component is invoked to recognize the expert information, extract the expert's identity information from the expert information, and match the expert's identity information with a preset expert database.

[0025] Optionally, S1 specifically includes:

[0026] S11. Perform keyword identification on the project proposal application documents to obtain the key pages to be approved, wherein the key pages include the page containing the expert approval opinions and the page containing the applicant's commitment letter;

[0027] S12. Call the Optical Character Recognition (OCR) component to recognize the key page and obtain the content to be approved.

[0028] Secondly, embodiments of the present invention also provide a project solution approval device based on RPA and AI, comprising:

[0029] The pending approval content identification module is configured to: identify the application documents of the project plan and obtain the pending approval content, which includes the expert information form and the applicant's commitment letter.

[0030] The matching module is configured to match each item pending approval with its corresponding standard template.

[0031] The approval module is configured such that if all items pending approval match successfully, the application document is deemed approved.

[0032] Optionally, the device further includes:

[0033] The approval module is configured to: after confirming that the application document has been approved, fill in the approval information in the electronic approval template and submit the completed electronic approval template. The approval information is used to indicate that the application document has been approved, and the approval information includes the project name and the applicant's information.

[0034] Optionally, the matching module includes:

[0035] The commitment letter content matching unit is configured to match the content of the applicant's commitment letter with the content of the standard commitment letter;

[0036] The stamp content matching unit is configured to: if it is identified that the content of the applicant's commitment letter contains the applicant's stamp content, then match the applicant's stamp content with the applicant's name;

[0037] The expert information matching unit is configured to: extract the expert's identity information from the expert information and match the expert's identity information with a preset expert database.

[0038] Optionally, the approval module is specifically configured as follows:

[0039] If the content of the applicant's commitment letter is consistent with the content of the standard commitment letter, and,

[0040] The content of the applicant's seal is consistent with the applicant's name, and,

[0041] If the expert's identity information exists in the preset expert database, then the application document is deemed approved.

[0042] Optionally, the commitment letter content matching unit is specifically configured as follows:

[0043] The Natural Language Processing (NLP) component is invoked to match the content of the applicant's commitment letter with the content of the standard commitment letter.

[0044] Optionally, the expert information matching unit is specifically configured as follows:

[0045] The Optical Character Recognition (OCR) component is invoked to recognize the expert information, extract the expert's identity information from the expert information, and match the expert's identity information with a preset expert database.

[0046] Optionally, the module for identifying content to be approved is specifically configured as follows:

[0047] Keyword identification is performed on the project proposal application documents to obtain the key pages pending approval, including the page containing expert approval opinions and the page containing the applicant's commitment letter;

[0048] The Optical Character Recognition (OCR) component is invoked to recognize the key page and obtain the content to be approved.

[0049] Thirdly, embodiments of the present invention also provide a computing device, comprising:

[0050] Memory containing executable program code;

[0051] A processor coupled to the memory;

[0052] The processor calls the executable program code stored in the memory to execute the project scheme approval method based on RPA and AI provided in any embodiment of the present invention.

[0053] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the project approval method based on RPA and AI provided in any embodiment of the present invention.

[0054] The technical solution provided in this invention employs RPA and AI technologies. After identifying the applicant's application documents and obtaining the content to be approved, each item to be approved is matched with the corresponding standard template. If all items to be approved match successfully, the application document is deemed approved. Specifically, by using an RPA robot, the approval process for project proposals is automated, replacing manual review, thus improving efficiency. Furthermore, by employing AI technology, the system automatically compares whether each item to be approved meets the requirements, ensuring efficiency while avoiding errors or irregularities that can occur with manual review. This also effectively shortens the approval time for project proposals and improves the accuracy of the approval results.

[0055] The innovative aspects of this invention include:

[0056] 1. By using RPA robots to replace manual acceptance and approval of project proposals, the automatic approval of project proposals is realized, which improves approval efficiency and is one of the innovations of this invention.

[0057] 2. In the process of RPA robot automatically approving project proposals, the content of the applicant's commitment letter is matched with the content of the standard commitment letter by calling the natural language processing (NLP) component, and the optical character recognition (OCR) component is called to automatically recognize the applicant's stamped content and expert information, which improves the accuracy and efficiency of recognition. This is one of the innovations of this invention. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1a This is a flowchart of a project solution approval method based on RPA and AI provided in Embodiment 1 of the present invention;

[0060] Figure 1b This is a screenshot of the workflow effect of the project plan approval method created on a workflow automation platform according to Embodiment 1 of the present invention.

[0061] Figure 1c This is a screenshot showing the effect of querying application documents on the approval interface, as provided in Embodiment 1 of the present invention.

[0062] Figure 2 This is a structural block diagram of a project solution approval device based on RPA and AI provided in Embodiment 2 of the present invention;

[0063] Figure 3 This is a schematic diagram of the structure of a computing device provided in Embodiment 3 of the present invention. Detailed Implementation

[0064] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] It should be noted that the terms "comprising" and "having," and any variations thereof, in the embodiments and drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0066] In the description of this embodiment of the invention, the "electronic approval template" is a template file used to fill in approval information after the application documents have been approved. After the completed electronic approval template is returned to the applicant, the applicant can learn about the approval status of the application documents from the electronic approval template.

[0067] In the description of this embodiment of the invention, the "approval information" filled in the "electronic approval template" includes the project name, applicant information, and approval result.

[0068] In the description of the embodiments of the present invention, the "preset expert database" refers to a database containing a large number of project review experts. The database contains the identity information of each review expert, such as name, age, gender, and qualifications.

[0069] In the description of the embodiments of this invention, different items pending approval have corresponding standard templates. For example, the standard template for the applicant's commitment letter is the standard commitment letter for project applications, and the standard template for expert information is the preset expert database. These standard templates can serve as standard documents, and during the approval process of the application documents, each item pending approval is reviewed according to the corresponding standard document. To more clearly explain the contents of the various embodiments of this invention, the basic working principle of the embodiments of this invention will be briefly introduced below.

[0070] Robotic Process Automation (RPA) is a process that uses specific "robot software" to simulate human operations on a computer and automatically execute process tasks according to rules.

[0071] AI (Artificial Intelligence) is a new technical science that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence. The technical solution of this invention primarily employs OCR (Optical Character Recognition) and NLP (Natural Language Processing) technologies from the AI ​​field. OCR technology is used to recognize the application documents uploaded by the project applicant to obtain the content to be approved within the application documents. NLP technology is used to match the applicant's commitment letter content with the standard commitment letter content to determine whether the applicant's commitment letter content meets the requirements of the standard commitment letter.

[0072] Currently, project approval relies on manual processes. Approval systems only provide document transfer operations and cannot approve the submitted content. The most significant drawback of manual approval is the high demand for qualified personnel with varying levels of expertise, coupled with high staff turnover and low return on investment. Relying on manual approval also carries the risk of irregular approvals or errors due to individual incompetence or negligence. This invention combines RPA technology with OCR and NLP technologies, and based on approval rules, enables automated approval and authorization of project proposals. The entire process requires no manual intervention, effectively improving both the accuracy and efficiency of the approval process.

[0073] The following is a detailed description of the project scheme approval method, apparatus, equipment and medium based on RPA and AI disclosed in the embodiments of the present invention.

[0074] Example 1

[0075] Figure 1a This is a flowchart of a project proposal approval method based on RPA and AI provided in Embodiment 1 of the present invention. This method can be applied to the approval process of different types of projects and can be executed by an RPA and AI-based project proposal approval device. This device can be implemented through software and / or hardware. In this embodiment, an RPA robot is used to achieve automatic approval of project proposals. The RPA robot can be mounted on a process automation platform.

[0076] Figure 1b This is a screenshot of the workflow effect of the project plan approval method created on the workflow automation platform, as provided in Embodiment 1 of the present invention. Figure 1c This is a screenshot showing the effect of querying application documents on the approval interface, as provided in Embodiment 1 of the present invention. Figure 1bAs shown, after logging into the approval platform, if the RPA robot finds a project proposal application document in the approval interface, it will trigger a process such as... Figure 1c The download instructions shown will download each application document sequentially. For each downloaded application document, the RPA robot will first conduct a preliminary review, which typically checks the completeness of the document. After the preliminary review is complete, switching accounts will switch the current preliminary review interface to the acceptance interface, where the application document will be received and automatically approved. Figure 1a As shown, the project solution approval method based on RPA and AI provided in this embodiment includes:

[0077] S110. Identify the applicant's application documents to obtain the content to be approved.

[0078] The application documents are generally in PDF (Portable Document Format) format. When recognizing the application documents, an OCR component can be used to identify the content of each page. Specifically, keyword recognition can be performed on the applicant's application documents to obtain the key pages awaiting approval, such as... Figure 1b The image shows the target pages for approval. The key pages to be approved include the pages containing expert review comments and the applicant's commitment letter. After obtaining the key pages, the content of each key page can be recognized by calling the OCR component to obtain the content to be approved. This content includes the expert information table and the applicant's commitment letter.

[0079] S120. Match each item pending approval with its corresponding standard template.

[0080] In this embodiment, each item pending approval is matched with its corresponding standard template, including matching the applicant's commitment letter with the standard commitment letter. Furthermore, if the applicant's commitment letter contains an applicant's seal, the seal is matched with the applicant's identity. The matching process also includes extracting expert identity information from the expert opinion information table and matching this information with a pre-defined expert database.

[0081] In this embodiment, the application document can only be considered approved after all items pending approval match the corresponding standard template. If any item pending approval fails to match the corresponding standard template, the application document is deemed to have failed the review.

[0082] S130. If all pending approval items are successfully matched, the application document is deemed approved.

[0083] The standard commitment letter includes regulations and rules governing project application and implementation. The RPA robot matches the applicant's commitment letter with the standard commitment letter; if the two match, the applicant's commitment letter is deemed approved.

[0084] For example, when matching the content of the applicant's commitment letter with the content of the standard commitment letter, each page of the applicant's commitment letter can be compared page by page with the corresponding page of the standard commitment letter. However, this comparison method does not take into account the relationship between the content of each page, which can easily lead to inaccurate comparison results.

[0085] In this embodiment, to avoid inaccurate text comparison results, the content of the applicant's commitment letter is first concatenated into a single page of continuous text, i.e., the text to be matched, in the original text order. Similarly, the content of the standard commitment letter is concatenated into a single page of continuous text, i.e., the standard text, in the original text order. Then, a difference comparison can be performed between the text to be matched and the standard text. This involves comparing the text content of the applicant's commitment letter and the text content of the standard commitment letter as separate wholes. This setup avoids inaccurate comparison results due to neglecting the interrelationships between pages, which is one of the innovative aspects of this invention.

[0086] For example, in the process of matching the text to be matched with the standard text, the matching can be performed by judging whether the characters or words used in the text content are the same. If the characters in the text to be matched and the standard text are the same, it can be determined that the content of the applicant's commitment letter has been approved.

[0087] Preferably, to improve the accuracy of the matching results between the text to be matched and the standard text, semantic analysis of the text to be matched and the standard text can be performed by calling NLP components. If the semantic similarity between the text to be matched and the standard text meets the set requirements, the content of the applicant's commitment letter can be determined to have passed the review.

[0088] Furthermore, if a match fails, the differences between the text to be matched and the standard text can be marked in the text to be matched. For example, content that does not meet the set semantic similarity requirements can be marked. These differences can be marked by bolding the font, changing the font color, adding a background color, highlighting the font, increasing the font size, changing to italics, or adding an underline. This setting allows applicants to understand the reasons for the failure and thus make targeted modifications to the application documents.

[0089] Specifically, regarding the content of the applicant's commitment letter, if the RPA robot identifies any stamped content from the applicant, it matches the stamped content with the applicant's entity, which is the company name of the project being applied for. If the stamped content matches the applicant's entity, the stamped content is deemed approved. In this embodiment, the stamped content can be recognized by calling an OCR component, leveraging OCR's signature recognition capabilities.

[0090] For example, the identification of the applicant's stamped content can also be achieved in the following ways:

[0091] In this embodiment, the RPA robot can be mounted on the UiBot Creator platform, a professional and powerful robot production tool that primarily provides a good platform for robot creation. The AI ​​platform with seal recognition functionality is the UiBot Mage platform, a tool-type product mainly providing AI capability support for RPA robot developers. This AI platform integrates pre-configured seal recognition templates. Using these templates, operations such as seal text recognition, seal color recognition, seal shape recognition, and seal position recognition can be performed. This embodiment primarily uses seal text recognition to obtain the applicant's name from the seal text.

[0092] Specifically, in this embodiment, a single target account can be used to simultaneously log in to both the RPA platform and the AI ​​platform, enabling the RPA robot to integrate with the AI ​​platform. After logging into both the RPA robot's platform and the AI ​​platform using this target account, a communication connection is established between the RPA robot's platform and the AI ​​platform. This means the RPA robot can directly call the seal recognition function already released by the AI ​​platform to recognize the content stamped by the applicant, which is one of the innovative aspects of this embodiment.

[0093] In this embodiment, the "preset expert database" refers to a database containing a large number of project review experts. This database includes the identity information of each review expert, such as name, age, gender, and qualifications. Specifically, the process of matching the expert's identity information with the preset expert database involves determining whether the review expert is an expert in the preset expert database. If the expert is an expert in the preset expert database, the expert information is deemed approved. The recognition of the expert information table can be achieved by calling an OCR component. Through the OCR's table recognition capabilities, the table content can be obtained, and the expert's identity information can be extracted from the expert information table.

[0094] In this embodiment, for each of the aforementioned items to be matched for approval, the application document can only be deemed approved if the content of the applicant's commitment letter matches the content of the corresponding standard commitment letter, the applicant's stamp matches the applicant's identity, and expert identity information exists in the preset expert database. If any one of these matching conditions fails, the application document is deemed to have failed to be approved.

[0095] Furthermore, after confirming the approval of the application documents, the RPA robot can return approval information to the applicant, indicating that the application documents have passed the review. This approval information includes the project name, applicant information, and approval status. Specifically, this approval information can be filled in an electronic approval template. After completing the electronic approval information according to the template, the completed template can be submitted to inform the applicant of the project proposal's approval status. In addition, the RPA robot can obtain the project proposal's license document by triggering the certificate issuance and licensing instruction on the approval interface, and can return this license document to the applicant.

[0096] Furthermore, after submitting the completed electronic approval template, the RPA robot can also fill in and submit the public notice of the project plan approval, thus completing the entire project plan approval process.

[0097] The technical solution provided in this embodiment employs RPA and AI technologies. After identifying the applicant's application documents and obtaining the content to be approved, each item to be approved is matched with the corresponding standard template. If all items to be approved match successfully, the application document is deemed approved. Specifically, by using an RPA robot, the approval process for project proposals is automated, replacing manual review, thus improving efficiency. Furthermore, by employing AI technology, the system automatically compares whether each item to be approved meets the requirements, ensuring efficiency while avoiding errors or irregularities that can occur with manual review. This also effectively shortens the approval time for project proposals and improves the accuracy of the approval results.

[0098] Example 2

[0099] Figure 2 This is a structural block diagram of a project solution approval device based on RPA and AI provided in Embodiment 2 of the present invention. The device includes: a content recognition module 210 to be approved, a matching module 220, and an approval module 230; wherein,

[0100] The pending approval content identification module 210 is configured to: identify the application documents of the project plan to obtain the pending approval content, which includes the expert information table and the applicant's commitment letter.

[0101] The matching module 220 is configured to match each item pending approval with its corresponding standard template.

[0102] The approval module 230 is configured to determine that the application document is approved if all items pending approval are successfully matched.

[0103] Optionally, the device further includes:

[0104] The approval module is configured to: after confirming that the application document has been approved, fill in the approval information in the electronic approval template and submit the completed electronic approval template. The approval information is used to indicate that the application document has been approved, and the approval information includes the project name and the applicant's information.

[0105] Optionally, the matching module includes:

[0106] The commitment letter content matching unit is configured to match the content of the applicant's commitment letter with the content of the standard commitment letter;

[0107] The stamp content matching unit is configured to: if it is identified that the content of the applicant's commitment letter contains the applicant's stamp content, then match the applicant's stamp content with the applicant's name;

[0108] The expert information matching unit is configured to: extract the expert's identity information from the expert information and match the expert's identity information with a preset expert database.

[0109] Optionally, the approval module is specifically configured as follows:

[0110] If the content of the applicant's commitment letter is consistent with the content of the standard commitment letter, and,

[0111] The content of the applicant's seal is consistent with the applicant's name, and,

[0112] If the expert's identity information exists in the preset expert database, then the application document is deemed approved.

[0113] Optionally, the commitment letter content matching unit is specifically configured as follows:

[0114] The Natural Language Processing (NLP) component is invoked to match the content of the applicant's commitment letter with the content of the standard commitment letter.

[0115] Optionally, the expert information matching unit is specifically configured as follows:

[0116] The Optical Character Recognition (OCR) component is invoked to recognize the expert information, extract the expert's identity information from the expert information, and match the expert's identity information with a preset expert database.

[0117] Optionally, the module for identifying content to be approved is specifically configured as follows:

[0118] Keyword identification is performed on the project proposal application documents to obtain the key pages pending approval, including the page containing expert approval opinions and the page containing the applicant's commitment letter;

[0119] The Optical Character Recognition (OCR) component is invoked to recognize the key page and obtain the content to be approved.

[0120] The RPA and AI-based project approval device provided in this embodiment of the invention can execute the RPA and AI-based project approval method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method. Technical details not described in detail in the above embodiments can be found in the RPA and AI-based project approval method provided in any embodiment of the invention.

[0121] Example 3

[0122] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a computing device provided in Embodiment 3 of the present invention. Figure 3 As shown, the computing device may include:

[0123] Memory 701 storing executable program code;

[0124] Processor 702 coupled to memory 701;

[0125] The processor 702 calls the executable program code stored in the memory 701 to execute the project scheme approval method based on RPA and AI provided in any embodiment of the present invention.

[0126] This invention discloses a computer-readable storage medium storing a computer program that enables a computer to execute the RPA and AI-based project approval method provided in any embodiment of this invention.

[0127] In various embodiments of the present invention, it should be understood that the sequence number of each process does not necessarily imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0128] In the embodiments provided by this invention, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information.

[0129] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0130] If the integrated units described above are implemented as software functional units and sold or used as independent products, they can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several requests to cause a computer device (which can be a personal computer, server, or network device, specifically a processor in the computer device) to execute some or all of the steps of the methods described in the various embodiments of the present invention.

[0131] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0132] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0133] Those skilled in the art will understand that the modules in the apparatus of the embodiments can be distributed in the apparatus of the embodiments as described in the embodiments, or they can be located in one or more devices different from this embodiment with corresponding changes. The modules of the above embodiments can be combined into one module, or they can be further divided into multiple sub-modules.

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A project scheme approval method based on RPA and AI, characterized in that, The method comprises the following steps: S1, after the RPA robot logs in the approval platform, if it is found in the approval interface that there is a project plan declaration file, the RPA robot triggers the download instruction of the approval interface to download the declaration files one by one, and for each downloaded declaration file, the RPA robot first performs a preliminary review on the declaration file, the preliminary review process is to check the integrity of the declaration file, after the preliminary review is completed, the current preliminary review interface is switched to the acceptance interface by switching the account, and the declaration file is received in the acceptance interface and automatic approval operation is started; The declaration file of the project plan is identified to obtain the content to be approved, and the content to be approved includes an expert information table and a declaration person commitment content; S2, each item of the content to be approved is matched with a corresponding standard template; S3, if each item of the content to be approved is matched successfully, it is determined that the declaration file is approved; Wherein, the S3 specifically comprises: If the declaration person commitment content is consistent with the standard commitment content, and, The declaration person seal content is consistent with the declaration person name, and, The identity information of the expert extracted from the expert information table exists in the preset expert library, it is determined that the declaration file is approved; Wherein, when the declaration person commitment content is matched with the standard commitment content, the declaration person commitment content is spliced into a page of context-continuous text according to the original order of the text to obtain a matching text, and the standard commitment content is spliced into a page of context-continuous text according to the original order of the text to obtain a standard text, and then the matching text is compared with the standard text; The declaration person seal content is identified by the following method: Through a target account logged in the RPA platform and the AI platform at the same time, the platform carried by the RPA robot is connected with the AI platform to establish a communication connection, after the communication connection is established, the RPA robot calls the seal recognition function published by the AI platform to identify the declaration person seal content, wherein the AI platform integrates a pre-configured seal recognition template for performing seal text recognition, seal color recognition, seal shape recognition and seal position recognition.

2. The method of claim 1, wherein, After it is determined that the declaration file is approved, the method further comprises: S4, filling the reply information in the electronic reply template, and submitting the filled electronic reply template, wherein the reply information indicates that the declaration file is approved, and the reply information includes the project name and the declaration person information.

3. The method of claim 1, wherein, S2 specifically comprises: S21, matching the declaration person commitment content with the standard commitment content; S22, if the declaration person seal content is identified in the declaration person commitment content, the declaration person seal content is matched with the declaration person name; S23, extracting the identity information of the expert from the expert information table, and matching the identity information of the expert with the preset expert library.

4. The method of claim 3, wherein, The S21 specifically comprises: Calling a natural language processing (NLP) component to match the declaration person commitment content with the standard commitment content.

5. The method of claim 3, wherein, The S23 specifically comprises: The optical character recognition (OCR) component is called to recognize the expert information, the identity information of the expert is extracted from the expert information table, and the identity information of the expert is matched with a preset expert library.

6. The method of claim 1, wherein, The S1 specifically comprises: S11, keyword recognition is performed on the project scheme declaration file to obtain a key page to be approved, wherein the key page comprises a page where expert approval opinions are located and a page where a declarer commitment is located; S12, an optical character recognition (OCR) component is called to recognize the key page to obtain the content to be approved. 7.A project plan approval device based on RPA and AI, characterized by Comprise: The content to be approved identification module is configured to: after an RPA robot logs in an approval platform, if it is found in an approval interface that there is a project scheme declaration file, a download instruction of the approval interface is triggered to download each declaration file in turn, for each downloaded declaration file, the RPA robot first performs pre-examination on the declaration file, the pre-examination process is to check the integrity of the declaration file, after the pre-examination is completed, the current pre-examination interface is switched to a receiving interface by switching accounts, the declaration file is received in the receiving interface and automatic approval operation is started; the declaration file of the project scheme is recognized to obtain the content to be approved, the content to be approved comprises an expert information table and a declarer commitment content; The matching module is configured to: each item of the content to be approved is matched with a corresponding standard template; The approval module is configured to: if each item of the content to be approved is matched successfully, it is determined that the declaration file is approved; The approval module is configured to: If the declarer commitment content is consistent with the standard commitment content, and The declarer seal content is consistent with the declarer name, and The identity information of the expert extracted from the expert information table exists in the preset expert library, it is determined that the declaration file is approved; When the declarer commitment content is matched with the standard commitment content, the declarer commitment content is spliced into a page of context-continuous text in the original order of the text to obtain a matching text, and the standard commitment content is spliced into a page of context-continuous text in the original order of the text to obtain a standard text, and then the matching text is compared with the standard text in difference; The declarer seal content is recognized by the following method: A target account that logs in an RPA platform and an AI platform at the same time is used to enable the platform carried by the RPA robot to establish a communication connection with the AI platform, after the communication connection is established, the RPA robot recognizes the declarer seal content by calling the seal recognition function published by the AI platform, wherein the AI platform integrates a pre-configured seal recognition template, and the seal recognition template is used to perform seal text recognition, seal color recognition, seal shape recognition and seal position recognition.

8. The apparatus of claim 7, wherein, The device further comprises: The reply module is configured to: after determining that the declaration file is approved, fill in the reply information in an electronic reply template, and submit the filled electronic reply template, wherein the reply information is used to indicate that the declaration file is approved, and the reply information includes a project name and declarer information.

9. The apparatus of claim 7, wherein, The matching module comprises: The commitment content matching unit is configured to match the declarer commitment content with standard commitment content; The seal content matching unit is configured to match the declarer seal content with the declarer name if the declarer seal content is identified in the declarer commitment content; The expert information matching unit is configured to extract the identity information of the expert from the expert information, and match the identity information of the expert with a preset expert library.

10. An apparatus, comprising: The device comprises: One or more processors; A storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the RPA and AI-based project scheme approval method according to any one of claims 1-6.

11. A computer readable storage medium having stored thereon a computer program, characterized in that The program is executed by the processor to implement the RPA and AI-based project scheme approval method according to any one of claims 1-6.

Images