Information processing system, information processing method, and information processing program

Abstract

We provide an information processing system that allows expense reimbursement to be performed with less workload. [Solution] The settlement information acquisition unit sends a prompt to the generation device that includes target information to be processed, subject information having multiple subject descriptions, a subject description inference instruction that causes the unit to infer which of the multiple subject descriptions the target information corresponds to, and a settlement information generation instruction that generates settlement information for performing expense settlement from the target information based on the subject descriptions, and acquires settlement information from the generation device. When the question unit acquires a question comment from the generation device, it displays the question comment acquired from the generation device and accepts a response from the user, and sends a prompt to the generation device that includes the response and an additional generation instruction that generates and outputs settlement information with the response as additional information.

Description

[Technical Field]

[0001] The present invention relates to an information processing system, an information processing method, and an information processing program. [Background Art]

[0002] Conventionally, in organizations such as companies, there exist information processing systems for supporting expense reimbursement. For example, Japanese Patent No. 7653741 discloses a detection device including: an acquisition unit that acquires application data representing the content of an application to a company, the application data including a plurality of application items; a detection unit that detects, for a plurality of detection items, that the application data includes an application that does not conform to a criterion; and a providing unit that determines whether there is any application item with missing information in the application data, and provides information related to undetectable detection items when detection cannot be performed for some of the plurality of detection items due to the missing information. [Prior Art Documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent No. 7653741 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] However, with conventional information processing systems, an accounting staff member who receives an expense reimbursement application must check whether the expense reimbursement application conforms to the criteria. Furthermore, when there is missing information, the accounting staff member must make an inquiry to the applicant, which imposes a large work load on the accounting staff member.

[0005] Accordingly, an object of embodiments of the present invention is to provide an information processing system, an information processing method, and an information processing program that enable expense reimbursement with a reduced work load. [Means for Solving the Problem]

[0006] The information processing system according to this embodiment includes a settlement information acquisition unit and a questioning unit. The settlement information acquisition unit includes target information to be processed, subject information having multiple subject descriptions, a subject description inference instruction that infers which of the multiple subject descriptions the target information corresponds to, and a unit for performing expense settlement from the target information based on the subject descriptions. Includes the subject, date, and amount corresponding to the above subject description. Instruction to generate settlement information. Furthermore, if it is inferred that the information instructed to be included in the settlement information in the instruction statement included in the subject description is not included in the target information, a question output instruction will be generated and output to ask the user for the missing information that is not included in the target information. A prompt including the settlement information is sent to the generating device, and the generating device receives the settlement information , or the above question comments The questioning unit obtains from the generating device. The aforementioned When a question comment is obtained, the question comment obtained from the generator is displayed. The aforementioned The system receives a response from the user and transmits the response, along with a prompt to the generating device that includes an additional generation instruction to generate and output the settlement information using the response as additional information. The aforementioned target information is text information that is subject to expense reimbursement processing, and the aforementioned text information is information based on document images or text received from the user.

[0007] The information processing method according to this embodiment includes, by the settlement information acquisition unit, target information to be processed, subject information having multiple subject descriptions, a subject description inference instruction that causes the unit to infer which of the multiple subject descriptions the target information corresponds to, and a method for performing expense settlement from the target information based on the subject descriptions. Includes the subject, date, and amount corresponding to the above subject description. Instruction to generate settlement information. Furthermore, if it is inferred that the information instructed to be included in the settlement information in the instruction statement included in the subject description is not included in the target information, a question output instruction will be generated and output to ask the user for the missing information that is not included in the target information. A prompt including the settlement information is sent to the generating device, and the generating device receives the settlement information , or the above question comments The questioning unit obtains the information from the generating device. The aforementioned When a question comment is obtained, the question comment obtained from the generator is displayed. The aforementioned The system receives a response from the user and transmits the response, along with a prompt to the generating device that includes an additional generation instruction to generate and output the settlement information using the response as additional information. The aforementioned target information is text information that is subject to expense reimbursement processing, and the aforementioned text information is information based on document images or text received from the user.

[0008] The information processing program according to this embodiment causes the computer to execute the code of the settlement information acquisition unit and the code of the questioning unit. The code of the settlement information acquisition unit includes target information to be processed, subject information having multiple subject descriptions, a subject description inference instruction that infers which of the multiple subject descriptions the target information corresponds to, and a method for performing expense settlement from the target information based on the subject descriptions. Includes the subject, date, and amount corresponding to the above subject description. Instruction to generate settlement information. Furthermore, if it is inferred that the information instructed to be included in the settlement information in the instruction statement included in the subject description is not included in the target information, a question output instruction will be generated and output to ask the user for the missing information that is not included in the target information. A prompt including the settlement information is sent to the generating device, and the generating device receives the settlement information , or the above question comments Obtain the following. The code of the question section is obtained from the generator. The aforementioned When a question comment is obtained, the question comment obtained from the generator is displayed. The aforementioned The system receives a response from the user and transmits the response, along with a prompt to the generating device that includes an additional generation instruction to generate and output the settlement information using the response as additional information. The aforementioned target information is text information that is subject to expense reimbursement processing, and the aforementioned text information is information based on document images or text received from the user. [Effects of the Invention]

[0009] According to embodiments of the present invention, it is possible to provide an information processing system, an information processing method, and an information processing program that enable expense settlement with less workload. [Brief explanation of the drawing]

[0010] [Figure 1] This is a block diagram showing an example of the overall configuration of an information processing system according to the embodiment. [Figure 2] This is a block diagram showing an example of the internal configuration of an information processing system according to the embodiment. [Figure 3] This is a block diagram showing an example of the functional configuration of an information processing system according to the embodiment. [Figure 4] This figure shows an example of subject information in the information processing system according to the embodiment. [Figure 5] This figure shows an example of a document image that is instructed to be input to the target information acquisition unit in the information processing system according to the embodiment. [Figure 6] FIG. 1 is a diagram showing an example of an expense settlement screen displayed on a display unit in the information processing system according to the embodiment. [Figure 7] FIG. 2 is a diagram showing an example of target information extracted by a target information extraction unit in the information processing system according to the embodiment. [Figure 8] FIG. 3 is a flowchart showing an example of the flow of expense settlement processing in the information processing system according to the embodiment. [Figure 9] FIG. 4 is a flowchart showing an example of the flow of expense settlement processing in the information processing system according to the embodiment. [Figure 10] FIG. 5 is a flowchart showing an example of the flow of expense settlement processing for a plurality of pieces of target information in the information processing system according to the embodiment. MODE FOR CARRYING OUT THE INVENTION

[0011] (Configuration of Embodiment) Hereinafter, embodiments will be described with reference to the drawings. FIG. 1 is a block diagram showing an example of the overall configuration of an information processing system 1.

[0012] As shown in FIG. 1, an information processing system 1 includes a terminal device 10 and a central device 20. The terminal device 10 is connected to the central device 20 and a business device B via a network such as a LAN, the Internet, or mobile communication. The central device 20 is connected to each of the business device B and a generation device M via a network.

[0013] FIG. 2 is a block diagram showing an example of the internal configuration of the information processing system 1. FIG. 3 is a block diagram showing an example of the functional configuration of the information processing system 1. In FIG. 2 and FIG. 3, connection lines that directly connect the terminal device 10 and the business device B are omitted.

[0014] As shown in Figures 2 and 3, business device B is a server and is capable of performing business processes such as accounting. Business device B stores programs for the pre-application group As, account information K, and settlement processing unit Bp. The programs for the pre-application group As, account information K, and settlement processing unit Bp are used for business processes performed in response to user input via terminal device 10, as well as for expense settlement processing performed in response to user input via terminal device 10 and central device 20.

[0015] The pre-application group As includes one or more pre-applications. A pre-application may include the user's identification information, date, and amount, as well as title, participants, purpose, etc. The information included in a pre-application is not limited to this and may include other information. Before expense reimbursement, the applicant submits the pre-application to the business device B via terminal device 10 and central device 20, and the application is stored there. The administrator can view and approve the pre-applications stored in business device B via terminal device 10.

[0016] Figure 4 shows an example of subject information K.

[0017] As shown in Figure 4, the account information K includes the corresponding account K1, tax classification K2, and account description K3. The business device B can display the account information K on the display unit 13, and in response to user instructions via the instruction input unit 14, it is possible to view, create, edit, and delete the account information K.

[0018] Subject K1 contains the account name used for expense reimbursement. Subject K1 may include keywords with symbols such as "#" to improve searchability and visibility.

[0019] Tax category K2 is information used for calculating consumption tax in accounting procedures.

[0020] Subject Description K3 is information that describes Subject K1. Subject Description K3 can be viewed by the user and is also input into the generation device M as part of a prompt, and is used in the subject description inference process and settlement information generation process described later. Subject Description K3 can include not only a descriptive text that describes Subject K1, but also instruction text that instructs the processing content in the subject description inference process and settlement information generation process.

[0021] The settlement processing unit Bp is capable of performing settlement flow processing. More specifically, the settlement processing unit Bp receives instruction inputs from the user for settlement input information such as date, amount, application title, document image, account, content information, and tax classification. Upon receiving instruction inputs for settlement input information, the settlement processing unit Bp starts settlement flow processing and puts the settlement input information into a state awaiting approval. Once the administrator approves, the settlement processing unit Bp puts the settlement input information into a state awaiting transfer, and once the administrator completes the transfer process, it registers journal entry information and performs accounting processing related to expense settlement.

[0022] The settlement processing unit Bp can execute settlement flow processing based on settlement input information received from the user, and can also execute settlement flow processing based on settlement information received from the central device 20.

[0023] The generation device M is a server with a large-scale language model. The large-scale language model can be, for example, Claude, ChatGPT, or Gemini, and is constructed by learning from a large amount of data. It is capable of analyzing natural language prompts and generating and outputting various types of information based on the prompts. The generation device M includes a text reading unit M1, a pre-application extraction unit M2, and a settlement information generation unit M3.

[0024] The text reading unit M1 reads text from image information and outputs the text information. The text information includes the text read from the image information and the label corresponding to the text. The text reading unit M1 may be implemented using functions provided by the Cloud Vision API or Document AI.

[0025] For example, when the text reading unit M1 reads the image information of a receipt, it outputs text information that includes the issuer, issuer's address, issuer's phone number, date, and amount. The text information is not limited to these; if it is a receipt for transportation expenses, it will also include information that can be read from the image information, such as the boarding station and alighting station.

[0026] Furthermore, if the image information contains multiple settlement targets, the text reading unit M1 outputs text information containing multiple target information items to be processed. More specifically, if the image information is a table containing multiple settlement targets, the text reading unit M1 divides the read text into cells, assigns labels corresponding to the text, and outputs text information containing multiple target information items. The text information may be in JSON format, for example.

[0027] The pre-application extraction unit M2 extracts pre-applications corresponding to the target information from the pre-application group As and outputs the extracted pre-applications.

[0028] Figure 5 shows an example of a document image that is instructed to be input to the target information acquisition unit P2.

[0029] The settlement information generation unit M3 performs a subject description inference process to infer which of the multiple subject descriptions K3 the target information corresponds to, and then performs a settlement information generation process to generate settlement information for expense settlement from the target information based on the subject description K3, thereby generating settlement information for expense settlement. The subject description inference process and the settlement information generation process can also be performed based on the target information and a pre-application corresponding to the target information. If there is any missing information that is insufficient to generate the settlement information instructed by the subject description K3, the settlement information generation unit M3 generates and outputs a question comment asking the user for the missing information.

[0030] More specifically, the settlement information generation unit M3 first performs account description inference processing. The settlement information generation unit M3 analyzes each of the account descriptions K3 included in the account information K and infers which account description K3 the target information corresponds to. As shown in Figure 5 as an example, if the target information is one of the electronic money usage histories showing transportation expenses, the settlement information generation unit M3 infers that the target information corresponds to "Please use this for application for payment of transportation" in account description K3 because it includes station names and amounts equivalent to small train fares.

[0031] Next, the settlement information generation unit M3 performs settlement information generation processing. Based on the instruction "Please include the purpose of travel (destination or reason for travel) and route section in the content" contained in the subject description K3, which was inferred to be suitable by the subject description inference processing, the settlement information generation unit M3 infers that the route section is included because the target information includes station names. Also, based on this instruction, if the settlement information generation unit M3 infers that the reason for travel is not included in the target information and the pre-application, it generates and outputs a question comment asking for the missing information, the reason for travel. On the other hand, if the settlement information generation unit M3 infers that there is no missing information based on the target information and the pre-application, it generates and outputs settlement information according to the comment contained in the subject description K3.

[0032] Returning to Figures 2 and 3, the terminal device 10 is, for example, an information terminal such as a personal computer or a smartphone. The terminal device 10 is used by users in organizations such as companies. Users include not only applicants who submit pre-approval requests or expense reimbursements, but also administrators who approve the applications submitted by applicants and process the approved applications for accounting purposes. The terminal device 10 has a terminal control unit 11, a terminal storage unit 12, a display unit 13, and an instruction input unit 14.

[0033] The terminal control unit 11 has a processing unit such as a CPU and realizes its function by executing a program read from the terminal storage unit 12.

[0034] The terminal storage unit 12 has memory elements such as ROM, RAM, HDD, and SSD, and stores various information for controlling the terminal device 10, as well as various programs that constitute the processing unit.

[0035] The display unit 13 is composed of an LCD, LED, OLED, etc., and is capable of displaying various types of information input from the terminal control unit 11.

[0036] The instruction input unit 14 has, for example, an input device such as a keyboard or mouse, and outputs various instruction inputs entered by the user to the terminal control unit 11.

[0037] The central device 20 is a server and includes a control unit 21 and a storage unit 22.

[0038] The control unit 21 has a processing unit such as a CPU and realizes its function by executing a program read from the storage unit 22.

[0039] The storage unit 22 has memory elements such as ROM, RAM, HDD, and SSD, and stores various programs that constitute the processing unit for controlling the central device 20, as well as various information used by the programs. For example, the storage unit 22 stores various programs such as the pre-application registration unit P1, the target information acquisition unit P2, the target information extraction unit P3, the pre-application acquisition unit P4, the account acquisition unit P5, the settlement information acquisition unit P6, the questioning unit P7, the amount determination unit P8, and the application unit P9.

[0040] The pre-application registration unit P1 receives pre-applications containing a date and amount from the user (applicant) via the terminal device 10 and registers them in the pre-application group As.

[0041] Figure 6 shows an example of the expense reimbursement screen G displayed on the display unit 13. Figure 7 shows an example of the target information extracted by the target information extraction unit P3.

[0042] The target information acquisition unit P2 acquires the target information to be processed and stores it in the storage unit 22. As shown in Figure 5, the target information acquisition unit P2 displays the expense settlement screen G on the display unit 13, receives instruction input from the user for document information consisting of document images, transmits the received document information to the text reading unit M1 of the generation device M, and acquires target information consisting of text information from the text reading unit M1. The target information acquisition unit P2 can also accept not only document images, but also documents composed of text, or instruction input from the user consisting only of text, and use them as target information.

[0043] When multiple target information items are acquired by the target information acquisition unit P2, the target information extraction unit P3 receives input from the user for selecting target information, extracts the target information to be processed selected by the user from among the multiple target information items, and stores it in the storage unit 22. The user can also select and input multiple target information items. For example, if the target information acquisition unit P2 reads a document image containing multiple target information items, such as the electronic money usage history shown in Figure 5, and acquires multiple target information items, the target information extraction unit P3 displays a selection screen showing a list of target information items, receives input from the user for selecting target information items, and extracts one or more target information items as shown in Figure 7.

[0044] The pre-application acquisition unit P4 extracts pre-applications corresponding to the target information from the pre-application group As in which pre-applications have been registered. More specifically, the pre-application acquisition unit P4 acquires the pre-application group As that the user has submitted in advance from the business device B based on the user's identification information. Subsequently, the pre-application acquisition unit P4 sends a prompt to the pre-application extraction unit M2 of the generation device M, which includes the target information, the pre-application group As, and a pre-application extraction instruction, and retrieves the extracted pre-applications from the pre-application extraction unit M2 and stores them in the storage unit 22. The pre-application extraction instruction is an instruction to extract pre-applications that are similar to the date and amount included in the target information from among the pre-applications included in the pre-application group As, and to extract them as extracted pre-applications. On the other hand, if an extracted pre-application could not be obtained, the pre-application acquisition unit P4 stores in the storage unit 22 that an extracted pre-application could not be obtained.

[0045] The subject acquisition unit P5 acquires subject information K from the business device B and stores it in the storage unit 22.

[0046] The settlement information acquisition unit P6 transmits a prompt to the settlement information generation unit M3 of the generation device M, which includes the target information to be processed, the account information K having multiple account descriptions K3, an account description inference instruction, and a settlement information generation instruction, and acquires settlement information from the generation device M. The account description inference instruction is an instruction to infer which of the multiple account descriptions K3 the target information corresponds to. The settlement information generation instruction is an instruction to generate settlement information for expense settlement from the target information based on the instruction in the account description K3.

[0047] Furthermore, if an extracted pre-application has been obtained by the pre-application acquisition unit P4, the settlement information acquisition unit P6 sends a prompt to the settlement information generation unit M3 of the generation device M, which includes a settlement information generation instruction to generate settlement information by inferring that includes the target information and the extracted pre-application, and obtains settlement information from the generation device M.

[0048] More specifically, the instruction to generate settlement information includes instructions to include information based on the target information, prior application, and account information K in the settlement information. For example, the settlement information may include the date and amount required for expense settlement processing, as well as at least one of the following: expense name, participant, purpose, issuer, registration number, or purpose. The settlement information is not limited to this and may include other information.

[0049] Furthermore, if the settlement information acquisition unit P6 has missing information necessary to generate the settlement information instructed in the account description K3, it includes a question output instruction in the prompt to generate and output a question comment asking the user for the missing information, and sends it to the generation device M.

[0050] When the question unit P7 obtains a question comment from the generation device M, it displays the question comment obtained from the generation device M, accepts a response from the user, and sends a prompt to the generation device M that includes the response and an additional generation instruction to generate and output settlement information using the response as additional information. More specifically, when the settlement information acquisition unit P6 obtains a question comment, it displays the question comment on the expense settlement screen G in the display unit 13 and provides a text box to accept a response from the user. When the user enters a response in the text box, the settlement information acquisition unit P6 sends a prompt to the settlement information generation unit M3 of the generation device M that includes the response received from the user and an additional generation instruction to generate settlement information.

[0051] The amount determination unit P8 outputs a determination result indicating whether the amount included in the settlement information is a fair amount suitable for expense reimbursement, based on the amount registered in the pre-application and under predetermined conditions. The predetermined conditions are, for example, a condition in which the amount included in the settlement information is determined to be a fair amount if it is less than or equal to the amount included in the pre-application. The predetermined conditions are not limited to this condition; the amount may be set to a range, and the amount included in the settlement information may be considered a fair amount if it does not exceed a predetermined range from the amount included in the pre-application, or other conditions may be used. If the amount determination unit P8 determines that the amount is not fair, it displays on the expense reimbursement screen G displayed on the display unit 13 that the amount subject to expense reimbursement is not fair, prompting the user to confirm.

[0052] If the settlement information acquisition unit P6 has successfully acquired the settlement information, the application unit P9 will execute the settlement flow processing based on the settlement information. More specifically, the application unit P9 generates content information by summarizing the expense name, participant, purpose, issuer, registration number, or purpose etc. included in the settlement information into text. The application unit P9 sends the content information, the date, amount, application title included in the settlement information, the account K1 and tax category K2 included in the account information K, and the document image to the settlement processing unit Bp of the business device B, and executes the settlement flow processing.

[0053] Next, we will explain the expense reimbursement process in Information Processing System 1.

[0054] Figures 8 to 10 are flowcharts illustrating an example of the expense reimbursement process.

[0055] As shown in Figures 8 and 9, when a user inputs a pre-application via the terminal device 10, the pre-application registration unit P1 registers the pre-application in the pre-application group As (S1). The pre-application includes, for example, an amount of "16,000 yen" and an application date of "YYYY / MM / DD", as well as the user's identification information, title, department, and department.

[0056] When the user scans the receipt, the target information acquisition unit P2 sends an image of the receipt document to the text reading unit M1, and the text reading unit M1 acquires the target information consisting of text data (S2). Once the acquisition of the target information is complete, the target information acquisition unit P2 displays a message on the expense settlement screen G in the display unit 13 indicating that the scan has been completed. In the example in Figure 6, the target information includes the amount "14,688 yen", the issue date "YYYY / MM / DD", as well as the issuer, registration number, and purchased items.

[0057] The pre-application acquisition unit P4 acquires the pre-application group As submitted by the user in advance from the business device B based on the user's identification information (S3). The pre-application acquisition unit P4 sends a prompt to the pre-application extraction unit M2 including the target information, the pre-application group As, and a pre-application extraction instruction, instructing it to extract pre-applications (S4). The pre-application extraction unit M2 extracts pre-applications with similar dates and amounts from the pre-application group As submitted by the user in advance, and outputs the extracted pre-applications (S5). The extracted pre-applications are acquired from the pre-application extraction unit M2, and the expense settlement screen G on the display unit 13 displays a message indicating that the pre-applications have been matched and the content inferred based on the pre-applications (S6). The expense settlement screen G displays an "Interrupt and Correct" button. If the extracted pre-applications have not been acquired correctly, the user can operate the button to interrupt the process and have the image information read again.

[0058] The subject acquisition unit P5 acquires subject information K from the business device B (S7).

[0059] The settlement information acquisition unit P6 sends a prompt to the settlement information generation unit M3 that includes the target information, pre-application, subject information K, subject description inference instruction, and settlement information generation instruction (S8).

[0060] If there is any missing information needed to generate the settlement information (S9:YES), the settlement information generation unit M3 generates and outputs a question comment (S10).

[0061] The question unit P7 displays a question comment on the expense settlement screen G in the display unit 13 and accepts the user's input of an answer (S11). When the user inputs an answer, the question unit P7 sends the answer and additional generation instructions to the settlement information generation unit M3 (S12) and returns to S9.

[0062] If there is no missing information (S9:NO), the settlement information generation unit M3 generates settlement information and displays it on the expense settlement screen G in the display unit 13 (S13). In the example in Figure 6, the settlement information includes the transaction date based on the receipt issuance date, the amount based on the receipt, the title based on the prior application, the purchased items based on the receipt, etc.

[0063] The amount determination unit P8 determines, based on the amount registered in the pre-application, whether the amount included in the settlement information is a fair amount under predetermined conditions. If it is a fair amount (S14: YES), the application unit P9 sends the settlement information to the settlement processing unit Bp to execute the settlement flow processing (S15).

[0064] On the other hand, if the amount included in the settlement information is not the correct amount (S14: NO), a message indicating that the amount is not correct will be displayed on the expense settlement screen G (S16), and the process will be terminated.

[0065] Next, we will explain the expense reimbursement process when the document information contains multiple pieces of relevant information. Here, as an example, we will explain the case where the document information is an electronic money usage history (Figure 5).

[0066] When a user has the terminal device 10 read the electronic money usage history, the target information acquisition unit P2 sends an image of the electronic money usage history to the text reading unit M1 and acquires multiple pieces of target information from the text reading unit M1 (S21).

[0067] The target information extraction unit P3 displays multiple target information on the display unit 13, accepts selection input of target information from the user, and extracts one or more target information selected by the user (S22).

[0068] If multiple pieces of target information are extracted, the process is repeated, processing each piece of target information one by one. In the repeated processing, the pre-application corresponding to the target information is obtained (S23), the account information K is obtained (S24), if there is any missing information, the user's response is accepted to generate and obtain settlement information (S25), a process is performed to determine if the amount in the settlement information is an appropriate amount (S26), the settlement information is sent to the settlement processing unit Bp to execute the settlement flow process (S27), and the process is repeated back to S23 until all pieces of target information have been processed (S28).

[0069] Each of the processes in S1 to S14 and the processes in S21 to S28 constitutes the expense reimbursement process.

[0070] As a result, when the user scans a document image, the information processing system 1 displays a question comment if there is any missing information, prompts the user to answer, generates settlement information, and allows the settlement processing unit Bp to seamlessly execute the settlement flow processing, enabling expense settlement with less workload.

[0071] Furthermore, the subject information K can be viewed, created, edited, and deleted by the user. Users can easily edit the subject description K3 included in the subject information K, and adjust the subject description inference process and settlement information generation process.

[0072] Furthermore, since settlement information is generated based on the target information and the corresponding pre-application, when inputting the target information, it is not necessary to re-enter information that overlaps with the pre-application, thereby reducing the workload.

[0073] Furthermore, the amount determination unit P8 can display a message to the user indicating that the amount included in the settlement information is not the correct amount, thereby eliminating incorrect expense settlements without increasing the workload of the administrator.

[0074] The target information extraction unit P3 extracts the target information to be processed from among multiple target information items. This makes it possible to process expense reimbursements using tabular document information such as electronic money usage history with minimal workload.

[0075] According to this embodiment, the information processing system 1 can perform expense reimbursement with less workload.

[0076] In other words, the information processing system 1 includes a settlement information acquisition unit P6 that sends a prompt to a generation device M including target information to be processed, subject information K having multiple subject descriptions K3, a subject description inference instruction that causes the system to infer which of the multiple subject descriptions K3 the target information corresponds to, and a settlement information generation instruction that generates settlement information for performing expense settlement from the target information based on the subject descriptions K3, and acquires settlement information from the generation device M; and a question unit P7 that, upon acquiring a question comment from the generation device M, displays the question comment acquired from the generation device M, accepts a response from the user, and sends a prompt to the generation device M including a response and an additional generation instruction that generates and outputs settlement information with the response as additional information.

[0077] If the settlement information acquisition unit P6 has missing information necessary to generate the settlement information instructed in the account description K3, it includes a question output instruction in the prompt to generate and output a question comment asking the user for the missing information, and sends it to the generation device M.

[0078] The pre-application registration unit P1 receives pre-applications from users that include a date and amount, and registers them in the pre-application group As.

[0079] The pre-application extraction unit M2 extracts pre-applications corresponding to the target information from the pre-application group As in which pre-applications have been registered. The settlement information acquisition unit P6 sends a prompt to the generation device M including the extracted pre-applications, target information, account information K, account description inference instruction, and settlement information generation instruction, and acquires settlement information from the generation device M. The settlement information generation instruction includes an instruction to generate settlement information for expense settlement from the target information based on the account description K3 and the extracted pre-applications.

[0080] If the settlement information acquisition unit P6 has successfully acquired the settlement information, the application unit P9 performs settlement flow processing based on the settlement information.

[0081] The settlement information includes the expense name, participant, purpose, issuer, registration number, or at least one of the purposes, the date, and the amount.

[0082] The pre-application registration unit P1 receives pre-applications with monetary amounts from users and registers them in the pre-application group As. The amount determination unit P8, based on the amount registered in the pre-application, outputs a determination result indicating whether the amount included in the settlement information is an appropriate amount suitable for expense settlement, under predetermined conditions.

[0083] The target information extraction unit P3 extracts the target information to be processed from among multiple pieces of target information.

[0084] The information processing method involves the settlement information acquisition unit P6 sending a prompt to the generation device M that includes the target information to be processed, subject information K having multiple subject descriptions K3, a subject description inference instruction to infer which of the multiple subject descriptions K3 the target information corresponds to, and a settlement information generation instruction to generate settlement information for expense settlement based on the subject descriptions K3 from the target information, and acquiring settlement information from the generation device M. When the question unit P7 acquires a question comment from the generation device M, it displays the question comment acquired from the generation device M to accept a response from the user, and sends a prompt to the generation device M that includes the response and an additional generation instruction to generate and output settlement information with the response as additional information.

[0085] The information processing program causes the computer to execute the following: a prompt that includes target information to be processed, subject information K having multiple subject descriptions K3, a subject description inference instruction that causes the computer to infer which of the multiple subject descriptions K3 the target information corresponds to, and a settlement information generation instruction that generates settlement information for performing expense settlement from the target information based on the subject descriptions K3, and the code of the settlement information acquisition unit P6 that acquires settlement information from the generation device M; and a prompt that includes an additional generation instruction that, when a question comment is acquired from the generation device M, displays the question comment acquired from the generation device M, accepts a response from the user, and generates and outputs settlement information with the response as additional information.

[0086] In this embodiment, an example was described in which the subject information K and the pre-application group As are stored in the business device B. However, the embodiment is not limited to this, and the subject information K and the pre-application group As may be stored in the central device 20, the terminal device 10, or other devices.

[0087] Furthermore, while the embodiments described examples in which the settlement information includes the date and amount required for expense settlement processing, as well as the expense name, participant, purpose, issuer, registration number, or purpose, it is not limited to these examples. The settlement information can also include information set in advance by the user, such as the setting of mandatory related applications, the setting of application routes, the recipients of application sharing, and the requirement of prior application.

[0088] As described above, embodiments of the present invention have been explained, but the embodiments are not limited thereto, and various changes and modifications are possible as long as they do not alter the essence of the present invention.

[0089] Furthermore, each step of each procedure in the embodiment may be executed in a different order, executed simultaneously, or executed in a different order each time, as long as it does not contradict its nature. In addition, all or part of each step of each procedure in this embodiment may be implemented by hardware or by program. [Explanation of Symbols]

[0090] 1. Information Processing System 10. Terminal device 11. Terminal Control Unit 12. Terminal Storage Unit 13...Display section 14. Instruction Input Section 20...Central device 21... Control Unit 22...Storage section As...Pre-application group B...Business equipment Bp... Settlement Processing Unit G ···Expense reimbursement screen K...Course information K1... Subject K2...Tax category K3... Subject Description M...Generation device M1...Text reading unit M2...Pre-application extraction section M3...Settlement information generation section P1... Pre-registration section P2...Target Information Acquisition Unit P3...Target Information Extraction Unit P4...Advance application acquisition department P5...Course acquisition department P6...Settlement information acquisition section P7...Question section P8...Amount judgment section P9...Application Department

Claims

1. A settlement information acquisition unit sends a prompt to a generation device that includes target information to be processed, subject information having multiple subject descriptions, a subject description inference instruction that causes the system to infer which of the multiple subject descriptions the target information corresponds to, a settlement information generation instruction that generates settlement information including the subject, date, and amount corresponding to the subject description for expense settlement based on the subject description of the target information, and a question output instruction that generates and outputs a question comment to inquire with the user about the missing information that is not included in the target information when the system infers that the information instructed to be included in the settlement information in the instruction statement included in the subject description is not included in the target information, and acquires the settlement information or the question comment from the generation device. The system includes a question unit that, upon obtaining the question comment from the generating device, displays the question comment obtained from the generating device, accepts a response from the user, and sends a prompt to the generating device that includes the response and an additional generation instruction to generate and output the settlement information as additional information from the response. The aforementioned target information is text information subject to expense reimbursement processing, and the aforementioned text information is information based on document images or text received from the user. Information processing system.

2. It has a pre-registration department, The aforementioned pre-application registration unit receives the aforementioned pre-application for expense settlement, which includes a date and amount, from the user and registers it in the pre-application group. The information processing system according to claim 1.

3. Having a pre-application acquisition unit, The pre-application acquisition unit transmits to the generation device the target information, the pre-application group, and a pre-application extraction instruction to extract pre-applications from among the pre-applications included in the pre-application group that are similar to the date and amount included in the target information, and acquires the extracted pre-applications from the generation device. The settlement information acquisition unit transmits the prompt, which includes the extraction pre-application, the target information, the subject information, the subject explanation inference instruction, and the settlement information generation instruction, to the generation device, and acquires the settlement information or the question comment from the generation device. The settlement information generation instruction includes an instruction to generate the settlement information for performing expense settlement from the target information based on the subject description and the prior extraction application, The information processing system according to claim 2.

4. It has an application department, If the settlement information acquisition unit has obtained the settlement information, the application unit will perform settlement flow processing based on the settlement information. The aforementioned settlement flow process involves placing the settlement input information in a pending approval state, then, upon approval by the administrator, placing the settlement input information in a pending transfer state, and finally, registering the journal entry information once the administrator has completed the transfer process. The information processing system according to claim 1.

5. The information processing system according to claim 4, wherein the settlement information includes at least one of the following: expense name, participant, purpose, issuer, or registration number, as well as a date and an amount.

6. It has an amount determination unit, The amount determination unit outputs a determination result indicating whether the amount included in the settlement information is an appropriate amount suitable for expense settlement, based on the amount registered in the prior application and under predetermined conditions. The information processing system according to claim 2.

7. It has a target information extraction unit, The aforementioned target information extraction unit extracts the target information to be processed from among a plurality of target information. The information processing system according to claim 1.

8. The settlement information acquisition unit sends a prompt to the generation device, which includes target information to be processed, subject information having multiple subject descriptions, a subject description inference instruction that causes the unit to infer which of the multiple subject descriptions the target information corresponds to, a settlement information generation instruction that generates settlement information including the subject, date, and amount corresponding to the subject description for expense settlement from the target information based on the subject description, and a question output instruction that generates and outputs a question comment to inquire with the user about the missing information if it is inferred that the information instructed to be included in the settlement information in the instruction statement included in the subject description is not included in the target information, and the unit acquires the settlement information or the question comment from the generation device. When the question unit obtains the question comment from the generation device, it displays the question comment obtained from the generation device and accepts the answer from the user, and sends a prompt to the generation device that includes the answer and an additional generation instruction to generate and output the settlement information using the answer as additional information. The aforementioned target information is text information subject to expense reimbursement processing, and the aforementioned text information is information based on document images or text received from the user. Information processing methods.

9. A code for a settlement information acquisition unit that sends a prompt to a generation device including target information to be processed, subject information having multiple subject descriptions, a subject description inference instruction that causes the generation device to infer which of the multiple subject descriptions the target information corresponds to, a settlement information generation instruction that generates settlement information including the subject, date, and amount corresponding to the subject description for performing expense settlement from the target information based on the subject description, and a question output instruction that generates and outputs a question comment to inquire with the user about the missing information that is not included in the target information when it is inferred that the information instructed to be included in the settlement information in the instruction statement included in the subject description is not included in the target information, and acquires the settlement information or the question comment from the generation device, The code of the question unit, upon obtaining the question comment from the generation device, displays the question comment obtained from the generation device, accepts a response from the user, and sends a prompt to the generation device that includes the response and an additional generation instruction to generate and output the settlement information as additional information from the response. Have the computer run it, The aforementioned target information is text information subject to expense reimbursement processing, and the aforementioned text information is information based on document images or text received from the user. Information processing program.

Images