AI-based travel expense report filling and auditing method, system, device and medium

Abstract

The application relates to an AI-based travel expense reimbursement filing and auditing method, system, device and medium, which is applied to the technical field of financial management, and the method comprises the following steps: obtaining user uploaded information; extracting ticket information in the user uploaded information based on a preset visual model and a document extraction tool; pre-auditing the ticket information based on an AI model; if the pre-auditing is passed, extracting structured data of a reimbursement form based on the ticket information; transmitting the structured data of the reimbursement form to an OA interface to create a travel expense reimbursement application page, uploading the tickets in the user uploaded information to the travel expense reimbursement application page through the OA interface, and showing a link of the travel expense reimbursement application page; if a user clicks a confirmation button of the travel expense reimbursement application page, a reimbursement form is generated, and the reimbursement form is submitted to a reimbursement auditing end. The application has the effects of improving the efficiency and accuracy of travel expense reimbursement.

Description

Technical Field

[0001] This application relates to the technical field of financial management, and in particular to an AI-based method, system, device, and medium for filling out and reviewing travel expense reports. Background Technology

[0002] Travel expense reimbursement is an important aspect of daily business operations, involving the reasonable expenditure and reimbursement of employee expenses, as well as the standardized management of corporate finances. An efficient and accurate travel expense reimbursement process can improve employee productivity, reduce the workload of the finance department, and help companies better control costs and mitigate risks. As companies expand and their businesses diversify, travel activities become increasingly frequent, leading to a significant increase in the volume of travel expense reimbursements, making the need for process optimization ever more urgent.

[0003] Traditionally, travel expense reimbursement processes are conducted manually. Employees need to manually organize all receipts and documents from their business trips and fill out reimbursement forms. Finance staff then review each submitted form. This method is not only time-consuming and labor-intensive but also prone to human error, making it difficult to guarantee the accuracy of reimbursements. Summary of the Invention

[0004] To improve the efficiency and accuracy of travel expense reimbursement, this application provides an AI-based method, system, device, and medium for travel expense reimbursement filling and review.

[0005] Firstly, this application provides an AI-based method for filling out and reviewing travel expense reports, employing the following technical solution: An AI-based method for travel expense reimbursement filling and approval, applied to the reimbursement filling terminal, includes: Obtain user-uploaded information; Based on a preset visual model and document extraction tools, the ticket information in the user-uploaded information is extracted; The invoice information is pre-screened based on an AI model; If the preliminary review is approved, the structured data of the expense report will be extracted based on the invoice information. The structured data of the expense report is transmitted to the OA interface to create a travel expense reimbursement application page. The receipts in the user-uploaded information are uploaded to the travel expense reimbursement application page through the OA interface, and a link to the travel expense reimbursement application page is displayed. If the user clicks the confirmation button on the travel expense reimbursement application page, an expense report will be generated and submitted to the expense review terminal.

[0006] By adopting the above technical solutions, the preset visual model and document extraction tool can automatically extract invoice information from user-uploaded information. The AI ​​model extracts structured data from the expense report based on the invoice information and fills out the travel expense reimbursement application through the OA interface, and uploads the invoices, simplifying the user's travel expense reimbursement process and thus improving the efficiency of the user's travel expense reimbursement. After the travel expense reimbursement application is generated, user confirmation is required before submission, which improves the accuracy of travel expense reimbursement. By pre-screening the invoice information, the AI ​​model can initially filter out invoices that do not meet the reimbursement requirements, improving the efficiency of subsequent review by financial personnel.

[0007] Optionally, the pre-screening of the invoice information based on the AI ​​model includes: The AI ​​model is used to analyze the ticket information and establish a business trip timeline. The OA interface is invoked to query whether a business trip application form exists based on the start time of the business trip timeline. If the business trip application form does not exist, the OA interface is called to query whether a business trip application form exists based on the location of the ticket in the ticket information; If the aforementioned business trip application form exists, then the preliminary review of the invoice information is approved.

[0008] By adopting the above technical solution, an AI model is used to analyze invoice information to establish a business trip timeline. Based on the start time of the business trip timeline and the location of the invoice in the invoice information, the OA interface is called to obtain the business trip application form. Only business trip application forms with corresponding information can pass the pre-approval, thereby effectively pre-approving the invoice information and improving the accuracy and compliance of travel expense reimbursement.

[0009] Optionally, the invoice information includes the invoice type, and the extraction of structured reimbursement data based on the invoice information includes: Based on the invoice type and the corresponding preset JSON field format, the initial expense report structured data is extracted from the invoice information; The structured data of the initial expense report is supplemented based on the business trip application form to obtain the structured data of the expense report.

[0010] By adopting the above technical solution, the structured data of the initial expense reimbursement form is extracted from the invoice information, which facilitates the subsequent creation of travel expense reimbursement application pages and the supplementation of data through travel application forms, thereby improving the completeness of the structured data of the expense reimbursement form.

[0011] Secondly, this application provides an AI-based method for filling out and reviewing travel expense reports, employing the following technical solution: An AI-based method for travel expense reimbursement filling and approval, applied to the reimbursement approval end, includes: Receive expense reports, which include invoice types; The reimbursement form is reviewed for compliance using an AI model to determine the compliance review result. If the compliance review result is "approved", then the deductible input tax is calculated based on the invoice type.

[0012] By adopting the above technical solution, AI models are used to assist financial personnel in reviewing expense reports, and deductible input tax is automatically calculated after the review is approved, which improves the efficiency and accuracy of financial approval.

[0013] Optionally, the step of using an AI model to conduct a compliance review of the expense report and determine the compliance review result includes: Call the OA interface to obtain the business trip application period associated with the expense report; Based on the business trip application period, the dates of the receipts in the expense report are reviewed to determine the date review results; Based on the type of invoice in the expense report, perform a completeness review on the expense report and determine the completeness review result; A business trip timeline is established based on the receipts in the expense report, wherein the business trip timeline is the timeline corresponding to the business trip time and the business trip location. The timeline of the business trip is reviewed for consistency in terms of time and space logic to determine the results of the timeline review. The reimbursement form is reviewed based on RAG technology to determine the amount review result. The compliance review result is determined based on the date review result, the integrity review result, the itinerary review result, and the amount review result.

[0014] By adopting the above technical solutions, the AI ​​model improves the accuracy of the review by examining multiple dimensions, including the reimbursement date, completeness, consistency of the trip's spatiotemporal logic, and the reimbursement amount.

[0015] Optionally, the step of reviewing the reimbursement amount on the reimbursement form based on RAG technology and determining the amount review result includes: Convert table segments in the preset standard into text format; A knowledge base for travel reimbursement systems is constructed based on the preset standards in text format. Based on the AI ​​model, the business trip location, business trip role, and business trip date are extracted from the expense report; The knowledge base of the travel reimbursement system is searched based on the business trip location to determine the hit segment; The expense report is reviewed based on the traveler's role, travel date, and hit segment to obtain the expense review result.

[0016] By adopting the above technical solution, a travel reimbursement system knowledge base is constructed using RAG technology, and the reimbursement amount is reviewed based on the travel reimbursement system knowledge base, thereby improving the accuracy of the amount review.

[0017] Thirdly, this application provides an AI-based travel expense reimbursement filling and review system, which adopts the following technical solution: An AI-based travel expense reimbursement system for submission and approval, applied to the reimbursement submission end, includes: The first acquisition module is used to acquire user-uploaded information; The information extraction module is used to extract ticket information from the user-uploaded information based on a preset visual model and document extraction tools; The information pre-screening module is used to pre-screen the invoice information based on an AI model; The reimbursement data structure extraction module is used to extract structured data of the reimbursement form based on the invoice information if the preliminary review is passed. The expense report filling module is used to transmit the structured data of the expense report to the OA interface to create a travel expense reimbursement application page, upload the receipts in the user-uploaded information to the travel expense reimbursement application page through the OA interface, and display a link to the travel expense reimbursement application page; The expense report submission module is used to generate an expense report and submit it to the expense review terminal if the user clicks the confirmation button on the travel expense report application page.

[0018] By adopting the above technical solutions, the preset visual model and document extraction tool can automatically extract invoice information from user-uploaded information. The AI ​​model extracts structured data from the expense report based on the invoice information and fills out the travel expense reimbursement application through the OA interface, and uploads the invoices, simplifying the user's travel expense reimbursement process and thus improving the efficiency of user travel expense reimbursement. After generating the travel expense reimbursement application, user confirmation is required before submission, which improves the accuracy of travel expense reimbursement. By pre-screening the invoice information, the AI ​​model can initially filter out invoices that do not meet the reimbursement requirements, improving the efficiency of subsequent review by financial personnel.

[0019] Fourthly, this application provides an AI-based travel expense reimbursement filling and review system, employing the following technical solution: An AI-based travel expense reimbursement form and approval system, applied to the reimbursement approval end, includes: The expense report receiving module is used to receive expense reports, which include invoice types. The expense report review module is used to conduct a compliance review of the expense report using an AI model and determine the compliance review result. The deductible input tax calculation module is used to calculate the deductible input tax based on the invoice type if the compliance review result is that the review is passed.

[0020] By adopting the above technical solution, AI models are used to assist financial personnel in reviewing expense reports, and deductible input tax is automatically calculated after the review is approved, which improves the efficiency and accuracy of financial approval.

[0021] Fifthly, this application provides an electronic device that adopts the following technical solution: An electronic device includes a processor coupled to a memory; The memory stores a computer program that can be loaded by a processor and executed by the AI-based travel expense reimbursement filling and review method described in any of the first aspects.

[0022] Sixthly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the AI-based travel expense reimbursement filling and review method described in any one of the first aspects. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating an AI-based travel expense reimbursement filling and review method applied to the reimbursement filling terminal, as provided in an embodiment of this application.

[0024] Figure 2 This is a flowchart illustrating an AI-based travel expense reimbursement filling and review method applied to the reimbursement review end, as provided in an embodiment of this application.

[0025] Figure 3 This is a structural block diagram of an AI-based travel expense reimbursement filling and review system provided in this application embodiment.

[0026] Figure 4 This is a structural block diagram of an AI-based travel expense reimbursement filling and review system provided in an embodiment of this application.

[0027] Figure 5 This is a structural block diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0028] The present application will be further described in detail below with reference to the accompanying drawings.

[0029] This application provides an AI-based method for filling out and reviewing travel expense reports. This AI-based method can be executed by an electronic device, which can be a server or a terminal device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, desktop computer, etc., but is not limited to these.

[0030] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0031] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0032] This application provides an AI-based method for filling out and reviewing travel expense reports. This method uses a report filling end and a report review end to jointly complete the travel expense reimbursement process. The report filling end is used to fill out the reimbursement application and generate the reimbursement form, while the report review end is used to assist financial personnel in reviewing the reimbursement form and automatically calculate the deductible input tax. No manual calculation by financial personnel is required. Both ends can be assisted by AI models (such as the DeepSeekV3 model), thereby improving the efficiency and accuracy of travel expense reimbursement.

[0033] like Figure 1 As shown, an AI-based method for travel expense reimbursement filling and approval is applied to the reimbursement filling terminal. The main process of this method is described as follows (steps S101 to S106): Step S101: Obtain user-uploaded information.

[0034] When users need to submit travel expense reports, they can upload relevant materials (i.e., user-uploaded information) through the user conversation interface provided by the AI ​​model. These materials include, but are not limited to, receipt images and electronic invoices (PDF, Word, etc.). The user conversation interface can be pre-set with prompts (e.g., "Hello, I am the travel expense pre-filling assistant. Please upload your expense receipts") to guide users in uploading materials.

[0035] Step S102: Extract ticket information from user-uploaded information based on a preset visual model and document extraction tools.

[0036] If the user-uploaded information includes images, the image content is extracted using a preset visual model (e.g., Tencent Hunyuan Model); if the user-uploaded information includes documents, the document content is extracted using a document extraction tool (e.g., MinerU); the extracted image content and document content are combined to form the ticket information, which includes ticket type (e.g., invoice, itinerary, boarding pass, etc.), invoice number, amount, tax rate, tax amount, date, ticket location, etc.

[0037] Step S103: Pre-examine the invoice information based on the AI ​​model.

[0038] Specifically, the preliminary review of invoice information based on the AI ​​model includes: analyzing the invoice information based on the AI ​​model to establish a business trip timeline; calling the OA interface to query whether a business trip application form exists based on the start time of the business trip timeline; if no business trip application form exists, calling the OA interface to query whether a business trip application form exists based on the location of the invoice in the invoice information; if a business trip application form exists, the preliminary review of the invoice information is approved.

[0039] In this embodiment, the AI ​​model establishes a business trip timeline based on the date and location information of each ticket in the ticket information. The business trip timeline displays the user's travel trajectory in chronological order. The OA system stores all users' business trip application forms. By calling the OA interface based on the start time of the business trip timeline, it checks whether a business trip application form for the corresponding time exists. If a business trip application form exists, the ticket information pre-approves. If no business trip application form exists, it calls the OA interface again based on the location of the ticket in the ticket information to check whether a business trip application form for the corresponding location exists. If a business trip application form exists, the ticket information pre-approves. If no business trip application form exists, the ticket information pre-approves. The user is then prompted to check whether the uploaded ticket is correct.

[0040] By pre-screening reimbursement materials through business trip application forms, the occurrence of errors in user material uploads or reimbursements made without actually traveling can be reduced, thereby reducing the workload of approvers and improving the efficiency of reimbursement approval.

[0041] Step S104: If the preliminary review is approved, extract the structured data of the reimbursement form based on the invoice information.

[0042] Specifically, the process of extracting structured data from expense reports based on invoice information includes: extracting initial structured data from expense reports based on invoice type and corresponding preset JSON field format; and supplementing the initial structured data with data from the business trip application form to obtain the actual structured data of the expense reports.

[0043] In this embodiment, different invoice types correspond to different JSON field formats. The AI ​​model converts the invoice information into initial expense report structured data in JSON format according to the preset JSON field format and invoice type. The model then supplements the initial expense report structured data with the business trip application form obtained during the pre-approval process (e.g., project ID, project name, etc.) to obtain the expense report structured data.

[0044] Step S105: Transmit the structured data of the expense report to the OA interface to create a travel expense reimbursement application page. Upload the receipts in the user-uploaded information to the travel expense reimbursement application page through the OA interface and display the link to the travel expense reimbursement application page.

[0045] After generating structured expense report data, the AI ​​model can automatically upload the data to the OA system via an OA interface, eliminating the need for user uploads. This creates the OA system's travel expense reimbursement application page. Once the OA system creates the application page, the AI ​​model uploads the receipts from the user-uploaded information to the page through the OA interface and displays a link to the corresponding application page. Users can click the link to view and modify the content of the application page, thus completing the expense reimbursement application. This significantly reduces the difficulty for users who are unsure how to complete the reimbursement process and improves the efficiency of expense reimbursement submission.

[0046] Step S106: If the user clicks the confirmation button on the travel expense reimbursement application page, an expense report will be generated and submitted to the expense review terminal.

[0047] Once the user confirms that the pre-filled travel expense reimbursement application page content is correct, they can click the confirmation button on the travel expense reimbursement application page. At this time, a reimbursement form will be generated based on the content of the travel expense reimbursement application page and submitted to the reimbursement review end, thereby completing the reimbursement filling process.

[0048] like Figure 2 As shown, an AI-based method for filling out and approving travel expense reports is applied to the expense approval process. The main workflow of this method is described below (steps S201 to S203): Step S201: Receive expense report.

[0049] The reimbursement form includes the type of receipt, such as invoice, itinerary, boarding pass, etc.

[0050] The expense reimbursement review system receives expense reports uploaded by users and reviews them using an AI model.

[0051] Step S202: Use an AI model to conduct a compliance review of the expense report and determine the compliance review result.

[0052] Finance staff only need to click the audit button, and the AI ​​model will conduct a compliance review of the expense report and output the compliance review result. The compliance review result includes two situations: compliance review passed and compliance review failed. If the compliance review fails, the problematic content can be marked with different colored fonts and the reasons for the problem can be noted, making it easy for finance staff to view.

[0053] Specifically, the compliance review of expense reports is conducted using an AI model to determine the compliance review results. This includes: calling the OA interface to obtain the business trip application period associated with the expense report; reviewing the dates of the receipts in the expense report based on the business trip application period to determine the date review results; conducting a completeness review of the expense report based on the receipt types to determine the completeness review results; establishing a business trip timeline based on the receipts in the expense report, with the business trip timeline including the business trip time and the business trip location; conducting a timeline consistency review of the itinerary's spatiotemporal logic to determine the itinerary review results; reviewing the expense report's reimbursement amount based on RAG technology to determine the amount review results; and determining the compliance review results based on the date review results, completeness review results, itinerary review results, and amount review results.

[0054] In this embodiment, the AI ​​model conducts compliance reviews through several aspects, including date reasonableness review, material completeness review, trip time and space logical consistency review, and reimbursement amount review, and obtains the compliance review results: When generating an expense report, the expense report system identifies the associated business trip application form. This is the business trip application form found during the preliminary review of invoice information. The system retrieves this associated business trip application form by calling the OA interface, thereby determining the business trip application period in the application form. The system then reviews whether the dates on the invoices in the expense report fall within the business trip application period. If all invoice dates fall within the business trip application period, the date review result is "Date Review Passed." If any invoice dates fall outside the business trip application period, the date review result is "Date Review Failed," and the invoice with an unreasonable date is marked.

[0055] Different types of receipts require different materials to be uploaded simultaneously. The association rules between receipt types and required materials are pre-set and not specifically limited here. For example, if the receipt type is an airplane invoice or itinerary, the required materials include a boarding pass; if the receipt type is an accommodation invoice, the required materials include an accommodation bill. The completeness of the receipt types and materials in the reimbursement form is reviewed according to the association rules between receipt types and required materials. If the materials are complete, the completeness review result is "materials approved"; if the materials are incomplete, the completeness review result is "materials rejected", and the problematic materials are marked.

[0056] The AI ​​model creates a business trip timeline based on information such as the date, type, and location of each receipt in the expense report, displaying the user's travel trajectory in chronological order. The AI ​​model then reviews the travel trajectory. If the user's travel trajectory is reasonable, the travel review result is "approved"; if the user's travel trajectory is unreasonable (e.g., 10:00 AM in Beijing, 10:30 AM in Shenzhen), the travel review result is "failed," and the unreasonable itinerary is marked.

[0057] A travel reimbursement policy knowledge base is constructed based on RAG technology. The knowledge base includes detailed reimbursement amount standards for different job levels, regions, and seasons. The AI ​​model reviews the reimbursement amounts on the reimbursement forms using the knowledge base. If the reimbursement amounts all meet the standards in the knowledge base, the review result is "approved." If there are any reimbursement amounts that do not meet the standards in the knowledge base, the review result is "failed," and the non-compliant reimbursement amounts and the corresponding amount range in the rules are marked.

[0058] The results of date review, completeness review, itinerary review, and amount review are collectively determined as the compliance review result. If all results of the date review, completeness review, itinerary review, and amount review are approved, the compliance review result is approved. If any one or more results of the date review, completeness review, itinerary review, and amount review are rejected, the compliance review result is rejected.

[0059] More specifically, the reimbursement amount is reviewed based on RAG technology to determine the review result, including: converting the table segments in the preset standards into text format; constructing a travel reimbursement policy knowledge base based on the preset standards in text format; extracting the business trip location, business trip role, and business trip date from the reimbursement form using an AI model; searching the travel reimbursement policy knowledge base based on the business trip location to determine the hit segment; and reviewing the reimbursement amount based on the business trip role, business trip date, and hit segment to obtain the review result.

[0060] In this embodiment, the table segments in the preset standard are converted into text format using RAG technology. The text format includes multiple segments, with one rule corresponding to one segment. By dividing the table into segments, with one region corresponding to one segment, the accuracy of subsequent searches is improved.

[0061] After converting the table segments in the preset standards into text format, a travel reimbursement system knowledge base is constructed using the preset standards in text format. The AI ​​model extracts the business trip location, business trip role (e.g., ministerial level personnel, bureau-level personnel, other personnel, etc.) and business trip date from the reimbursement form. The travel reimbursement system knowledge base is searched according to the business trip location, and the segment corresponding to the business trip location in the travel reimbursement system knowledge base is determined as the hit segment. The reimbursement amount of the reimbursement form is reviewed based on the amount standard corresponding to the business trip role and business trip date in the hit segment, and the amount review result is obtained.

[0062] Step S203: If the compliance review result is "approved", then calculate the deductible input tax based on the invoice type.

[0063] If the compliance review is passed, the calculation rules for deductible input tax are retrieved from the database according to the type of invoice (determined according to national standards, which will not be elaborated here), and the deductible input tax is calculated according to the calculation rules, thereby reducing the workload of financial personnel and improving review efficiency.

[0064] If the compliance review fails, the result will be shown to the finance staff to facilitate their subsequent processing.

[0065] like Figure 3 As shown, the AI-based travel expense reimbursement filling and approval system 300, applied to the reimbursement filling terminal, mainly includes: The first acquisition module 301 is used to acquire user-uploaded information; Information extraction module 302 is used to extract ticket information from user-uploaded information based on a preset visual model and document extraction tools; Information pre-screening module 303 is used to pre-screen invoice information based on an AI model; The reimbursement data structure extraction 304 is used to extract structured data from the reimbursement form based on the invoice information if the preliminary review is approved. The expense report filling module 305 is used to transmit the structured data of the expense report to the OA interface to create a travel expense reimbursement application page, upload the receipts in the user-uploaded information to the travel expense reimbursement application page through the OA interface, and display the link to the travel expense reimbursement application page; The expense report submission module 306 is used to generate an expense report and submit it to the expense review terminal if the user clicks the confirmation button on the travel expense report application page.

[0066] As an optional implementation in this embodiment, the information pre-review module 303 is also specifically used to pre-review the invoice information based on the AI ​​model, including: analyzing the invoice information based on the AI ​​model and establishing a business trip timeline; calling the OA interface to query whether a business trip application form exists based on the start time of the business trip timeline; if no business trip application form exists, calling the OA interface to query whether a business trip application form exists based on the invoice location in the invoice information; if a business trip application form exists, the invoice information pre-review is passed.

[0067] As an optional implementation in this embodiment, the expense report generation module 304 is also specifically used to extract structured expense report data based on invoice information, including: extracting initial structured expense report data from invoice information based on invoice type and corresponding preset JSON field format; supplementing the initial structured expense report data based on the business trip application form to obtain structured expense report data.

[0068] like Figure 4 As shown, the AI-based travel expense reimbursement filling and approval system 400, applied to the reimbursement approval end, mainly includes: The expense report receiving module 401 is used to receive expense reports, which include invoice types. The expense report review module 402 is used to conduct compliance reviews of expense reports using an AI model and determine the compliance review results. The deductible input tax calculation module 403 is used to calculate the deductible input tax based on the invoice type if the compliance review result is passed.

[0069] As an optional implementation in this embodiment, the expense report review module 402 is also specifically used to conduct compliance review of expense reports using an AI model and determine the compliance review result, including: calling the OA interface to obtain the business trip application period associated with the expense report; reviewing the time of the invoices in the expense report based on the business trip application period to determine the date review result; conducting a completeness review of the expense report based on the invoice type to determine the completeness review result; establishing a business trip timeline based on the invoices in the expense report, where the business trip timeline is the timeline corresponding to the business trip time and the business trip location; conducting a trip timeline consistency review of the trip timeline to determine the trip review result; reviewing the expense report amount based on RAG technology to determine the amount review result; and determining the compliance review result based on the date review result, completeness review result, trip review result, and amount review result.

[0070] As an optional implementation in this embodiment, the expense report review module 402 is also specifically used to review the expense report amount based on RAG technology and determine the amount review result, including: converting the table segments in the preset standard into text format; constructing a travel expense reimbursement system knowledge base based on the preset standard in text format; extracting the business trip location, business trip role, and business trip date from the expense report based on an AI model; searching the travel expense reimbursement system knowledge base based on the business trip location to determine the hit segment; and reviewing the expense report amount based on the business trip role, business trip date, and hit segment to obtain the amount review result.

[0071] In one example, a module in any of the above systems may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0072] For example, when modules in a system can be implemented through a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor capable of calling programs. Alternatively, these modules can be integrated together to form a system-on-a-chip (SOC).

[0073] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the system and modules described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0074] Figure 5 This is a structural block diagram of an electronic device 500 provided in an embodiment of this application.

[0075] like Figure 5 As shown, the electronic device 500 includes a processor 501 and a memory 502, and may further include one or more of an information input / output (I / O) interface 503, a communication component 504, and a communication bus 505.

[0076] The processor 501 controls the overall operation of the electronic device 500 to complete all or part of the steps of the AI-based travel expense reimbursement filling and review method described above. The memory 502 stores various types of data to support the operation of the electronic device 500. This data may include, for example, instructions for any application or method operating on the electronic device 500, as well as application-related data. The memory 502 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as one or more of 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 storage, flash memory, magnetic disk, or optical disk.

[0077] I / O interface 503 provides an interface between processor 501 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 504 is used for wired or wireless communication between electronic device 500 and other devices. Wireless communication includes Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 504 may include a Wi-Fi component, a Bluetooth component, and an NFC component.

[0078] The electronic device 500 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the AI-based travel expense reimbursement filling and review method given in the above embodiments.

[0079] The communication bus 505 may include a path for transmitting information between the aforementioned components. The communication bus 505 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 505 can be divided into an address bus, a data bus, a control bus, etc.

[0080] Electronic device 500 may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers, and may also include servers.

[0081] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described AI-based travel expense reimbursement filling and review method.

[0082] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0083] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0084] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. An AI-based method for filling out and reviewing travel expense reports, characterized in that, Applications include: (The text abruptly ends here, likely due to an incomplete sentence or a formatting error.) Obtain user-uploaded information; Based on a preset visual model and document extraction tools, the ticket information in the user-uploaded information is extracted; The invoice information is pre-screened based on an AI model; If the preliminary review is approved, the structured data of the expense report will be extracted based on the invoice information. The structured data of the expense report is transmitted to the OA interface to create a travel expense reimbursement application page. The receipts in the user-uploaded information are uploaded to the travel expense reimbursement application page through the OA interface, and a link to the travel expense reimbursement application page is displayed. If the user clicks the confirmation button on the travel expense reimbursement application page, an expense report will be generated and submitted to the expense review terminal. The preliminary review of the invoice information based on the AI ​​model includes: The AI ​​model is used to analyze the ticket information and establish a business trip timeline. The OA interface is invoked to query whether a business trip application form exists based on the start time of the business trip timeline. If the business trip application form does not exist, the OA interface is called to query whether a business trip application form exists based on the location of the ticket in the ticket information; If the aforementioned business trip application form exists, then the preliminary review of the invoice information is approved; The invoice information includes the invoice type, and the extraction of structured data from the expense reimbursement form based on the invoice information includes: Based on the invoice type and the corresponding preset JSON field format, the initial expense report structured data is extracted from the invoice information; The structured data of the initial expense report is supplemented based on the business trip application form to obtain the structured data of the expense report.

2. An AI-based method for filling out and reviewing travel expense reports, characterized in that, Applications include: Receive expense reports, which include invoice types; The reimbursement form is reviewed for compliance using an AI model to determine the compliance review result. If the compliance review result is "approved", then the deductible input tax is calculated based on the invoice type. The process of using an AI model to conduct a compliance review of the expense report and determine the compliance review result includes: Call the OA interface to obtain the business trip application period associated with the expense report; Based on the business trip application period, the dates of the receipts in the expense report are reviewed to determine the date review results; Based on the type of invoice in the expense report, perform a completeness review on the expense report and determine the completeness review result; A business trip timeline is established based on the receipts in the expense report, wherein the business trip timeline is the timeline corresponding to the business trip time and the business trip location. The timeline of the business trip is reviewed for consistency in terms of time and space logic to determine the results of the timeline review. The reimbursement form is reviewed based on RAG technology to determine the amount review result. The compliance review result is determined based on the date review result, the integrity review result, the itinerary review result, and the amount review result; The process of reviewing the reimbursement amount on the expense report based on RAG technology and determining the review result includes: Convert table segments in the preset standard into text format; A knowledge base for travel reimbursement systems is constructed based on the preset standards in text format. Based on the AI ​​model, the business trip location, business trip role, and business trip date are extracted from the expense report; The knowledge base of the travel reimbursement system is searched based on the business trip location to determine the hit segment; The expense report is reviewed based on the traveler's role, travel date, and hit segment to obtain the expense review result.

3. An AI-based travel expense reimbursement form and approval system, characterized in that, Applied to the reimbursement form filling terminal, for implementing the method as described in claim 1, including: The first acquisition module is used to acquire user-uploaded information; The information extraction module is used to extract ticket information from the user-uploaded information based on a preset visual model and document extraction tools; The information pre-screening module is used to pre-screen the invoice information based on an AI model; The reimbursement data structure extraction module is used to extract structured data of the reimbursement form based on the invoice information if the preliminary review is passed. The expense report filling module is used to transmit the structured data of the expense report to the OA interface to create a travel expense reimbursement application page, upload the receipts in the user-uploaded information to the travel expense reimbursement application page through the OA interface, and display a link to the travel expense reimbursement application page; The expense report submission module is used to generate an expense report and submit it to the expense review terminal if the user clicks the confirmation button on the travel expense report application page.

4. An AI-based travel expense reimbursement form and approval system, characterized in that, Applied to the expense reimbursement approval terminal, for implementing the method as described in claim 2, including: The expense report receiving module is used to receive expense reports, which include invoice types. The expense report review module is used to conduct a compliance review of the expense report using an AI model and determine the compliance review result. The deductible input tax calculation module is used to calculate the deductible input tax based on the invoice type if the compliance review result is that the review is passed.

5. An electronic device, characterized in that, Includes a processor, which is coupled to a memory; The processor is configured to execute a computer program stored in the memory, causing the electronic device to perform the method as described in any one of claims 1 to 2.

6. A computer-readable storage medium, characterized in that, It includes a computer program or instructions that, when run on a computer, cause the computer to perform the method as described in any one of claims 1 to 2.

Images