Reimbursement receipt generation method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a reimbursement receipt generation method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring each reimbursement invoice corresponding to a target user, and extracting invoice key information corresponding to each reimbursement invoice; generating bill detail information based on the invoice key information; generating each piece of travel information based on the time sequence displacement information in the document detail information, and extracting travel detail information corresponding to each piece of travel information from the document detail information; and obtaining a target reimbursement receipt corresponding to the target user based on each piece of travel information and the travel detail information. By adopting the method, the reimbursement receipt generation accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, apparatus, computer device, computer-readable storage medium and computer program product for generating a reimbursement document. Background Art

[0002] Financial reimbursement is an essential part of every unit's operation. Traditional financial reimbursement usually uses paper reimbursement application forms and reimbursement vouchers. The reimbursement agent needs to fill out the reimbursement, calculate the corresponding reimbursement-related expenses based on the reimbursement voucher, and paste the paper reimbursement voucher. This method is not only inefficient, but also with the increase in business activities, the preservation and search of related paper documents are inconvenient.

[0003] With the popularization of electronic reimbursement vouchers such as electronic invoices and electronic itineraries, in the prior art, although the reimbursement agent can fill out the reimbursement application form and insert the electronic reimbursement voucher through a special program, the entire operation process requires the reimbursement agent to perform manual operations. For business trips that may generate a large number of reimbursement vouchers, not only is it cumbersome to collect and organize the vouchers, but also it is usually necessary to fill in a lot of business trip itinerary content when making reimbursements, which takes a lot of time to fill in, and there is a problem of low efficiency in generating reimbursement documents. Summary of the invention

[0004] Based on this, it is necessary to provide a reimbursement document generation method, apparatus, computer equipment, computer-readable storage medium and computer program product that can improve the efficiency of reimbursement document generation in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for generating a reimbursement document, comprising:

[0006] Obtain each reimbursement invoice corresponding to the target user, and extract the key invoice information corresponding to each reimbursement invoice;

[0007] Generate document details based on invoice key information;

[0008] Generate each trip information based on the time series displacement information in the document detail information, and extract the trip detail information corresponding to each trip information from the document detail information;

[0009] Based on each itinerary information and itinerary detail information, the target reimbursement document corresponding to the target user is obtained.

[0010] In a second aspect, the present application also provides a reimbursement document generating device, comprising:

[0011] The invoice extraction module is used to obtain each reimbursement invoice corresponding to the target user and extract the key invoice information corresponding to each reimbursement invoice;

[0012] Detail generation module, used to generate document detail information based on invoice key information;

[0013] A trip generation module, used to generate each trip information based on the time sequence displacement information in the document detail information, and extract the trip detail information corresponding to each trip information from the document detail information;

[0014] The document generation module is used to obtain the target reimbursement document corresponding to the target user based on each itinerary information and itinerary detail information.

[0015] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0016] Obtain each reimbursement invoice corresponding to the target user, and extract the key invoice information corresponding to each reimbursement invoice;

[0017] Generate document details based on invoice key information;

[0018] Generate each trip information based on the time series displacement information in the document detail information, and extract the trip detail information corresponding to each trip information from the document detail information;

[0019] Based on each itinerary information and itinerary detail information, the target reimbursement document corresponding to the target user is obtained.

[0020] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0021] Obtain each reimbursement invoice corresponding to the target user, and extract the key invoice information corresponding to each reimbursement invoice;

[0022] Generate document details based on invoice key information;

[0023] Generate each trip information based on the time series displacement information in the document detail information, and extract the trip detail information corresponding to each trip information from the document detail information;

[0024] Based on each itinerary information and itinerary detail information, the target reimbursement document corresponding to the target user is obtained.

[0025] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0026] Obtain each reimbursement invoice corresponding to the target user, and extract the key invoice information corresponding to each reimbursement invoice;

[0027] Generate document details based on invoice key information;

[0028] Generate each trip information based on the time series displacement information in the document detail information, and extract the trip detail information corresponding to each trip information from the document detail information;

[0029] Based on each itinerary information and itinerary detail information, the target reimbursement document corresponding to the target user is obtained.

[0030] The above-mentioned reimbursement document generation method, device, computer equipment, computer-readable storage medium and computer program product, by extracting invoice key information from each reimbursement invoice and generating document detail information based on the invoice key information, without the need for manual editing, improves the efficiency of generating document detail information; then generates each itinerary information based on the time series displacement information in the document detail information, and extracts the itinerary detail information of each itinerary information from the document detail information, thereby realizing automatic division of itineraries of the document detail information; then, based on each itinerary information and itinerary detail information, obtains the target reimbursement document of the target user, and, when there are reimbursement invoices for multiple itineraries, by automatically dividing the itineraries of the document detail information, the corresponding itinerary information and itinerary detail information are quickly generated, without the need for manual sorting of reimbursement invoices, and also saves the time for entering reimbursement information, thereby improving the efficiency of generating reimbursement documents. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 An application environment diagram of a method for generating a reimbursement document in an embodiment;

[0033] Figure 2 A schematic diagram of a process for generating a reimbursement document in one embodiment;

[0034] Figure 3 A schematic diagram of a process for generating a reimbursement document in one embodiment;

[0035] Figure 4 A schematic diagram of a processing flow of a reimbursement document in one embodiment;

[0036] Figure 5 It is a structural block diagram of a reimbursement document generating device in one embodiment;

[0037] Figure 6is an internal structure diagram of a computer device in one embodiment;

[0038] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in another embodiment. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0040] The reimbursement document generation method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 obtains each reimbursement invoice corresponding to the target user, and extracts the key invoice information corresponding to each reimbursement invoice; the terminal 102 generates the document detail information based on the key invoice information; the terminal 102 generates each trip information based on the time sequence displacement information in the document detail information, and extracts the trip detail information corresponding to each trip information from the document detail information; the terminal 102 obtains the target reimbursement document corresponding to the target user based on the each trip information and the trip detail information, and the terminal 102 can store the target reimbursement document to the server 104. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. The server 104 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0041] In one embodiment, Figure 2 As shown, a method for generating a reimbursement document is provided, and the method is applied to Figure 1 The terminal in the example is used for explanation. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0042] Step 202, obtaining each reimbursement invoice corresponding to the target user, and extracting the key invoice information corresponding to each reimbursement invoice.

[0043] The target user refers to the user who needs to make an invoice reimbursement. The reimbursement invoice refers to the consumption voucher that can be used for reimbursement. The invoice key information refers to the information in the reimbursement invoice that represents the consumption details.

[0044] Specifically, the terminal receives the reimbursement application initiated by the target user, obtains each reimbursement invoice corresponding to the target user, and extracts the key invoice information of each reimbursement invoice, such as the invoice date, invoice location, departure place, destination, invoice type, project name, amount, tax rate, tax amount, etc. The terminal can extract the key invoice information from each reimbursement invoice according to keywords, or input each reimbursement invoice into the information extraction model to obtain the key invoice information of each reimbursement invoice output by the information extraction model. The information extraction model is, for example, an LLM (Large Language Model) model.

[0045] Step 204: Generate document detail information based on the invoice key information.

[0046] The document detail information is the detail information used to generate the reimbursement document. The detail information may refer to the consumption details of the target user, including transportation, catering, accommodation and other consumption details.

[0047] Specifically, the terminal sorts the key invoice information of each reimbursement invoice according to the date information in the key invoice information, such as the invoice issuance date and consumption date of the reimbursement invoice, to obtain the invoice sequence information. Then, the document requirement information corresponding to the reimbursement document type is obtained. The document requirement information refers to the necessary information required to generate the reimbursement document, such as the travel type, expense items, reimbursement reasons, etc. The terminal performs data conversion on the invoice sequence information according to the document requirement information to obtain the document detail information. It can be that the reimbursement supplementary information is generated for the invoice sequence information according to the document requirement information, the reimbursement supplementary information is added to the invoice sequence information, and the format of the added invoice sequence information is converted according to the preset document detail format to obtain the document detail information.

[0048] Step 206, generating each trip information based on the time series displacement information in the document detail information, and extracting the trip detail information corresponding to each trip information from the document detail information.

[0049] Among them, the temporal displacement information is used to represent the target user's location movement information in chronological order. The itinerary information refers to the information of the target user's corresponding movement route. The itinerary detail information refers to the consumption details generated by the target user in the corresponding itinerary.

[0050] Specifically, after generating the document details, the terminal extracts each date information and the traffic information corresponding to each date information from the document details in chronological order, where the traffic information includes the location information of the target user, and then determines the time sequence displacement information corresponding to the document details based on the date information and the location information in the corresponding traffic information, for example: Date 2024-10-11: Location 1—2024-10-11: Location 2—2024-10-11: Location 2— … —2024-10-13: Location 3.

[0051] The terminal divides the time series displacement information into itineraries according to the position change characteristics in the time series displacement information to obtain each itinerary information, such as location 1-location 2, location 2-location 3, etc., and then extracts the consumption details of the corresponding itinerary from the document details according to each itinerary information to obtain the itinerary detail information corresponding to each itinerary information.

[0052] Step 208, based on each itinerary information and itinerary detail information, obtain the target reimbursement document corresponding to the target user.

[0053] The target reimbursement document refers to the document that the target user can reimburse.

[0054] Specifically, the document requirement information also includes the document data format. After obtaining each itinerary information and the corresponding itinerary detail information, the terminal performs data format conversion on each itinerary information and the corresponding itinerary detail information according to the document data format, which may be a data structure format conversion to obtain the target reimbursement document. For example, each itinerary detail information is added to the corresponding itinerary information, converted into a parent-child data structure between the itinerary information and the itinerary detail information, and the target reimbursement document is generated according to each itinerary information and the itinerary detail information after the data structure conversion.

[0055] In the above-mentioned reimbursement document generation method, by extracting invoice key information from each reimbursement invoice and generating document detail information based on the invoice key information, manual editing is not required, thereby improving the efficiency of generating document detail information; then, each itinerary information is generated based on the time series displacement information in the document detail information, and the itinerary detail information of each itinerary information is extracted from the document detail information, thereby realizing automatic division of itineraries of the document detail information; then, based on each itinerary information and itinerary detail information, the target reimbursement document of the target user is obtained, and when there are reimbursement invoices for multiple itineraries, the itinerary of the document detail information is automatically divided, thereby realizing the rapid generation of corresponding itinerary information and itinerary detail information, without the need for manual sorting of reimbursement invoices, and also saving the time for entering reimbursement information, thereby improving the efficiency of generating reimbursement documents.

[0056] In one embodiment, Figure 3As shown, step 204 generates document details based on the invoice key information, including:

[0057] Step 302, inputting the key information of the invoice into the classification model to obtain the objective reimbursement field determined from the candidate objective fields;

[0058] Step 304, obtaining the historical reimbursement documents corresponding to the target user, inputting the key invoice information and the historical reimbursement documents into the generative model, and obtaining the subjective reimbursement field;

[0059] Step 306, obtaining document detail information based on the objective reimbursement field and the subjective reimbursement field.

[0060] Among them, the classification model is used to determine the objective reimbursement field of the classification to which the invoice key information belongs from the candidate objective fields. The candidate objective fields refer to the objective fields corresponding to the multiple attribute categories related to the reimbursement expenses that are pre-set. The attribute categories related to the reimbursement expenses are, for example, "travel type", "expense items", etc., and the objective fields are pre-defined expense attribute fields corresponding to the expense types, such as "accommodation expenses", "railway tickets", "business trip allowances", etc. corresponding to "travel type", and "entertainment expenses", "office expenses", "sales expenses", etc. corresponding to "expense items". The objective reimbursement field refers to the expense attribute field of the corresponding expense type determined based on the invoice key information.

[0061] The generative model is used to generate text fields related to user reimbursement, such as description fields or notes fields such as "I went on a business trip to xx place from xx to xx time due to xx reason". The text fields related to user reimbursement output by the generative model can be used as subjective fields. Subjective reimbursement fields refer to subjective fields output by the generative model based on invoice key information and historical reimbursement documents.

[0062] Specifically, the terminal obtains the key invoice information corresponding to each reimbursement invoice, sorts the key invoice information of each reimbursement invoice in chronological order, and obtains the invoice chronological information. The document requirement information corresponding to the reimbursement document type is obtained, and the invoice chronological information is input into a classification model, such as a decision tree model, according to the document requirement information. The classification model determines the objective reimbursement fields corresponding to each reimbursement invoice from each candidate objective field according to the invoice chronological information. Then, the historical reimbursement documents corresponding to the target user are obtained, and the invoice chronological information and the historical reimbursement documents are input into a generative model, such as an LLM (large language model) model such as ChatGPT. The generative model outputs the subjective reimbursement fields corresponding to the target user according to the historical subjective fields in the historical reimbursement documents and the invoice chronological information. The objective reimbursement fields corresponding to each reimbursement invoice output by the classification model and the subjective reimbursement fields output by the generative model are used as reimbursement supplementary information, the reimbursement supplementary information is added to the invoice sequence information, and the format of the added invoice sequence information is converted according to the preset document detail format to obtain the document detail information.

[0063] In this embodiment, the objective reimbursement fields output by the classification model and the subjective reimbursement fields output by the generation model are added to the invoice timing information as reimbursement supplementary information according to the document requirement information, and the document detail information is generated, thereby ensuring the accuracy of the document detail information.

[0064] In one embodiment, step 206, generating each trip information based on the time sequence displacement information in the document detail information, and extracting the trip detail information corresponding to each trip information from the document detail information, includes:

[0065] Based on the date information and traffic information in the document details, the time series displacement information is obtained;

[0066] Based on the departure and destination corresponding to the target traffic information in the time series displacement information, the information of each trip is obtained;

[0067] The sub-document detail information generated at the destination in the itinerary information is obtained from the document detail information as the itinerary detail information corresponding to the itinerary information, and the itinerary detail information corresponding to each itinerary information is obtained.

[0068] Among them, time series displacement information refers to location information arranged in time series. Itinerary information refers to the itinerary information of the target user across regions, generally itinerary information between different cities. Target transportation information is transportation information used to determine the specified transportation type in the itinerary information, such as cross-regional transportation types such as railways and airplanes.

[0069] Specifically, the terminal extracts various traffic information from the document details, including traffic information between different regions, such as air ticket information between different cities, and traffic information between the same region, such as online car-hailing information between the same city. Then, the location information corresponding to each traffic information is extracted, including the departure place and destination, and the location information corresponding to each traffic information is sorted according to the date information corresponding to each traffic information to obtain the time series displacement information. The departure place and destination corresponding to the target traffic information are extracted from the time series displacement information, such as the departure place and destination corresponding to the railway ticket, and the departure place and destination corresponding to the air ticket, to obtain the cross-regional itinerary information.

[0070] Then, the sub-document details generated by the destination in the itinerary information are obtained from the document details. For example, if the target user's itinerary 1 is Shenzhen-Shanghai, the consumption details in Shanghai are determined based on the consumption date and consumption location of each consumption detail in the document details. If so, the consumption details of the target user in Shanghai are obtained from the document details to obtain the sub-document details; if not, an abnormal location reminder is issued. Based on the sub-document details corresponding to the destination in each itinerary information, the itinerary details corresponding to each itinerary information are obtained.

[0071] In this embodiment, each trip information is determined based on the document detail information combined with cross-regional traffic information, thereby ensuring the accuracy of the trip information and the accuracy of the trip detail information, so that the target reimbursement document can be subsequently generated based on the each trip information and trip detail information, thereby improving the accuracy of the reimbursement document.

[0072] In one embodiment, the reimbursement document generation method further includes:

[0073] Based on the date information and traffic information in the document details, determine each traffic travel time and the corresponding traffic travel location; based on the document details, when there is abnormal consumption details in each traffic travel time, generate an abnormal time consumption alarm;

[0074] In a preset time period after the travel time, when there is consumption details at a different location from the destination in the travel location, an abnormal travel consumption alert is generated.

[0075] The travel time refers to the time information when taking a transportation vehicle, including the start time and the end time of the ride. The travel location refers to the start location and the end destination of the ride when taking a transportation vehicle.

[0076] Specifically, after generating the document details, the terminal determines each travel time and corresponding travel location according to the date information and traffic information in the document details. For example, the travel time is the take-off time and landing time of the plane from Shenzhen to Shanghai, where Shenzhen and Shanghai are the travel locations of the travel time. Abnormal time consumption monitoring and off-site consumption monitoring are performed on each travel time and corresponding travel location.

[0077] Abnormal time consumption monitoring can be that the terminal matches each consumption location and consumption time in the document details with each travel time and the corresponding travel location. According to the matching results, when it is monitored that there is abnormal consumption detail information within each travel time, for example, there is a reimbursement invoice within the time period of Shenzhen-Shanghai flight take-off time and plane landing time, the travel information corresponding to the travel time with abnormal consumption detail information is determined as abnormal time consumption, and an abnormal time consumption alarm is generated.

[0078] The out-of-town consumption monitoring can be when it is detected that within a preset time period after the travel time, there is out-of-town consumption details that are different from the travel location. For example, in a Shenzhen-Shanghai trip, within one day after arriving in Shanghai, there is out-of-town consumption in non-Shanghai places such as Shenzhen or Hangzhou. In this case, the travel location where the out-of-town consumption details exist will be determined as abnormal travel consumption, and an abnormal travel consumption alert will be generated.

[0079] In a specific embodiment, the document detail information can also be used to monitor unconventional consumption, and the consumption frequency of the same consumption type in the same time period can be counted based on the document detail information. When the statistical number reaches an upper limit, it is determined that unconventional consumption exists in the time period, such as 5 dining consumptions in one day, and an unconventional consumption alert is generated.

[0080] In this embodiment, by monitoring consumption at abnormal times and consumption in different locations on the document detail information, the reasonable validity and accuracy of the document detail information can be ensured.

[0081] In one embodiment, the reimbursement document generation method further includes:

[0082] Obtain a reference reimbursement file, and determine an initial document approval standard based on the reference reimbursement file; the initial document approval standard includes reference reimbursement standards corresponding to multiple reference reimbursement items;

[0083] Identify target reimbursement items from reference reimbursement items;

[0084] Obtain a historical reimbursement document set corresponding to the user set, and determine a reimbursement warning value corresponding to the target reimbursement item based on the historical reimbursement document set and a reference reimbursement standard corresponding to the target reimbursement item;

[0085] Add the reimbursement warning value corresponding to the target reimbursement item to the initial document approval standard to obtain the target document approval standard, which is used to approve the target reimbursement document.

[0086] Among them, the reference reimbursement file is a file used to provide document approval standards. The target reimbursement item refers to the reimbursement item involving expenses, such as accommodation expenses, food expenses, etc. The reference reimbursement standard refers to the expense threshold of the reimbursement expenses corresponding to the target reimbursement item. The reimbursement warning value refers to the excess value allowed to exceed the reference reimbursement standard for the target reimbursement item until the warning is reached. The initial document approval standard is the document approval standard in the initial format and the reimbursement warning value is to be supplemented. The target document trial standard is the trial standard generated according to the approval format, which includes the reimbursement warning value. The document trial standard is used to verify whether the reimbursement expenses in the reimbursement document exceed the reference reimbursement standard.

[0087] Specifically, the terminal obtains a reference reimbursement file, such as a reimbursement rule issued by an enterprise, and generates an initial document approval standard based on the reference reimbursement file, such as an initial document approval standard that converts the reference reimbursement file into an initial format through an LLM (large language model) model. The initial document approval standard can be in a table form. Then, the historical reimbursement document set corresponding to the user set is obtained. The historical reimbursement document set includes historical reimbursement documents applied for by each user in the user set in a historical time period, as well as reimbursement expenses and approval results corresponding to the historical reimbursement documents. The approval result can be a manual approval result. The terminal searches the historical reimbursement document set for the reimbursement expenses of the target reimbursement project that exceed the corresponding reference reimbursement standard, and the approval result is the approved historical reimbursement document. According to the difference between the reimbursement expenses in the historical reimbursement document found and the corresponding reference reimbursement standard, the reimbursement warning value corresponding to the target reimbursement project is determined, which can be a percentage value. The target reimbursement project can refer to a reimbursement project of the expense type. Then, the reimbursement warning value corresponding to the target reimbursement project is added to the initial document approval standard to obtain the target document approval standard, which is used to approve the target reimbursement document. The target document approval standard can be in table form, such as the target document approval standard corresponding to the reimbursement person Zhang San shown in Table 1 below:

[0088] Table 1

[0089] City Level City Name Reimbursement Peak and off season Transportation Seat Standard Accommodation Standards Warning line Catering standards Warning line Warning level First Line Beijing Zhang San Peak Season railway First Class 500 15% 300 5% Strong reminder First Line Beijing Zhang San Peak Season airplane Economy Class 500 15% 300 5% Strong reminder Second Line Dongguan Zhang San Off-season airplane Economy Class 300 12% 200 10% Weak reminder

[0090] As shown in Table 1, non-expense-type reimbursement items such as city level, city name, peak and off-peak season, transportation, and seat standards and their corresponding reference item values, expense-type reimbursement items such as accommodation standards and catering standards and their corresponding reference reimbursement standards, as well as warning levels are included to indicate the warning intensity when the reimbursement expenses exceed the reference reimbursement standards.

[0091] In this embodiment, the reimbursement warning value corresponding to the target reimbursement item is determined based on the historical reimbursement document set and the reference reimbursement standard corresponding to the target reimbursement item. The reimbursement warning value enables flexibility in the approval of reimbursement documents, eliminating the need for repeated manual review and improving the approval efficiency of reimbursement documents.

[0092] In one embodiment, the reimbursement warning value is obtained by inputting the historical reimbursement document set and the reference reimbursement standard corresponding to the target reimbursement item into the reimbursement standard analysis model, and based on the output of the reimbursement standard analysis model; the reimbursement document generation method further includes:

[0093] When the target reimbursement document fails to be approved, obtain the manual approval result corresponding to the target reimbursement document;

[0094] The target reimbursement document and the manual approval result are used as new training samples, and the new training samples are added to the training sample set;

[0095] Based on the training sample set, optimize the reimbursement standard analysis model.

[0096] Specifically, a reimbursement standard analysis model can be set in advance. The reimbursement standard analysis model can be an RNN neural network model, which is used to analyze and set the reimbursement warning value in the document approval standard. After the terminal determines the initial document approval standard based on the reference reimbursement file, the terminal can input the historical reimbursement document set and the initial document approval standard corresponding to the user set into the reimbursement standard analysis model, and obtain the reimbursement warning value corresponding to the target reimbursement item through the reimbursement standard analysis model. Then, the reimbursement warning value corresponding to the target reimbursement item is added to the initial document approval standard to obtain the target document approval standard.

[0097] The terminal uses the target document approval standard to approve the target reimbursement documents within the preset time period. When the target reimbursement document approval fails, the manual approval result corresponding to the target reimbursement document is obtained. When there are multiple target reimbursement documents that fail to be approved and the corresponding manual approval result is approved, it means that the reimbursement warning value corresponding to the existing target reimbursement item in the target document approval standard is no longer applicable, and the reimbursement warning value needs to be adjusted. The terminal uses the target reimbursement documents that fail to be approved and whose corresponding manual approval result is approved as new training samples, adds the new training samples to the training sample set corresponding to the reimbursement standard analysis model, uses the training sample set to optimize the reimbursement standard analysis model, and obtains the updated reimbursement warning value corresponding to the target reimbursement item output by the optimized reimbursement standard analysis model. Update the target document approval standard according to the updated reimbursement warning value.

[0098] In this embodiment, when the approval result of the target document approval standard is inconsistent with the approval result given by manual confirmation during the approval process of the target reimbursement document, the reimbursement warning value corresponding to the target reimbursement item is updated through the reimbursement standard analysis model, thereby realizing dynamic adjustment of the reimbursement warning value and ensuring the approval accuracy of the reimbursement document according to the target document approval standard.

[0099] In one embodiment, the reimbursement document generation method further includes:

[0100] From the target reimbursement document, obtain the abnormal reimbursement items with approval exceptions and the current reimbursement expenses of the abnormal reimbursement items in the current item value;

[0101] Determine the reference item value of the abnormal reimbursement item from the target document approval criteria, and query the actual cost of the abnormal reimbursement item based on the reference item value;

[0102] When the current reimbursement expenses do not exceed the actual expenses, the approval result of the abnormal reimbursement item will be updated to approved.

[0103] Among them, abnormal reimbursement items refer to reimbursement items whose item values ​​are inconsistent with the reference item values ​​of the corresponding reimbursement items in the target document approval standard. Current item value refers to the item value corresponding to the abnormal reimbursement item in the target reimbursement document. Current reimbursement expenses refer to the reimbursement expenses corresponding to the abnormal reimbursement item in the target reimbursement document.

[0104] Specifically, the terminal verifies each reimbursement item in the target reimbursement document through the target document approval standard, which can be to match the item value of each reimbursement item in the target reimbursement document with the reference item value of the corresponding reimbursement item in the target document approval standard. When the item value of the expense type reimbursement item (i.e., the reimbursement expense) meets the item value corresponding to the reference reimbursement standard, but the item value of the non-expense type reimbursement item is inconsistent with the reference item value, the abnormal reimbursement item corresponding to the approval abnormality is determined. Then, it is identified whether the abnormal reimbursement item is a specific reimbursement item. The specific reimbursement item is generally a reimbursement item corresponding to a means of transportation, and its corresponding item value includes railways, airplanes, etc. For example, the target reimbursement document of the target user shows that the means of transportation to the first-tier city of Beijing is airplane, and the target document approval standard shows that the means of transportation allowed for the target user to travel to the background first-tier city is railway. Then, the item value (airplane) corresponding to the reimbursement item (transportation) in the target reimbursement document is inconsistent with the reference item value (railway) corresponding to the corresponding reimbursement item (transportation) in the target document approval standard, and there is an abnormal reimbursement item with abnormal transportation, and the abnormal reimbursement item is a specific reimbursement item.

[0105] When the abnormal reimbursement item is not a specific reimbursement item, the approval result corresponding to the target reimbursement document is determined to be approval failure, the reimbursement operation of the target reimbursement document is stopped, and an alarm prompt is generated. When the abnormal reimbursement item is a specific reimbursement item, the current reimbursement cost of the abnormal reimbursement item in the current item value is obtained from the target reimbursement document, such as the reimbursement cost of the air ticket of the target reimbursement document with the transportation tool being an airplane is 400, and the reference item value of the abnormal reimbursement item is determined from the approval standard of the target document, such as the reference item value of the abnormal reimbursement item with the transportation tool being the railway. Then, the actual cost of the abnormal reimbursement item in the reference item value is queried, which can be determined based on the traffic information corresponding to the abnormal reimbursement item in the target reimbursement document. The abnormal trip refers to the trip corresponding to the traffic information of the abnormal reimbursement item, and the actual cost of the reference item value in the abnormal trip is queried, such as the high-speed rail cost of the same trip is queried through the official website. When the current reimbursement cost does not exceed the actual cost, the approval result of the abnormal reimbursement item is updated to approved. It can be understood that the permitted means of transportation is second-class railway seats, but when traveling on business, the airfare for the same itinerary is lower than the price of second-class railway seats, which can be judged to meet the standards.

[0106] In this embodiment, by verifying the actual expenses according to the approval exceptions of the specific reimbursement items, the approval flexibility of the reimbursement documents is reflected and the approval efficiency of the reimbursement documents is improved.

[0107] In a specific embodiment, Figure 4 As shown, a schematic diagram of the processing flow of consistent reimbursement documents is provided. The method for generating reimbursement documents can be applied to a document reimbursement device, and the document reimbursement device can process reimbursement documents with the business type of travel reimbursement. The document reimbursement device includes a document generation sub-device, a travel standard sub-device, and a document verification sub-device, wherein:

[0108] The document generation sub-device is used to generate reimbursement documents, and its execution steps are as follows: extract the invoice key information such as the invoice date, invoice location, departure place, destination, invoice type, project name, amount, tax rate, and tax amount from the collected reimbursement invoices, such as invoice 1, invoice 2, ... invoice N, through LLM (large language model); sort the invoice key information in chronological order according to the invoice date of the invoice key information, and generate the document detailed information. Among them, the decision tree model is used to output the objective fields such as travel type and expense items that need to be determined in the document detailed information, and the LLM model such as ChatGPT is used to generate the subjective fields such as description and remarks required in the document detailed information.

[0109] Then, each itinerary information is determined based on the generated document detail information and the time displacement information reflected in the cross-regional transportation invoices (such as air tickets, train tickets, ship tickets, passenger tickets, etc.) in the document detail information. It can also be determined based on the invoice date and location change information of non-transportation invoices; the itinerary detail information corresponding to each itinerary information is obtained, the itinerary detail information is incorporated into the corresponding itinerary information to form a parent-child data structure, and a reimbursement document is generated based on the formed parent-child data structure, and the reimbursement document is stored in the database.

[0110] The travel standard sub-device is used to generate travel standard data (i.e., target document approval standards) based on other external reference reimbursement documents (including reimbursement standard data, reimbursement standard documents, Word, Excel, etc. authorized for use by other similar customers in the system) and the historical reimbursement document set of the user set. The travel standard data is used to approve, verify and issue early warnings for the reimbursement documents generated by the document generation sub-device. When, during the approval process of a reimbursement document, the approval result of the document verification device is inconsistent with the approval result given by manual confirmation, the data of this reimbursement document will be included in the training set of the travel standard sub-device, and the early warning line in the travel standard data will be fine-tuned through the RNN neural network algorithm.

[0111] The document verification sub-device is used to review and approve the reimbursement documents generated by the document generation sub-device and to issue warnings. The verification or warning level can be set according to the actual situation, such as: termination (prohibit initiation), strong reminder (allow initiation but the approval process requires secondary confirmation), weak reminder (allow initiation without secondary confirmation). The verification steps include:

[0112] Itinerary matching verification is used to verify whether the itinerary timeline is complete (abnormal time consumption monitoring and abnormal itinerary consumption monitoring), whether there is consumption details within the transportation travel time, such as the existence of out-of-town invoices and abnormal details that are not from the current itinerary location, and whether the start and end time and location between the itineraries are consistent with the consumption details; whether there is detailed information about non-current itinerary locations within the itinerary (unconventional consumption monitoring), such as abnormal consumption frequency, etc.

[0113] Excessive standard verification, including excessive standard verification of travel standards and excessive standard verification of transportation tools:

[0114] The travel standard over-limit check is used to determine whether the relevant expense items in the document details exceed the travel standard. If it exceeds the warning line, additional explanation is required. If it exceeds the standard but does not exceed the warning line, a strong reminder is given.

[0115] The over-standard check of means of transportation is used to determine whether the means of transportation in the document details exceeds the standard. If so, it is necessary to determine whether the expense limit exceeds the standard. If not, a weak reminder is given. Based on the attention paid to reminders during the approval process, the travel standard device fine-tunes the reminder level. For example, if the permitted means of transportation is a second-class railway seat, but the airfare for the same itinerary during a business trip is lower than that of a second-class railway seat, it can be judged to meet the standard and a weak reminder will be issued initially. After similar reimbursement documents with abnormal transportation approval have been reviewed and approved multiple times, they will be directly marked as meeting the standard and no reminder will be issued. The default is compliant reimbursement data.

[0116] In this embodiment, the document generation sub-device can reduce the number of human-computer interactions, improve the automation rate and user experience of invoice-to-document, advance the document verification to the background after the document is saved, rationally use the device resources, and improve the efficiency of document verification and review. Intelligent verification and early warning are realized through the document verification sub-device, high risks are controlled before the document is approved, and early warning and auxiliary decision-making are carried out throughout the review process to reduce the risk of corporate reimbursement. Through the in-depth analysis of the historical documents of the reimbursement person through machine learning technology, the travel standards of the reimbursement person are dynamically adjusted, making management more humane under the premise of controllable risks, and without manual preset or adjustment. The entire device has sufficient self-evolution capabilities to better fit the usage scenarios that change at any time.

[0117] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0118] Based on the same inventive concept, the embodiment of the present application also provides a reimbursement document generation device for implementing the reimbursement document generation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more reimbursement document generation device embodiments provided below can refer to the limitations of the reimbursement document generation method above, and will not be repeated here.

[0119] In one embodiment, Figure 5As shown, a reimbursement document generation device 500 is provided, comprising: an invoice extraction module 502, a detail generation module 504, an itinerary generation module 506 and a document generation module 508, wherein:

[0120] The invoice extraction module 502 is used to obtain each reimbursement invoice corresponding to the target user and extract the key invoice information corresponding to each reimbursement invoice;

[0121] A detail generation module 504 is used to generate document detail information based on the invoice key information;

[0122] The itinerary generation module 506 is used to generate each itinerary information based on the time sequence displacement information in the document detail information, and extract the itinerary detail information corresponding to each itinerary information from the document detail information;

[0123] The document generation module 508 is used to obtain the target reimbursement document corresponding to the target user based on each itinerary information and itinerary detail information.

[0124] In one embodiment, the detail generation module 504 is also used to input the invoice key information into the classification model to obtain the objective reimbursement field determined from the candidate objective fields; obtain the historical reimbursement documents corresponding to the target user, input the invoice key information and the historical reimbursement documents into the generation model to obtain the subjective reimbursement field; based on the objective reimbursement field and the subjective reimbursement field, obtain the document detail information.

[0125] In one embodiment, the itinerary generation module 506 is also used to obtain time displacement information based on the date information and traffic information in the document detail information; obtain each itinerary information based on the departure place and destination corresponding to the target traffic information in the time displacement information; obtain the sub-document detail information generated at the destination in the itinerary information from the document detail information as the itinerary detail information corresponding to the itinerary information, and obtain the itinerary detail information corresponding to each itinerary information.

[0126] In one embodiment, the reimbursement document generation device 500 is also used to obtain a reference reimbursement file, and determine an initial document approval standard based on the reference reimbursement file; the initial document approval standard includes reference reimbursement standards corresponding to multiple reference reimbursement items; determine a target reimbursement item from the reference reimbursement items; obtain a historical reimbursement document set corresponding to a user set, and determine a reimbursement warning value corresponding to the target reimbursement item based on the historical reimbursement document set and the reference reimbursement standard corresponding to the target reimbursement item; add the reimbursement warning value corresponding to the target reimbursement item to the initial document approval standard to obtain the target document approval standard, and the target document approval standard is used to approve the target reimbursement document.

[0127] The reimbursement document generating device 500 is also used to determine each travel time and the corresponding travel location based on the date information and traffic information in the document detail information; based on the document detail information, when there is abnormal consumption detail information in each travel time, an abnormal time consumption alarm is generated; within a preset time period after the travel time, when there is out-of-town consumption detail information that is different from the destination in the travel location, an abnormal travel consumption alarm is generated.

[0128] In one embodiment, the reimbursement warning value is obtained by inputting the reference reimbursement standards corresponding to the historical reimbursement document set and the target reimbursement item into the reimbursement standard analysis model, and based on the output of the reimbursement standard analysis model; the reimbursement document generation device 500 is also used to obtain the manual approval result corresponding to the target reimbursement document when the target reimbursement document fails to be approved; the target reimbursement document and the manual approval result are used as new training samples, and the new training samples are added to the training sample set; based on the training sample set, the reimbursement standard analysis model is optimized.

[0129] In one embodiment, the reimbursement document generating device 500 is also used to obtain the abnormal reimbursement items with approval exceptions and the current reimbursement expenses of the abnormal reimbursement items in the current project value from the target reimbursement document; determine the reference project value of the abnormal reimbursement items from the target document approval standard, and query the actual expenses of the abnormal reimbursement items in the reference project value; when the current reimbursement expenses do not exceed the actual expenses, update the approval result of the abnormal reimbursement item to approved.

[0130] Each module in the above-mentioned reimbursement document generation device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0131] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as various reimbursement invoices. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for generating a reimbursement document is implemented.

[0132] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for generating a reimbursement document is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0133] Those skilled in the art will understand that Figure 6-Figure 7The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0134] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0135] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0136] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0137] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0138] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0139] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0140] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for generating a reimbursement document, characterized in that: The method comprises: Obtain each reimbursement invoice corresponding to the target user, and extract the key invoice information corresponding to each reimbursement invoice; Generate document detail information based on the invoice key information; Generate each trip information based on the time series displacement information in the document detail information, and extract the trip detail information corresponding to each trip information from the document detail information; Based on the various itinerary information and the itinerary detail information, a target reimbursement document corresponding to the target user is obtained.

2. The method according to claim 1, characterized in that The generating of document detail information based on the invoice key information includes: Inputting the invoice key information into a classification model to obtain an objective reimbursement field determined from candidate objective fields; Obtaining the historical reimbursement documents corresponding to the target user, inputting the key invoice information and the historical reimbursement documents into the generative model to obtain a subjective reimbursement field; Based on the objective reimbursement field and the subjective reimbursement field, document detail information is obtained.

3. The method according to claim 1, characterized in that The generating each piece of itinerary information based on the time series displacement information in the document detail information, and extracting the itinerary detail information corresponding to each piece of itinerary information from the document detail information, comprises: Based on the date information and traffic information in the document detail information, obtaining time series displacement information; Based on the departure place and destination corresponding to the target traffic information in the time series displacement information, obtain each trip information; The sub-document detail information generated at the destination in the itinerary information is obtained from the document detail information as the itinerary detail information corresponding to the itinerary information, and the itinerary detail information corresponding to each of the itinerary information is obtained.

4. The method according to claim 3, characterized in that The method further comprises: Determine each transportation journey time and corresponding transportation journey location based on the date information and transportation information in the document details; generate an abnormal time consumption alarm when there is abnormal consumption detail information within each transportation journey time based on the document details; or When there is off-site consumption detail information that is different from the destination in the transportation itinerary location within a preset time period after the transportation itinerary time, an abnormal itinerary consumption alarm is generated.

5. The method according to claim 1, characterized in that The method further comprises: Obtaining a reference reimbursement file, and determining an initial document approval standard based on the reference reimbursement file; the initial document approval standard includes reference reimbursement standards corresponding to a plurality of reference reimbursement items respectively; determining a target reimbursement item from the reference reimbursement items; Acquire a historical reimbursement document set corresponding to the user set, and determine a reimbursement warning value corresponding to the target reimbursement item based on the historical reimbursement document set and a reference reimbursement standard corresponding to the target reimbursement item; The reimbursement warning value corresponding to the target reimbursement item is added to the initial document approval standard to obtain a target document approval standard, and the target document approval standard is used to approve the target reimbursement document.

6. The method according to claim 5, characterized in that The reimbursement warning value is obtained by inputting the historical reimbursement document set and the reference reimbursement standard corresponding to the target reimbursement item into the reimbursement standard analysis model, and based on the output of the reimbursement standard analysis model; The method further comprises: When the target reimbursement document fails to be approved, obtaining the manual approval result corresponding to the target reimbursement document; Taking the target reimbursement document and the manual approval result as new training samples, and adding the new training samples to the training sample set; Based on the training sample set, the reimbursement standard analysis model is optimized.

7. The method according to claim 1, characterized in that The method further comprises: Acquire, from the target reimbursement document, the abnormal reimbursement item with abnormal approval and the current reimbursement expense of the abnormal reimbursement item at the current item value; Determine the reference item value of the abnormal reimbursement item from the target document approval standard, and query the actual cost of the abnormal reimbursement item at the reference item value; When the current reimbursement expense does not exceed the actual expense, the approval result of the abnormal reimbursement item is updated to approved.

8. A reimbursement document generating device, characterized in that: The device comprises: An invoice extraction module is used to obtain each reimbursement invoice corresponding to the target user and extract the key invoice information corresponding to each reimbursement invoice; A detail generation module, used to generate document detail information based on the invoice key information; A trip generation module, used to generate each trip information based on the time sequence displacement information in the document detail information, and extract the trip detail information corresponding to each trip information from the document detail information; The document generation module is used to obtain the target reimbursement document corresponding to the target user based on the various itinerary information and the itinerary detail information.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images