WO2022054165A1 - Data processing device, data processing method, and program - Google Patents

Data processing device, data processing method, and program Download PDF

Info

Publication number
WO2022054165A1
WO2022054165A1 PCT/JP2020/034103 JP2020034103W WO2022054165A1 WO 2022054165 A1 WO2022054165 A1 WO 2022054165A1 JP 2020034103 W JP2020034103 W JP 2020034103W WO 2022054165 A1 WO2022054165 A1 WO 2022054165A1
Authority
WO
WIPO (PCT)
Prior art keywords
text information
data
past
departments
apportionment
Prior art date
Application number
PCT/JP2020/034103
Other languages
French (fr)
Japanese (ja)
Inventor
鴻鵬 葛
顕 松田
智 小俣
啓太郎 森
Original Assignee
ファーストアカウンティング株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ファーストアカウンティング株式会社 filed Critical ファーストアカウンティング株式会社
Priority to PCT/JP2020/034103 priority Critical patent/WO2022054165A1/en
Priority to JP2020547247A priority patent/JP6810306B1/en
Priority to JP2020205042A priority patent/JP2022045868A/en
Publication of WO2022054165A1 publication Critical patent/WO2022054165A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes

Definitions

  • the present invention relates to a data processing device, a data processing method and a program for processing voucher data.
  • the read amount may be apportioned to multiple departments.
  • a journal entry rule is provided for voucher data to be apportioned by a plurality of departments, and when journal data is created from the voucher data, journal data is generated based on the journal entry rule. is doing.
  • the voucher data is new voucher data for which journal data has not been registered so far, there is no journal entry rule.
  • the accounting staff can manually apportion the data to multiple departments. It was necessary to create data, which reduced business efficiency.
  • the present invention has been made in view of these points, and an object thereof is to be able to create journal data corresponding to the apportionment to a plurality of departments for new voucher data.
  • the data processing device is extracted from a data acquisition unit for acquiring voucher data, an extraction unit for extracting text information from the voucher data, and voucher data acquired in the past.
  • the text information is based on at least a part of the past text information and the journal data created in the past corresponding to the past text information, in which the proportion of the amount to each of one or more departments is specified. It has an apportionment rate determining unit that determines the apportionment rate of an amount corresponding to at least a part thereof to each of the one or more departments, and an output unit that outputs the determined apportionment rate.
  • the past text information includes a product name or a service name and a transaction amount corresponding to the product name or the service name
  • the apportionment rate determination unit includes the product name or the service name included in the text information. And, the apportionment rate to each of the above-mentioned one or more departments corresponding to the transaction amount corresponding to the product name or the service name may be determined.
  • the past text information includes date information, and the apportionment rate determination unit may determine the apportionment rate for each of the one or more departments corresponding to the dates included in the text information.
  • the apportionment rate determination unit corresponds to at least a part of the text information and information about the business partner corresponding to the text information based on the information about the business partner corresponding to the past text information and the journal data.
  • the apportionment rate to each of the above-mentioned one or more departments may be determined.
  • the apportionment rate determination unit Based on the past text information, the information about the person who entered the journal data, and the journal data, the apportionment rate determination unit has at least a part of the text information and the journal data corresponding to the text information. The apportionment rate to each of the above-mentioned one or more departments corresponding to the input person may be determined.
  • the apportionment rate determination unit is based on the result of learning from the teacher data including the past text information and the journal data registered in the data processing device in the past corresponding to the past text information.
  • the apportionment rate to each of the above-mentioned one or more departments corresponding to the product name or service name included in the above and the transaction amount corresponding to the product name or service name may be determined.
  • the data processing device includes at least a part of journal data registered in the accounting system corresponding to the apportionment rate output by the output unit, and at least a part of the past text information corresponding to the journal data. It may further have a learning unit that performs the learning based on the teacher data.
  • the apportionment rate determination unit When the extracted text information does not correspond to a preset rule for determining the apportionment rate, the apportionment rate determination unit at least the text information based on the result learned from the teacher data. When the apportionment rate to each of the one or more departments corresponding to a part is determined and the extracted text information corresponds to the rule, the apportionment rate to each of the one or more departments is based on the rule. May be determined.
  • the apportionment rate determination unit has at least the text information based on the result of learning from the second teacher data including the past text information and information indicating whether or not the past text information is apportioned to a plurality of departments.
  • the text is based on the result learned from the teacher data.
  • the apportionment rate of information to each of a plurality of departments may be determined.
  • the data processing method is a step of acquiring voucher data executed by a computer, a step of extracting text information from the voucher data, and a voucher data acquired in the past.
  • the text is based on at least a portion of the past text information extracted and journal data created in the past corresponding to the past text information that specifies the proportion of the amount to each of one or more departments. It has a step of determining the apportionment rate of the amount of money corresponding to at least a part of the information to each of the one or more departments, and a step of outputting the determined apportionment rate.
  • the computer is extracted from the data acquisition unit for acquiring voucher data, the extraction unit for extracting text information from the voucher data, and the voucher data acquired in the past. At least some of the past text information and at least the text information, based on the journal data created in the past corresponding to the past text information, with the apportionment rate of the amount to each of one or more departments specified. It functions as an apportionment rate determination unit that determines the apportionment rate of the corresponding amount of money to each of the one or more departments, and an output unit that outputs the determined apportionment rate.
  • journal data corresponding to the apportionment to a plurality of departments for new voucher data.
  • FIG. 1 is a diagram for explaining an outline of the data processing device 1.
  • the data processing device 1 generates journal data using the result of analyzing a character string included in the voucher data generated by reading the voucher by a reading device 2 such as a scanner or a digital camera, and registers the journal data in the accounting system 3.
  • It is a device for, for example, a computer.
  • the data processing device 1 may be configured by one computer or may be configured by a plurality of computers.
  • a voucher is a voucher that shows information about a product such as a product name and a price of the product, such as an invoice or a purchase order. In the present embodiment, the description will proceed by taking as an example the case where the voucher is an invoice received by the user of the data processing device 1 as the demandee.
  • FIG. 2 is a diagram showing an example of an invoice received by the demandee.
  • the invoice shown in FIG. 2 describes the name, address, and contact information of the business operator that issued the invoice, that is, the requester who charges the price in the transaction.
  • the invoice includes the subject (January 2020 shown in Fig. 2), the purchase date of the product to be billed, the product name, the price of the product (that is, the subtotal), and the total price of multiple products (that is, the billing amount). ) Is described.
  • the data processing device 1 extracts text information by performing OCR processing from the invoice data generated by the reading device 2 reading the invoice, and based on the extracted text information, the amount of money to each of one or more departments. Create journal data with a specified apportionment rate.
  • the data processing device 1 has at least a part of past text information extracted from invoice data acquired in the past and one or more departments created in the past corresponding to the past text information. Determine the apportionment rate for each of one or more departments corresponding to at least a portion of the textual information, based on the results learned from the teacher data, including at least a portion of the specified journal data. .. By doing so, the data processing device 1 can create journal data including the amount corresponding to the apportionment to a plurality of departments and the department information of the apportionment destination for the new invoice data.
  • FIG. 3 is a diagram showing a functional configuration of the data processing device 1.
  • the data processing device 1 includes a communication unit 11, an operation unit 12, a display unit 13, a storage unit 14, and a control unit 15.
  • the control unit 15 includes a data acquisition unit 151, an extraction unit 152, an account item determination unit 153, a proportional division rate determination unit 154, a journal data creation unit 155 as an output unit, a registration unit 156, and a learning unit 157.
  • the communication unit 11 is a communication interface for connecting to a network (for example, an intranet or the Internet), and provides a communication controller for receiving data from the reading device 2 and transmitting / receiving data to / from another computer.
  • a network for example, an intranet or the Internet
  • the operation unit 12 performs an operation of registering the journal data created from the invoice data by the data processing device 1, a keyboard, a mouse, a display, etc. for correcting the apportionment ratio of the journal data to one or more departments.
  • the display unit 13 is a display for displaying information.
  • the display unit 13 displays information based on the instructions of the control unit 15.
  • the storage unit 14 has a storage medium such as a ROM (ReadOnlyMemory), a RAM (RandomAccessMemory), and a hard disk.
  • the storage unit 14 stores a program executed by the control unit 15. Further, the storage unit 14 stores the invoice data generated by the reading device 2 reading the invoice.
  • the storage unit 14 further stores a journal database for storing journal data created from the invoice data.
  • FIG. 4 is a diagram showing an example of a journal database.
  • the journal database stores, for example, the same information as the information registered in the accounting system or some information among the information registered in the accounting system.
  • the journal database is used for learning the account item determination model, the apportionment determination model, and the apportionment rate determination model, which will be described later.
  • the journal database the date on which the transaction corresponding to the journal was made, the debit account, the debit sub-item, the debit amount, the credit account, the credit sub-item, and the credit amount are displayed. It is associated.
  • the journal database may further include voucher identification information (eg, invoice number, etc.) for identifying vouchers such as invoices.
  • the control unit 15 is, for example, a CPU (Central Processing Unit). By executing the program stored in the storage unit 14, the control unit 15 executes the data acquisition unit 151, the extraction unit 152, the account item determination unit 153, the apportionment rate determination unit 154, the journal data creation unit 155, the registration unit 156, and the control unit 15. And functions as a learning unit 157.
  • a CPU Central Processing Unit
  • the data acquisition unit 151 acquires the invoice data generated by the reading device 2 reading the invoice as a voucher.
  • the data acquisition unit 151 may directly acquire the invoice data from the reading device 2, or may acquire the invoice data from a terminal communicably connected to the data processing device 1.
  • the extraction unit 152 extracts text information from the voucher data acquired by the data acquisition unit 151.
  • the extraction unit 152 extracts the character string included in the invoice indicated by the invoice data, for example, by executing the OCR process.
  • the extraction unit 152 includes date information indicating the issue date of the invoice, a product name or service name, and a transaction amount corresponding to the product name or service name.
  • extract extract.
  • the extraction unit 152 extracts a plurality of product names or service names and transaction amounts included in the invoice indicated by the invoice data line by line, thereby extracting the product name or service name and the product name or service name. Extract the transaction amount corresponding to.
  • the account item determination unit 153 determines an account item corresponding to at least a part of the text information extracted from the invoice data by the extraction unit 152.
  • the account determination unit 153 determines an account using an account determination model that outputs an account.
  • the account item determination model is a model that outputs a debit account item and a credit account item when a part of the text information extracted by the extraction unit 152 is input.
  • the account item determination model is generated by learning in advance using the first teacher data in which the input information to the model and the output information corresponding to the input information are set.
  • the input information included in the first teacher data is, for example, information obtained by combining date information extracted from invoice data acquired in the past, a product or service name, and a transaction amount corresponding to the product or service name. Is.
  • the output information is a debit account and a credit account corresponding to the input information.
  • the generated account determination model is stored in the storage unit 14.
  • the account item determination unit 153 selects one product name or service name from a plurality of product names or service names extracted by the extraction unit 152, and specifies the transaction amount corresponding to the selected product name or service name.
  • the account item determination unit 153 inputs the combination of the date information indicating the date of the invoice, the selected product name or service name, and the specified transaction amount into the account item determination model as input information.
  • the account determination unit 153 determines the debit account and the credit account by acquiring the debit account and the credit account output from the account determination model.
  • the account item determination unit 153 sequentially selects each of the plurality of product names or service names extracted by the extraction unit 152, and determines the debit account item and the credit account item using the account item determination model.
  • the account determination unit 153 uses the account rule information. Accounts may be determined based on. Then, the account item determination unit 153 may determine an account item by using the account item determination model when it does not correspond to the preset account item rule information.
  • the storage unit 14 stores at least a part of the text information extracted from the invoice data and the account rule information associated with the debit account and the credit account. At least a portion of the textual information is, for example, a combination of customer information indicating the business partner corresponding to the invoice, date information indicating the date of the invoice, and goods or services contained in the invoice.
  • the account item determination unit 153 stores the customer information corresponding to the invoice data, the date information of the invoice extracted by the extraction unit 152, and the selected product name or service name in association with each other in the storage unit 14. It is determined whether or not it is stored in the account rule information. When the account determination unit 153 determines that it is stored in the account rule information, the debit account and the credit account associated with the account information, date information, product name or service name are set to the product. Determine the debit and credit accounts that correspond to the name or service name. By doing so, the account item determination unit 153 can accurately identify the account item for the invoice received from the business partner who has made a transaction in the past.
  • the apportionment rate determination unit 154 determines the apportionment rate for each of one or more departments corresponding to at least a part of the text information extracted by the extraction unit 152.
  • the apportionment rate determination unit 154 is based on the past text information and the journal data for which the apportionment rate of the amount to each of one or more departments is specified, which was created in the past in response to the past text information. Determine the apportionment rate for each of one or more departments of the amount corresponding to at least a part of.
  • the apportionment rate determination unit 154 extracts based on the result of learning from the teacher data including the past text information and at least a part of the journal data acquired in the past corresponding to the past text information.
  • the proportion of the amount of money corresponding to at least a part of the text information extracted by the department 152 to one or more departments is determined.
  • the past text information is at least a part of the information extracted from the invoice data of the invoice acquired in the past by the data acquisition unit 151.
  • the journal data is journal data in which the proportion of the amount of money to each of one or more departments is specified.
  • the apportionment rate determination unit 154 is based on the result of learning from the second teacher data including the past text information and the information indicating whether or not the past text information is apportioned to a plurality of departments. Determine whether to apportion the text information to multiple departments for at least a part of. Then, when the apportionment rate determination unit 154 determines that the text information is apportioned to a plurality of departments, the apportionment rate determination unit 154 includes the past text information and at least a part of the journal data created in the past corresponding to the past text information. Based on the result of learning from the third teacher data, the apportionment rate of the text information to each of the plurality of departments is determined.
  • the apportionment rate determination unit 154 has an apportionment determination model that determines whether or not to apportion each of a plurality of departments when at least a part of the text information extracted by the extraction unit 152 is input, and an extraction unit.
  • the apportionment rate is determined by using the apportionment rate determination model that outputs the apportionment rate corresponding to each of the plurality of departments.
  • the apportionment determination model is generated by learning in advance from the second teacher data in which the input information to the model and the output information corresponding to the input information are set.
  • the input information included in the second teacher data is, for example, past text information extracted from invoice data acquired in the past.
  • the output information is determination information indicating whether or not to apportion the input information to a plurality of departments.
  • the generated apportionment determination model is stored in the storage unit 14.
  • the apportionment rate determination model is generated by learning in advance from the third teacher data, which is a set of the input information to the model and the output information corresponding to the input information.
  • the input information included in the third teacher data is, for example, past text information.
  • the output information is a proportional division rate corresponding to each of the plurality of departments corresponding to the input information.
  • the generated apportionment rate determination model is stored in the storage unit 14.
  • the past text information included in the teacher data used for learning the apportionment rate determination model corresponds to the product name or service name extracted from the invoice data of the invoice acquired in the past and the product name or service name. It contains the transaction amount to be processed and date information indicating the date of the invoice.
  • the apportionment rate determination unit 154 selects one product name or service name from a plurality of product names or service names extracted by the extraction unit 152, and specifies the transaction amount corresponding to the selected product name or service name.
  • the apportionment rate determination unit 154 inputs the date information indicating the date of the invoice, the selected product name or service name, and the specified transaction amount into the apportionment determination model stored in the storage unit 14, and the apportionment determination unit Based on the determination result output from the model, it is determined whether or not to perform proportional division.
  • the apportionment rate determination unit 154 determines that the apportionment is to be performed, the apportionment rate storing the date information indicating the date of the invoice, the selected product name or service name, and the specified transaction amount in the storage unit 14. Enter in the decision model. Then, the apportionment rate determination unit 154 obtains the apportionment rate corresponding to each of the plurality of departments output from the apportionment rate determination model, so that the selected product name or service name and the product name or service name can be obtained. Determine the apportionment rate for each of the corresponding transaction amounts and one or more departments.
  • the apportionment rate determination unit 154 is the text information extracted by the extraction unit 152 based on the result of learning from the teacher data including the past text information, the information about the business partner corresponding to the past text information, and the journal data. The apportionment rate to at least a part and one or more departments corresponding to the information about the business partner corresponding to the text information may be determined.
  • the apportionment determination model and the apportionment rate determination model are generated by learning the past text information and the information about the business partner corresponding to the past text information as input information.
  • the apportionment rate determination unit 154 inputs date information indicating the date of the invoice, the selected product name or service name, the specified transaction amount, and information on the business partner corresponding to the invoice into the apportionment determination model. Acquire the judgment result.
  • the apportionment rate determination unit 154 determines that the specified transaction amount is apportioned to a plurality of departments based on the determination result, the date information indicating the date of the invoice, the selected product name or service name, and the specified transaction amount are determined.
  • the apportionment rate determination model determines that the specified transaction amount is apportioned to a plurality of departments based on the determination result.
  • the apportionment rate determination unit 154 is based on the result of learning from the past text information, the information regarding the input person of the journal data corresponding to the past text information, and the teacher data including the journal data, and at least one of the text information.
  • the apportionment rate to each of one or more departments corresponding to the department and the person who inputs the journal data corresponding to the text information may be determined.
  • Information about the person who entered the journal data corresponding to the past text information includes, for example, user identification information for identifying the person, a department code for identifying the department to which the person belongs, and a project in charge of the person. Project code for identification.
  • the apportionment determination model and the apportionment rate determination model are generated by learning past text information and information about an input person of journal data corresponding to the past text information as input information.
  • the apportionment rate determination unit 154 inputs the date information indicating the date of the invoice, the selected product name or service name, the specified transaction amount, and the user ID of the input person who inputs the text information into the apportionment determination model. And get the judgment result.
  • the apportionment rate determination unit 154 determines that the apportionment rate is to be apportioned to a plurality of departments based on the determination result, the date information indicating the date of the invoice, the selected product name or service name, the specified transaction amount, and the data processing device 1
  • One or more departments by inputting the user ID for identifying the user who is operating the Determine the apportionment rate for each.
  • the apportionment rate determination unit 154 can determine the apportionment rate to a plurality of departments according to the relationship between the journal data input person, the department to which the journal data belongs, the project to which the journal data belongs, and the text information.
  • the apportionment rate determination unit 154 corresponds to the apportionment rate rule information.
  • the apportionment rate may be determined based on the apportionment rate rule information, and when the apportionment rate rule information is not supported, the apportionment rate may be determined using the apportionment determination model and the apportionment rate determination model.
  • the storage unit 14 stores at least a part of the text information extracted from the invoice data and the apportionment rate rule information in which the apportionment rate to each of one or more departments is associated with each other.
  • At least a portion of the textual information is, for example, a combination of customer information indicating the business partner corresponding to the invoice, date information indicating the date of the invoice, and goods or services contained in the invoice.
  • the apportionment rate determination unit 154 stores the customer information corresponding to the invoice data, the date information of the invoice extracted by the extraction unit 152, and the selected product name or service name in association with each other in the storage unit 14. By determining whether or not the information is stored in the apportionment rate rule information, it is determined whether or not these information correspond to the apportionment rate determination rule.
  • the apportionment rate determination unit 154 when the customer information, the date information, and the selected product name or service name correspond to the apportionment rate determination rule, each of one or more departments based on the apportionment rate rule information. Determine the apportionment rate to.
  • the apportionment rate determination unit 154 determines the apportionment rate to the customer information corresponding to the invoice data, the date information, and the apportionment rate to one or more departments associated with the selected product name or service name. , Determine the proportion to one or more departments corresponding to the product name or service name. By doing so, the apportionment rate determination unit 154 can accurately specify the apportionment rate for the invoice received from the business partner who has made a transaction in the past.
  • the journal data creation unit 155 outputs the account item determined by the account item determination unit 153 and the apportionment rate determined by the apportionment rate determination unit 154. For example, the journal data creation unit 155 prorates the date information extracted by the extraction unit 152 and the account items determined by the account item determination unit 153 for each of the one or more goods or services extracted by the extraction unit 152. Journal data including the apportionment rate determined by the rate determination unit 154 and the department name of the apportionment destination is created.
  • the journal data creation unit 155 outputs a registration screen for accepting registration of the created journal data to the accounting system to the display unit 13.
  • FIG. 5 is a diagram showing an example of a registration screen output by the journal data creation unit 155.
  • the journal data shown in FIG. 5 is the journal data generated based on the invoice corresponding to FIG.
  • the registration unit 156 registers the journal data displayed on the registration screen in the accounting system.
  • the registration unit 156 may accept the correction of the journal data displayed on the registration screen from the user.
  • the registration unit 156 registers the journal data shown on the registration screen in the accounting system 3 and stores the journal data in the journal database in response to the pressing of the registration button on the registration screen.
  • the learning unit 157 corresponds to the apportionment rate output by the journal data creation unit 155, and at least a part of the journal data registered in the accounting system 3 by the registration unit 156 and at least one of the past text information corresponding to the journal data.
  • the account item determination model, the apportionment judgment model, and the apportionment rate determination model are learned from the teacher data including the department.
  • the learning unit 157 learns the account item determination model, the apportionment determination model, and the apportionment rate determination model using the newly stored information as teacher data in response to the information stored in the journal entry database by the registration unit 156. conduct. By doing so, in the data processing device 1, the more the journal data is registered in the accounting system 3, the more the model can be learned. Therefore, the account output from the model is apportioned or not. It is possible to improve the accuracy of the determination result and the proportional division rate.
  • FIG. 6 is a flowchart showing an example of the processing flow in the data processing apparatus 1.
  • the data acquisition unit 151 acquires the invoice data generated by the reading device 2 reading the invoice (S1).
  • the extraction unit 152 extracts text information from the invoice data acquired by the data acquisition unit 151 (S2).
  • the account item determination unit 153 selects a product name or service name that has never been selected from a plurality of products or services included in the text information extracted by the extraction unit 152 (S3). Further, the account item determination unit 153 specifies the transaction amount corresponding to the selected product name or service name from the plurality of transaction amounts included in the text information extracted by the extraction unit 152 (S4).
  • the account determination unit 153 inputs the date information of the invoice, the selected product name or service name, and the specified transaction amount into the account determination model, and the account output from the account determination model.
  • the account item is determined by acquiring (S5).
  • the apportionment rate determination unit 154 determines the apportionment rate corresponding to the date information of the invoice, the selected product name or service name, and the specified transaction amount (S6). Specifically, the apportionment rate determination unit 154 inputs the date information of the invoice, the selected product name or service name, and the specified transaction amount into the apportionment determination model, and the determination result output from the apportionment determination model. Determines whether or not to perform proportional division.
  • the apportionment rate determination unit 154 determines that the apportionment is to be performed, the date information of the invoice, the selected product name or service name, and the specified transaction amount are input to the apportionment rate determination model and output from the apportionment rate determination model. By acquiring the apportionment rate for each of the plurality of departments, the apportionment rate for each of the plurality of departments is determined.
  • control unit 15 determines whether or not all of the plurality of products or services included in the text information extracted by the extraction unit 152 have been selected (S7).
  • the process is transferred to S8, and when it is determined that all the plurality of products or services have not been selected, the control unit 15 transfers the process to S3.
  • the journal data creation unit 155 includes the date information extracted by the extraction unit 152 and the account items determined by the account item determination unit 153, corresponding to each of the one or more products or services extracted by the extraction unit 152. , Creates journal data including the apportionment rate determined by the apportionment rate determination unit 154 and the department name of the apportionment destination, and outputs a registration screen for accepting registration of the journal data to the accounting system to the display unit 13 (S8). ..
  • the registration unit 156 appropriately accepts corrections of the journal data displayed on the registration screen from the user, and then registers the journal data displayed on the registration screen in the accounting system (S9).
  • the learning unit 157 determines the account item based on the teacher data including at least a part of the journal data registered in the accounting system 3 by the registration unit 156 and at least a part of the past text information corresponding to the journal data. Further learning of the model, the apportionment determination model, and the apportionment rate determination model is performed (S10).
  • the data processing device 1 has created at least a part of the past text information extracted from the invoice data of the invoice as a voucher acquired in the past and the past text information corresponding to the past text information. Based on the journal data for which the apportionment rate of the amount to each of one or more departments is specified, the apportionment of the amount corresponding to at least a part of the text information extracted by the extraction unit 152 to each of one or more departments. Determine the rate and output the determined apportionment rate. By doing so, the data processing device 1 can create journal data corresponding to the apportionment to a plurality of departments for the new invoice data.
  • the apportionment rate determination unit 154 determines the apportionment rate for each of one or more departments using the apportionment determination model and the apportionment rate determination model, but the present invention is not limited to this.
  • the apportionment rate determination unit 154 uses an apportionment rate determination model that outputs the apportionment rate to one or more departments in response to the input of the text information extracted by the extraction unit 152, and apportionments to each of the one or more departments. The rate may be determined.
  • all or a part of the device can be functionally or physically distributed / integrated in any unit.
  • new embodiments resulting from any combination of the plurality of embodiments are also included in the embodiments of the present invention. The effect of the new embodiment produced by the combination has the effect of the original embodiment together.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Finance (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A data processing device (1) comprises: a data acquisition unit (151) which acquires voucher data; an extraction unit (152) which extracts text information from the voucher data; a proportion determination unit (154) which determines the proportion of the amount of money corresponding to at least a part of the text information extracted by the extraction unit (152) to each of one or more departments, on the basis of at least a part of past text information extracted from the voucher data on vouchers issued in the past, and journal data that was created in the past in response to the past text information and that specifies the proportion of the amount of money to each of the one or more departments; and a journal data creation unit (155) which outputs the determined proportion.

Description

データ処理装置、データ処理方法及びプログラムData processing equipment, data processing methods and programs
　本発明は、証憑データを処理するデータ処理装置、データ処理方法及びプログラムに関する。 The present invention relates to a data processing device, a data processing method and a program for processing voucher data.
　請求書等の証憑データの画像データから商品名及び金額を読み取り、読み取った商品名及び金額に基づいて仕訳データを作成する会計処理システムが知られている（例えば、特許文献１を参照）。 There is known an accounting system that reads a product name and amount from image data of voucher data such as an invoice and creates journal data based on the read product name and amount (see, for example, Patent Document 1).
特開２０１８－１７３９３５号公報JP-A-2018-173935
　一つの証憑データから仕訳を行う場合に、読み取った金額を、複数の部署に按分することがある。これに対し、従来の会計処理システムでは、証憑データに対し、複数の部署で按分する仕訳ルールを設けておき、当該証憑データから仕訳データを作成する場合に、仕訳ルールに基づいて仕訳データを生成している。しかしながら、証憑データがこれまで仕訳データを登録したことのない新規の証憑データである場合には、仕訳ルールが存在しない。このため、従来の会計処理システムでは、新規の証憑データに対して、複数の部署に按分する仕訳データを作成することができず、経理担当者が手動で複数の部署への按分に対応した仕訳データを作成する必要があり、業務効率が低下していた。 When journalizing from one voucher data, the read amount may be apportioned to multiple departments. On the other hand, in the conventional accounting system, a journal entry rule is provided for voucher data to be apportioned by a plurality of departments, and when journal data is created from the voucher data, journal data is generated based on the journal entry rule. is doing. However, if the voucher data is new voucher data for which journal data has not been registered so far, there is no journal entry rule. For this reason, in the conventional accounting system, it is not possible to create journal data that is apportioned to multiple departments for new voucher data, and the accounting staff can manually apportion the data to multiple departments. It was necessary to create data, which reduced business efficiency.
　そこで、本発明はこれらの点に鑑みてなされたものであり、新規の証憑データに対し、複数の部署への按分に対応した仕訳データを作成できるようにすることを目的とする。 Therefore, the present invention has been made in view of these points, and an object thereof is to be able to create journal data corresponding to the apportionment to a plurality of departments for new voucher data.
　本発明の第１の態様に係るデータ処理装置は、証憑データを取得するデータ取得部と、前記証憑データから、テキスト情報を抽出する抽出部と、過去に取得された証憑の証憑データから抽出された少なくとも一部の過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された、一以上の部署それぞれへの金額の按分率が指定された仕訳データとに基づいて、前記テキスト情報の少なくとも一部に対応する金額の前記一以上の部署それぞれへの按分率を決定する按分率決定部と、決定された前記按分率を出力する出力部と、を有する。 The data processing device according to the first aspect of the present invention is extracted from a data acquisition unit for acquiring voucher data, an extraction unit for extracting text information from the voucher data, and voucher data acquired in the past. The text information is based on at least a part of the past text information and the journal data created in the past corresponding to the past text information, in which the proportion of the amount to each of one or more departments is specified. It has an apportionment rate determining unit that determines the apportionment rate of an amount corresponding to at least a part thereof to each of the one or more departments, and an output unit that outputs the determined apportionment rate.
　前記過去テキスト情報には、商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額とが含まれており、前記按分率決定部は、前記テキスト情報に含まれる商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額とに対応する前記一以上の部署それぞれへの按分率を決定してもよい。 The past text information includes a product name or a service name and a transaction amount corresponding to the product name or the service name, and the apportionment rate determination unit includes the product name or the service name included in the text information. And, the apportionment rate to each of the above-mentioned one or more departments corresponding to the transaction amount corresponding to the product name or the service name may be determined.
　前記過去テキスト情報には、日付情報が含まれており、前記按分率決定部は、前記テキスト情報に含まれる日付に対応する前記一以上の部署それぞれへの按分率を決定してもよい。 The past text information includes date information, and the apportionment rate determination unit may determine the apportionment rate for each of the one or more departments corresponding to the dates included in the text information.
　前記按分率決定部は、前記過去テキスト情報に対応する取引先に関する情報と、前記仕訳データとに基づいて、前記テキスト情報の少なくとも一部と、前記テキスト情報に対応する取引先に関する情報とに対応する前記一以上の部署それぞれへの按分率を決定してもよい。 The apportionment rate determination unit corresponds to at least a part of the text information and information about the business partner corresponding to the text information based on the information about the business partner corresponding to the past text information and the journal data. The apportionment rate to each of the above-mentioned one or more departments may be determined.
　前記按分率決定部は、前記過去テキスト情報と、前記仕訳データの入力者に関する情報と、前記仕訳データとに基づいて、前記テキスト情報の少なくとも一部と、前記テキスト情報に対応する前記仕訳データの入力者とに対応する前記一以上の部署それぞれへの按分率を決定してもよい。 Based on the past text information, the information about the person who entered the journal data, and the journal data, the apportionment rate determination unit has at least a part of the text information and the journal data corresponding to the text information. The apportionment rate to each of the above-mentioned one or more departments corresponding to the input person may be determined.
　前記按分率決定部は、前記過去テキスト情報と、当該過去テキスト情報に対応して前記データ処理装置に過去に登録された前記仕訳データとを含む教師データにより学習した結果に基づいて、前記テキスト情報に含まれる商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額とに対応する前記一以上の部署それぞれへの按分率を決定してもよい。 The apportionment rate determination unit is based on the result of learning from the teacher data including the past text information and the journal data registered in the data processing device in the past corresponding to the past text information. The apportionment rate to each of the above-mentioned one or more departments corresponding to the product name or service name included in the above and the transaction amount corresponding to the product name or service name may be determined.
　前記データ処理装置は、前記出力部が出力した前記按分率に対応して会計システムに登録された仕訳データの少なくとも一部と、当該仕訳データに対応する前記過去テキスト情報の少なくとも一部とを含む教師データにより前記学習を行う学習部をさらに有してもよい。 The data processing device includes at least a part of journal data registered in the accounting system corresponding to the apportionment rate output by the output unit, and at least a part of the past text information corresponding to the journal data. It may further have a learning unit that performs the learning based on the teacher data.
　前記按分率決定部は、抽出された前記テキスト情報が、予め設定されている按分率を決定するためのルールに対応していない場合、教師データにより学習した結果に基づいて、前記テキスト情報の少なくとも一部に対応する前記一以上の部署それぞれへの按分率を決定し、抽出された前記テキスト情報が前記ルールに対応している場合、前記ルールに基づいて前記一以上の部署それぞれへの按分率を決定してもよい。 When the extracted text information does not correspond to a preset rule for determining the apportionment rate, the apportionment rate determination unit at least the text information based on the result learned from the teacher data. When the apportionment rate to each of the one or more departments corresponding to a part is determined and the extracted text information corresponds to the rule, the apportionment rate to each of the one or more departments is based on the rule. May be determined.
　前記按分率決定部は、前記過去テキスト情報と、前記過去テキスト情報を複数の部署に按分するか否かを示す情報とを含む第２教師データにより学習した結果に基づいて、前記テキスト情報の少なくとも一部に対して前記テキスト情報を複数の部署に按分するか否かを判定し、前記テキスト情報を複数の部署に按分すると判定した場合に、前記教師データにより学習した結果に基づいて、前記テキスト情報の複数の部署それぞれへの按分率を決定してもよい。 The apportionment rate determination unit has at least the text information based on the result of learning from the second teacher data including the past text information and information indicating whether or not the past text information is apportioned to a plurality of departments. When it is determined whether or not to apportion the text information to a plurality of departments for a part and it is determined to apportion the text information to a plurality of departments, the text is based on the result learned from the teacher data. The apportionment rate of information to each of a plurality of departments may be determined.
　本発明の第２の態様に係るデータ処理方法は、コンピュータが実行する、証憑データを取得するステップと、前記証憑データから、テキスト情報を抽出するステップと、過去に取得された証憑の証憑データから抽出された少なくとも一部の過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された、一以上の部署それぞれへの金額の按分率が指定された仕訳データとに基づいて、前記テキスト情報の少なくとも一部に対応する金額の前記一以上の部署それぞれへの按分率を決定するステップと、決定された前記按分率を出力するステップと、を有する。 The data processing method according to the second aspect of the present invention is a step of acquiring voucher data executed by a computer, a step of extracting text information from the voucher data, and a voucher data acquired in the past. The text is based on at least a portion of the past text information extracted and journal data created in the past corresponding to the past text information that specifies the proportion of the amount to each of one or more departments. It has a step of determining the apportionment rate of the amount of money corresponding to at least a part of the information to each of the one or more departments, and a step of outputting the determined apportionment rate.
　本発明の第３の態様に係るプログラムは、コンピュータを、証憑データを取得するデータ取得部、前記証憑データから、テキスト情報を抽出する抽出部、過去に取得された証憑の証憑データから抽出された少なくとも一部の過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された、一以上の部署それぞれへの金額の按分率が指定された仕訳データとに基づいて、前記テキスト情報の少なくとも一部に対応する金額の前記一以上の部署それぞれへの按分率を決定する按分率決定部、及び、決定された前記按分率を出力する出力部、として機能させる。 In the program according to the third aspect of the present invention, the computer is extracted from the data acquisition unit for acquiring voucher data, the extraction unit for extracting text information from the voucher data, and the voucher data acquired in the past. At least some of the past text information and at least the text information, based on the journal data created in the past corresponding to the past text information, with the apportionment rate of the amount to each of one or more departments specified. It functions as an apportionment rate determination unit that determines the apportionment rate of the corresponding amount of money to each of the one or more departments, and an output unit that outputs the determined apportionment rate.
　本発明によれば、新規の証憑データに対し、複数の部署への按分に対応した仕訳データを作成できるという効果を奏する。 According to the present invention, it is possible to create journal data corresponding to the apportionment to a plurality of departments for new voucher data.
データ処理装置の概要を説明するための図である。It is a figure for demonstrating the outline of a data processing apparatus. 被請求者が受領する請求書の一例を示す図である。It is a figure which shows an example of the invoice received by the demandee. データ処理装置の機能構成を示す図である。It is a figure which shows the functional structure of a data processing apparatus. 仕訳データベースの一例を示す図である。It is a figure which shows an example of a journal entry database. 出力部が出力した登録画面の一例を示す図である。It is a figure which shows an example of the registration screen output by an output part. データ処理装置における処理の流れを示すフローチャートである。It is a flowchart which shows the flow of processing in a data processing apparatus.
［データ処理装置１の概要］
　図１は、データ処理装置１の概要を説明するための図である。データ処理装置１は、スキャナ又はデジタルカメラ等の読取装置２が証憑を読み取ることによって生成された証憑データに含まれる文字列を解析した結果を用いて仕訳データを生成し、会計システム３に登録するための装置であり、例えばコンピュータである。データ処理装置１は、１台のコンピュータにより構成されていてもよく、複数のコンピュータにより構成されていてもよい。証憑は、例えば、請求書や発注書等の、商品名、商品の金額等の商品に関する情報が示された証憑である。本実施形態では、証憑が、被請求者としてのデータ処理装置１のユーザが受領する請求書である場合を例として説明を進める。 [Overview of data processing device 1]
FIG. 1 is a diagram for explaining an outline of the data processing device 1. The data processing device 1 generates journal data using the result of analyzing a character string included in the voucher data generated by reading the voucher by a reading device 2 such as a scanner or a digital camera, and registers the journal data in the accounting system 3. It is a device for, for example, a computer. The data processing device 1 may be configured by one computer or may be configured by a plurality of computers. A voucher is a voucher that shows information about a product such as a product name and a price of the product, such as an invoice or a purchase order. In the present embodiment, the description will proceed by taking as an example the case where the voucher is an invoice received by the user of the data processing device 1 as the demandee.
　図２は、被請求者が受領する請求書の一例を示す図である。図２に示す請求書には、請求書を発行した事業者、すなわち取引における代金を請求する請求者の名称、住所、連絡先が記載されている。また、請求書には、件名（図２に示す２０２０年１月分）、請求の対象となる商品の購入日、品名、商品の金額（すなわち小計）、複数の商品の合計金額（すなわち請求額）が記載されている。データ処理装置１は、読取装置２が請求書を読み取ることにより生成した請求書データからＯＣＲ処理を行うことによりテキスト情報を抽出し、抽出したテキスト情報に基づいて一以上の部署それぞれへの金額の按分率が指定された仕訳データを作成する。 FIG. 2 is a diagram showing an example of an invoice received by the demandee. The invoice shown in FIG. 2 describes the name, address, and contact information of the business operator that issued the invoice, that is, the requester who charges the price in the transaction. In addition, the invoice includes the subject (January 2020 shown in Fig. 2), the purchase date of the product to be billed, the product name, the price of the product (that is, the subtotal), and the total price of multiple products (that is, the billing amount). ) Is described. The data processing device 1 extracts text information by performing OCR processing from the invoice data generated by the reading device 2 reading the invoice, and based on the extracted text information, the amount of money to each of one or more departments. Create journal data with a specified apportionment rate.
　請求書データがこれまで仕訳データを登録したことのない取引先等に対応する新規の請求書データである場合には、当該請求書データに対応する仕訳ルールが存在しない。これに対し、データ処理装置１は、過去に取得された請求書データから抽出された少なくとも一部の過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された、一以上の部署それぞれへの金額の按分率が指定された仕訳データの少なくとも一部とを含む教師データにより学習した結果に基づいて、テキスト情報の少なくとも一部に対応する一以上の部署それぞれへの按分率を決定する。このようにすることで、データ処理装置１は、新規の請求書データに対し、複数の部署への按分に対応した金額と、按分先の部署情報とを含む仕訳データを作成することができる。 If the invoice data is new invoice data corresponding to a business partner who has not registered journal data so far, there is no journal rule corresponding to the invoice data. On the other hand, the data processing device 1 has at least a part of past text information extracted from invoice data acquired in the past and one or more departments created in the past corresponding to the past text information. Determine the apportionment rate for each of one or more departments corresponding to at least a portion of the textual information, based on the results learned from the teacher data, including at least a portion of the specified journal data. .. By doing so, the data processing device 1 can create journal data including the amount corresponding to the apportionment to a plurality of departments and the department information of the apportionment destination for the new invoice data.
［データ処理装置１の機能構成及び動作］
　図３は、データ処理装置１の機能構成を示す図である。データ処理装置１は、通信部１１と、操作部１２と、表示部１３と、記憶部１４と、制御部１５とを有する。制御部１５は、データ取得部１５１と、抽出部１５２と、勘定科目決定部１５３と、按分率決定部１５４と、出力部としての仕訳データ作成部１５５と、登録部１５６と、学習部１５７とを有する。 [Functional configuration and operation of data processing device 1]
FIG. 3 is a diagram showing a functional configuration of the data processing device 1. The data processing device 1 includes a communication unit 11, an operation unit 12, a display unit 13, a storage unit 14, and a control unit 15. The control unit 15 includes a data acquisition unit 151, an extraction unit 152, an account item determination unit 153, a proportional division rate determination unit 154, a journal data creation unit 155 as an output unit, a registration unit 156, and a learning unit 157. Have.
　通信部１１は、ネットワーク（例えばイントラネット又はインターネット）に接続するための通信インターフェースであり、読取装置２からデータを受信したり、他のコンピュータとの間でデータを送受信したりするための通信コントローラを有する。 The communication unit 11 is a communication interface for connecting to a network (for example, an intranet or the Internet), and provides a communication controller for receiving data from the reading device 2 and transmitting / receiving data to / from another computer. Have.
　操作部１２は、データ処理装置１が請求書データから作成した仕訳データを登録する操作を行ったり、仕訳データにおける一以上の部署への按分率を訂正したりするためのキーボード、マウス及びディスプレイ等のデバイスを有する。
　表示部１３は、情報を表示するディスプレイである。表示部１３は、制御部１５の指示に基づいて情報を表示する。 The operation unit 12 performs an operation of registering the journal data created from the invoice data by the data processing device 1, a keyboard, a mouse, a display, etc. for correcting the apportionment ratio of the journal data to one or more departments. Have a device.
The display unit 13 is a display for displaying information. The display unit 13 displays information based on the instructions of the control unit 15.
　記憶部１４は、例えばＲＯＭ（Read Only Memory）、ＲＡＭ（Random Access Memory）及びハードディスク等の記憶媒体を有する。記憶部１４は、制御部１５が実行するプログラムを記憶する。また、記憶部１４は、読取装置２が請求書を読み取ることによって生成された請求書データを記憶する。記憶部１４は、さらに、請求書データから作成された仕訳データを格納する仕訳データベースを記憶する。 The storage unit 14 has a storage medium such as a ROM (ReadOnlyMemory), a RAM (RandomAccessMemory), and a hard disk. The storage unit 14 stores a program executed by the control unit 15. Further, the storage unit 14 stores the invoice data generated by the reading device 2 reading the invoice. The storage unit 14 further stores a journal database for storing journal data created from the invoice data.
　図４は、仕訳データベースの一例を示す図である。仕訳データベースには、例えば、会計システムに登録された情報と同じ情報又は会計システムに登録された情報のうち一部の情報が格納される。仕訳データベースは、後述する勘定科目決定モデル、按分判定モデル、按分率決定モデルの学習に用いられる。図４に示すように仕訳データベースでは、仕訳に対応する取引が行われた日付と、借方勘定科目と、借方補助科目と、借方金額と、貸方勘定科目と、貸方補助科目と、貸方金額とが関連付けられている。仕訳データベースにおいては、請求書等の証憑を特定するための証憑識別情報（例えば請求書番号等）がさらに含まれていてもよい。 FIG. 4 is a diagram showing an example of a journal database. The journal database stores, for example, the same information as the information registered in the accounting system or some information among the information registered in the accounting system. The journal database is used for learning the account item determination model, the apportionment determination model, and the apportionment rate determination model, which will be described later. As shown in FIG. 4, in the journal database, the date on which the transaction corresponding to the journal was made, the debit account, the debit sub-item, the debit amount, the credit account, the credit sub-item, and the credit amount are displayed. It is associated. The journal database may further include voucher identification information (eg, invoice number, etc.) for identifying vouchers such as invoices.
　制御部１５は、例えばＣＰＵ（Central Processing Unit）である。制御部１５は、記憶部１４に記憶されたプログラムを実行することにより、データ取得部１５１、抽出部１５２、勘定科目決定部１５３、按分率決定部１５４、仕訳データ作成部１５５、登録部１５６、及び学習部１５７として機能する。 The control unit 15 is, for example, a CPU (Central Processing Unit). By executing the program stored in the storage unit 14, the control unit 15 executes the data acquisition unit 151, the extraction unit 152, the account item determination unit 153, the apportionment rate determination unit 154, the journal data creation unit 155, the registration unit 156, and the control unit 15. And functions as a learning unit 157.
　データ取得部１５１は、読取装置２が証憑としての請求書を読み取ることによって生成された請求書データを取得する。データ取得部１５１は、読取装置２から請求書データを直接取得してもよいし、データ処理装置１と通信可能に接続されている端末から請求書データを取得してもよい。 The data acquisition unit 151 acquires the invoice data generated by the reading device 2 reading the invoice as a voucher. The data acquisition unit 151 may directly acquire the invoice data from the reading device 2, or may acquire the invoice data from a terminal communicably connected to the data processing device 1.
　抽出部１５２は、データ取得部１５１が取得した証憑データからテキスト情報を抽出する。抽出部１５２は、例えばＯＣＲ処理を実行することにより、請求書データが示す請求書に含まれている文字列を抽出する。抽出部１５２は、請求書データが示す請求書に含まれているテキスト情報として、請求書の発行日を示す日付情報と、商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額とを抽出する。抽出部１５２は、請求書データが示す請求書に含まれている複数の商品名又はサービス名と取引金額とを行ごとに抽出することにより、商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額とを抽出する。 The extraction unit 152 extracts text information from the voucher data acquired by the data acquisition unit 151. The extraction unit 152 extracts the character string included in the invoice indicated by the invoice data, for example, by executing the OCR process. As text information included in the invoice indicated by the invoice data, the extraction unit 152 includes date information indicating the issue date of the invoice, a product name or service name, and a transaction amount corresponding to the product name or service name. And extract. The extraction unit 152 extracts a plurality of product names or service names and transaction amounts included in the invoice indicated by the invoice data line by line, thereby extracting the product name or service name and the product name or service name. Extract the transaction amount corresponding to.
　勘定科目決定部１５３は、抽出部１５２が請求書データから抽出したテキスト情報の少なくとも一部に対応する勘定科目を決定する。勘定科目決定部１５３は、勘定科目を出力する勘定科目決定モデルを用いて勘定科目を決定する。勘定科目決定モデルは、抽出部１５２が抽出したテキスト情報の一部を入力すると借方勘定科目及び貸方勘定科目を出力するモデルである。 The account item determination unit 153 determines an account item corresponding to at least a part of the text information extracted from the invoice data by the extraction unit 152. The account determination unit 153 determines an account using an account determination model that outputs an account. The account item determination model is a model that outputs a debit account item and a credit account item when a part of the text information extracted by the extraction unit 152 is input.
　勘定科目決定モデルは、当該モデルへの入力情報と、当該入力情報に対応する出力情報とをセットにした第１教師データを用いて予め学習することにより生成される。第１教師データに含まれる入力情報は、例えば、過去に取得された請求書データから抽出された日付情報と、商品又はサービス名と、当該商品又はサービス名に対応する取引金額とを組み合わせた情報である。出力情報は、当該入力情報に対応する借方勘定科目及び貸方勘定科目である。生成された勘定科目決定モデルは記憶部１４に記憶されている。 The account item determination model is generated by learning in advance using the first teacher data in which the input information to the model and the output information corresponding to the input information are set. The input information included in the first teacher data is, for example, information obtained by combining date information extracted from invoice data acquired in the past, a product or service name, and a transaction amount corresponding to the product or service name. Is. The output information is a debit account and a credit account corresponding to the input information. The generated account determination model is stored in the storage unit 14.
　勘定科目決定部１５３は、抽出部１５２が抽出した複数の商品名又はサービス名から、一つの商品名又はサービス名を選択するとともに、選択した商品名又はサービス名に対応する取引金額を特定する。 The account item determination unit 153 selects one product name or service name from a plurality of product names or service names extracted by the extraction unit 152, and specifies the transaction amount corresponding to the selected product name or service name.
　勘定科目決定部１５３は、請求書の日付を示す日付情報と、選択した商品名又はサービス名と、特定した取引金額との組み合わせを入力情報として勘定科目決定モデルに入力する。勘定科目決定部１５３は、勘定科目決定モデルから出力される借方勘定科目及び貸方勘定科目を取得することにより、借方勘定科目及び貸方勘定科目を決定する。勘定科目決定部１５３は、抽出部１５２が抽出した複数の商品名又はサービス名それぞれを順次選択し、勘定科目決定モデルを用いて借方勘定科目及び貸方勘定科目を決定する。 The account item determination unit 153 inputs the combination of the date information indicating the date of the invoice, the selected product name or service name, and the specified transaction amount into the account item determination model as input information. The account determination unit 153 determines the debit account and the credit account by acquiring the debit account and the credit account output from the account determination model. The account item determination unit 153 sequentially selects each of the plurality of product names or service names extracted by the extraction unit 152, and determines the debit account item and the credit account item using the account item determination model.
　なお、勘定科目決定部１５３は、請求書データから抽出されたテキスト情報が、予め設定されている勘定科目を決定するための勘定科目ルール情報に対応している場合には、勘定科目ルール情報に基づいて勘定科目を決定してもよい。そして、勘定科目決定部１５３は、予め設定されている勘定科目ルール情報に対応していない場合に、勘定科目決定モデルを用いて勘定科目を決定してもよい。 If the text information extracted from the invoice data corresponds to the account rule information for determining the preset account, the account determination unit 153 uses the account rule information. Accounts may be determined based on. Then, the account item determination unit 153 may determine an account item by using the account item determination model when it does not correspond to the preset account item rule information.
　この場合、記憶部１４は、請求書データから抽出されたテキスト情報の少なくとも一部と、借方勘定科目及び貸方勘定科目と関連付けた勘定科目ルール情報と、を記憶する。テキスト情報の少なくとも一部は、例えば、請求書に対応する取引先を示す取引先情報と、請求書の日付を示す日付情報と、請求書に含まれる商品又はサービスとの組み合わせである。 In this case, the storage unit 14 stores at least a part of the text information extracted from the invoice data and the account rule information associated with the debit account and the credit account. At least a portion of the textual information is, for example, a combination of customer information indicating the business partner corresponding to the invoice, date information indicating the date of the invoice, and goods or services contained in the invoice.
　勘定科目決定部１５３は、請求書データに対応する取引先情報と、抽出部１５２が抽出した請求書の日付情報と、選択した商品名又はサービス名とが関連付けられて、記憶部１４に記憶されている勘定科目ルール情報に記憶されているか否かを判定する。勘定科目決定部１５３は、勘定科目ルール情報に記憶されていると判定すると、これらの取引先情報、日付情報、商品名又はサービス名に関連付けられている借方勘定科目及び貸方勘定科目を、当該商品名又はサービス名に対応する借方勘定科目及び貸方勘定科目に決定する。このようにすることで、勘定科目決定部１５３は、過去に取引したことがある取引先から受領した請求書に対して、勘定科目を精度良く特定することができる。 The account item determination unit 153 stores the customer information corresponding to the invoice data, the date information of the invoice extracted by the extraction unit 152, and the selected product name or service name in association with each other in the storage unit 14. It is determined whether or not it is stored in the account rule information. When the account determination unit 153 determines that it is stored in the account rule information, the debit account and the credit account associated with the account information, date information, product name or service name are set to the product. Determine the debit and credit accounts that correspond to the name or service name. By doing so, the account item determination unit 153 can accurately identify the account item for the invoice received from the business partner who has made a transaction in the past.
　按分率決定部１５４は、抽出部１５２が抽出したテキスト情報の少なくとも一部に対応する一以上の部署それぞれへの按分率を決定する。按分率決定部１５４は、過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された、一以上の部署それぞれへの金額の按分率が指定された仕訳データとに基づいて、テキスト情報の少なくとも一部に対応する金額の一以上の部署それぞれへの按分率を決定する。 The apportionment rate determination unit 154 determines the apportionment rate for each of one or more departments corresponding to at least a part of the text information extracted by the extraction unit 152. The apportionment rate determination unit 154 is based on the past text information and the journal data for which the apportionment rate of the amount to each of one or more departments is specified, which was created in the past in response to the past text information. Determine the apportionment rate for each of one or more departments of the amount corresponding to at least a part of.
　具体的には、按分率決定部１５４は、過去テキスト情報と、当該過去テキスト情報に対応して過去に取得された仕訳データの少なくとも一部とを含む教師データにより学習した結果に基づいて、抽出部１５２が抽出したテキスト情報の少なくとも一部に対応する金額の一以上の部署それぞれへの按分率を決定する。ここで、過去テキスト情報は、データ取得部１５１が過去に取得した請求書の請求書データから抽出された少なくとも一部の情報である。仕訳データは、一以上の部署それぞれへの金額の按分率が指定された仕訳データである。 Specifically, the apportionment rate determination unit 154 extracts based on the result of learning from the teacher data including the past text information and at least a part of the journal data acquired in the past corresponding to the past text information. The proportion of the amount of money corresponding to at least a part of the text information extracted by the department 152 to one or more departments is determined. Here, the past text information is at least a part of the information extracted from the invoice data of the invoice acquired in the past by the data acquisition unit 151. The journal data is journal data in which the proportion of the amount of money to each of one or more departments is specified.
　具体的には、按分率決定部１５４は、過去テキスト情報と、過去テキスト情報を複数の部署に按分するか否かを示す情報とを含む第２教師データにより学習した結果に基づいて、テキスト情報の少なくとも一部に対してテキスト情報を複数の部署に按分するか否かを判定する。そして、按分率決定部１５４は、テキスト情報を複数の部署に按分すると判定した場合に、過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された仕訳データの少なくとも一部とを含む第３教師データにより学習した結果に基づいて、テキスト情報の複数の部署それぞれへの按分率を決定する。 Specifically, the apportionment rate determination unit 154 is based on the result of learning from the second teacher data including the past text information and the information indicating whether or not the past text information is apportioned to a plurality of departments. Determine whether to apportion the text information to multiple departments for at least a part of. Then, when the apportionment rate determination unit 154 determines that the text information is apportioned to a plurality of departments, the apportionment rate determination unit 154 includes the past text information and at least a part of the journal data created in the past corresponding to the past text information. Based on the result of learning from the third teacher data, the apportionment rate of the text information to each of the plurality of departments is determined.
　より具体的には、按分率決定部１５４は、抽出部１５２が抽出したテキスト情報の少なくとも一部を入力すると、複数の部署のそれぞれに按分するか否かを判定する按分判定モデルと、抽出部１５２が抽出したテキスト情報の少なくとも一部を入力すると、複数の部署のそれぞれに対応する按分率を出力する按分率決定モデルとを用いて按分率を決定する。 More specifically, the apportionment rate determination unit 154 has an apportionment determination model that determines whether or not to apportion each of a plurality of departments when at least a part of the text information extracted by the extraction unit 152 is input, and an extraction unit. When at least a part of the text information extracted by 152 is input, the apportionment rate is determined by using the apportionment rate determination model that outputs the apportionment rate corresponding to each of the plurality of departments.
　按分判定モデルは、当該モデルへの入力情報と、当該入力情報に対応する出力情報とをセットにした第２教師データにより予め学習することにより生成される。第２教師データに含まれる入力情報は、例えば、過去に取得された請求書データから抽出された過去テキスト情報である。出力情報は、当該入力情報に対し、複数の部署に按分するか否かを示す判定情報である。生成された按分判定モデルは記憶部１４に記憶されている。 The apportionment determination model is generated by learning in advance from the second teacher data in which the input information to the model and the output information corresponding to the input information are set. The input information included in the second teacher data is, for example, past text information extracted from invoice data acquired in the past. The output information is determination information indicating whether or not to apportion the input information to a plurality of departments. The generated apportionment determination model is stored in the storage unit 14.
　按分率決定モデルは、当該モデルへの入力情報と、当該入力情報に対応する出力情報とをセットにした第３教師データにより予め学習することにより生成される。第３教師データに含まれる入力情報は、例えば、過去テキスト情報である。出力情報は、当該入力情報に対応する複数の部署のそれぞれに対応する按分率である。生成された按分率決定モデルは記憶部１４に記憶されている。 The apportionment rate determination model is generated by learning in advance from the third teacher data, which is a set of the input information to the model and the output information corresponding to the input information. The input information included in the third teacher data is, for example, past text information. The output information is a proportional division rate corresponding to each of the plurality of departments corresponding to the input information. The generated apportionment rate determination model is stored in the storage unit 14.
　按分率決定モデルの学習に用いる教師データに含まれる過去テキスト情報には、過去に取得された請求書の請求書データから抽出された、商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額と、請求書の日付を示す日付情報とが含まれている。 The past text information included in the teacher data used for learning the apportionment rate determination model corresponds to the product name or service name extracted from the invoice data of the invoice acquired in the past and the product name or service name. It contains the transaction amount to be processed and date information indicating the date of the invoice.
　按分率決定部１５４は、抽出部１５２が抽出した複数の商品名又はサービス名から、一つの商品名又はサービス名を選択するとともに、選択した商品名又はサービス名に対応する取引金額を特定する。按分率決定部１５４は、請求書の日付を示す日付情報と、選択した商品名又はサービス名と、特定した取引金額とを、記憶部１４に記憶されている按分判定モデルに入力し、按分判定モデルから出力される判定結果に基づいて、按分を行うか否かを判定する。 The apportionment rate determination unit 154 selects one product name or service name from a plurality of product names or service names extracted by the extraction unit 152, and specifies the transaction amount corresponding to the selected product name or service name. The apportionment rate determination unit 154 inputs the date information indicating the date of the invoice, the selected product name or service name, and the specified transaction amount into the apportionment determination model stored in the storage unit 14, and the apportionment determination unit Based on the determination result output from the model, it is determined whether or not to perform proportional division.
　按分率決定部１５４は、按分を行うと判定すると、請求書の日付を示す日付情報と、選択した商品名又はサービス名と、特定した取引金額とを、記憶部１４に記憶されている按分率決定モデルに入力する。そして、按分率決定部１５４は、按分率決定モデルから出力される、複数の部署のそれぞれに対応する按分率を取得することにより、選択した商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額とに対応する一以上の部署それぞれへの按分率を決定する。 When the apportionment rate determination unit 154 determines that the apportionment is to be performed, the apportionment rate storing the date information indicating the date of the invoice, the selected product name or service name, and the specified transaction amount in the storage unit 14. Enter in the decision model. Then, the apportionment rate determination unit 154 obtains the apportionment rate corresponding to each of the plurality of departments output from the apportionment rate determination model, so that the selected product name or service name and the product name or service name can be obtained. Determine the apportionment rate for each of the corresponding transaction amounts and one or more departments.
　なお、按分率決定部１５４は、過去テキスト情報と、過去テキスト情報に対応する取引先に関する情報と、仕訳データとを含む教師データにより学習した結果に基づいて、抽出部１５２が抽出したテキスト情報の少なくとも一部と、当該テキスト情報に対応する取引先に関する情報とに対応する一以上の部署それぞれへの按分率を決定してもよい。 The apportionment rate determination unit 154 is the text information extracted by the extraction unit 152 based on the result of learning from the teacher data including the past text information, the information about the business partner corresponding to the past text information, and the journal data. The apportionment rate to at least a part and one or more departments corresponding to the information about the business partner corresponding to the text information may be determined.
　この場合、按分判定モデル及び按分率決定モデルは、過去テキスト情報と、過去テキスト情報に対応する取引先に関する情報とを入力情報として学習することにより生成される。按分率決定部１５４は、請求書の日付を示す日付情報と、選択した商品名又はサービス名と、特定した取引金額と、請求書に対応する取引先に関する情報とを按分判定モデルに入力し、判定結果を取得する。 In this case, the apportionment determination model and the apportionment rate determination model are generated by learning the past text information and the information about the business partner corresponding to the past text information as input information. The apportionment rate determination unit 154 inputs date information indicating the date of the invoice, the selected product name or service name, the specified transaction amount, and information on the business partner corresponding to the invoice into the apportionment determination model. Acquire the judgment result.
　按分率決定部１５４は、判定結果に基づいて、特定した取引金額を複数の部署に按分すると判定すると、請求書の日付を示す日付情報と、選択した商品名又はサービス名と、特定した取引金額と、請求書に対応する取引先に関する情報とを按分率決定モデルに入力し、按分率決定モデルから出力される一以上の部署それぞれへの按分率を取得することにより、一以上の部署それぞれへの按分率を決定する。このようにすることで、按分率決定部１５４は、新規の取引先に対応する請求書データに対し、適切である蓋然性が高い按分率を決定することができる。 When the apportionment rate determination unit 154 determines that the specified transaction amount is apportioned to a plurality of departments based on the determination result, the date information indicating the date of the invoice, the selected product name or service name, and the specified transaction amount are determined. By inputting information about the business partner corresponding to the invoice into the apportionment rate determination model and acquiring the apportionment rate for each of one or more departments output from the apportionment rate determination model, each of the one or more departments Determine the apportionment rate of. By doing so, the apportionment rate determination unit 154 can determine an appropriate apportionment rate for the invoice data corresponding to the new business partner.
　また、按分率決定部１５４は、過去テキスト情報と、当該過去テキスト情報に対応する仕訳データの入力者に関する情報と、仕訳データとを含む教師データにより学習した結果に基づいて、テキスト情報の少なくとも一部と、当該テキスト情報に対応する仕訳データの入力者とに対応する一以上の部署それぞれへの按分率を決定してもよい。 Further, the apportionment rate determination unit 154 is based on the result of learning from the past text information, the information regarding the input person of the journal data corresponding to the past text information, and the teacher data including the journal data, and at least one of the text information. The apportionment rate to each of one or more departments corresponding to the department and the person who inputs the journal data corresponding to the text information may be determined.
　過去テキスト情報に対応する仕訳データの入力者に関する情報は、例えば、当該入力者を識別するためのユーザ識別情報、当該入力者が所属する部署を識別する部署コード、当該入力者が担当するプロジェクトを識別するためのプロジェクトコードである。按分判定モデル及び按分率決定モデルは、過去テキスト情報と、過去テキスト情報に対応する仕訳データの入力者に関する情報とを入力情報として学習することにより生成される。 Information about the person who entered the journal data corresponding to the past text information includes, for example, user identification information for identifying the person, a department code for identifying the department to which the person belongs, and a project in charge of the person. Project code for identification. The apportionment determination model and the apportionment rate determination model are generated by learning past text information and information about an input person of journal data corresponding to the past text information as input information.
　按分率決定部１５４は、請求書の日付を示す日付情報と、選択した商品名又はサービス名と、特定した取引金額と、当該テキスト情報を入力する入力者のユーザＩＤとを按分判定モデルに入力し、判定結果を取得する。按分率決定部１５４は、判定結果に基づいて複数の部署に按分すると判定すると、請求書の日付を示す日付情報と、選択した商品名又はサービス名と、特定した取引金額と、データ処理装置１を操作しているユーザを識別するためのユーザＩＤとを按分率決定モデルに入力し、按分率決定モデルから出力される一以上の部署それぞれへの按分率を取得することにより、一以上の部署それぞれへの按分率を決定する。このようにすることで、按分率決定部１５４は、仕訳データの入力者、所属部署、所属プロジェクトと、テキスト情報との関係に対応して複数の部署への按分率を決定することができる。 The apportionment rate determination unit 154 inputs the date information indicating the date of the invoice, the selected product name or service name, the specified transaction amount, and the user ID of the input person who inputs the text information into the apportionment determination model. And get the judgment result. When the apportionment rate determination unit 154 determines that the apportionment rate is to be apportioned to a plurality of departments based on the determination result, the date information indicating the date of the invoice, the selected product name or service name, the specified transaction amount, and the data processing device 1 One or more departments by inputting the user ID for identifying the user who is operating the Determine the apportionment rate for each. By doing so, the apportionment rate determination unit 154 can determine the apportionment rate to a plurality of departments according to the relationship between the journal data input person, the department to which the journal data belongs, the project to which the journal data belongs, and the text information.
　また、按分率決定部１５４は、請求書データから抽出されたテキスト情報が、予め設定されている按分率を決定するための按分率決定ルールを示す按分率ルール情報に対応している場合には、按分率ルール情報に基づいて按分率を決定し、按分率ルール情報に対応していない場合に、按分判定モデル及び按分率決定モデルを用いて按分率を決定してもよい。 Further, when the text information extracted from the invoice data corresponds to the apportionment rate rule information indicating the apportionment rate determination rule for determining the apportionment rate set in advance, the apportionment rate determination unit 154 corresponds to the apportionment rate rule information. , The apportionment rate may be determined based on the apportionment rate rule information, and when the apportionment rate rule information is not supported, the apportionment rate may be determined using the apportionment determination model and the apportionment rate determination model.
　この場合、記憶部１４は、請求書データから抽出されたテキスト情報の少なくとも一部と、一以上の部署それぞれへの按分率とを関連付けた按分率ルール情報を記憶する。テキスト情報の少なくとも一部は、例えば、請求書に対応する取引先を示す取引先情報と、請求書の日付を示す日付情報と、請求書に含まれる商品又はサービスとの組み合わせである。 In this case, the storage unit 14 stores at least a part of the text information extracted from the invoice data and the apportionment rate rule information in which the apportionment rate to each of one or more departments is associated with each other. At least a portion of the textual information is, for example, a combination of customer information indicating the business partner corresponding to the invoice, date information indicating the date of the invoice, and goods or services contained in the invoice.
　按分率決定部１５４は、請求書データに対応する取引先情報と、抽出部１５２が抽出した請求書の日付情報と、選択した商品名又はサービス名とが関連付けられて、記憶部１４に記憶されている按分率ルール情報に記憶されているか否かを判定することにより、これらの情報が按分率決定ルールに対応しているか否かを判定する。 The apportionment rate determination unit 154 stores the customer information corresponding to the invoice data, the date information of the invoice extracted by the extraction unit 152, and the selected product name or service name in association with each other in the storage unit 14. By determining whether or not the information is stored in the apportionment rate rule information, it is determined whether or not these information correspond to the apportionment rate determination rule.
　そして、按分率決定部１５４は、取引先情報と、日付情報と、選択した商品名又はサービス名が按分率決定ルールに対応している場合、按分率ルール情報に基づいて、一以上の部署それぞれへの按分率を決定する。按分率決定部１５４は、按分率ルール情報において、請求書データに対応する取引先情報、日付情報、及び、選択された商品名又はサービス名に関連付けられている一以上の部署への按分率を、当該商品名又はサービス名に対応する一以上の部署への按分率に決定する。このようにすることで、按分率決定部１５４は、過去に取引したことがある取引先から受領した請求書に対して、按分率を精度良く特定することができる。 Then, the apportionment rate determination unit 154, when the customer information, the date information, and the selected product name or service name correspond to the apportionment rate determination rule, each of one or more departments based on the apportionment rate rule information. Determine the apportionment rate to. In the apportionment rate rule information, the apportionment rate determination unit 154 determines the apportionment rate to the customer information corresponding to the invoice data, the date information, and the apportionment rate to one or more departments associated with the selected product name or service name. , Determine the proportion to one or more departments corresponding to the product name or service name. By doing so, the apportionment rate determination unit 154 can accurately specify the apportionment rate for the invoice received from the business partner who has made a transaction in the past.
　仕訳データ作成部１５５は、勘定科目決定部１５３が決定した勘定科目と、按分率決定部１５４が決定した按分率とを出力する。例えば、仕訳データ作成部１５５は、抽出部１５２が抽出した一以上の商品又はサービスのそれぞれに対して、抽出部１５２が抽出した日付情報と、勘定科目決定部１５３が決定した勘定科目と、按分率決定部１５４が決定した按分率と、按分先の部署名とを含む仕訳データを作成する。 The journal data creation unit 155 outputs the account item determined by the account item determination unit 153 and the apportionment rate determined by the apportionment rate determination unit 154. For example, the journal data creation unit 155 prorates the date information extracted by the extraction unit 152 and the account items determined by the account item determination unit 153 for each of the one or more goods or services extracted by the extraction unit 152. Journal data including the apportionment rate determined by the rate determination unit 154 and the department name of the apportionment destination is created.
　仕訳データ作成部１５５は、作成した仕訳データの会計システムへの登録を受け付ける登録画面を表示部１３に出力する。図５は、仕訳データ作成部１５５が出力した登録画面の一例を示す図である。図５に示す仕訳データは、図３に対応する請求書に基づいて生成された仕訳データである。図５に示す例では、仕訳データの他に、登録者として、ユーザの氏名が表示されているとともに、仕訳データの会計システムへの登録を受け付ける登録ボタンとが表示されていることが確認できる。 The journal data creation unit 155 outputs a registration screen for accepting registration of the created journal data to the accounting system to the display unit 13. FIG. 5 is a diagram showing an example of a registration screen output by the journal data creation unit 155. The journal data shown in FIG. 5 is the journal data generated based on the invoice corresponding to FIG. In the example shown in FIG. 5, in addition to the journal data, it can be confirmed that the name of the user as the registrant is displayed and the registration button for accepting the registration of the journal data in the accounting system is displayed.
　登録部１５６は、登録画面に表示された仕訳データを会計システムに登録する。登録部１５６は、登録画面に表示されている仕訳データの修正をユーザから受け付けてもよい。登録部１５６は、登録画面において登録ボタンが押下されたことに応じて、登録画面に示されている仕訳データを会計システム３に登録するとともに、当該仕訳データを仕訳データベースに格納する。 The registration unit 156 registers the journal data displayed on the registration screen in the accounting system. The registration unit 156 may accept the correction of the journal data displayed on the registration screen from the user. The registration unit 156 registers the journal data shown on the registration screen in the accounting system 3 and stores the journal data in the journal database in response to the pressing of the registration button on the registration screen.
　学習部１５７は、仕訳データ作成部１５５が出力した按分率に対応し、登録部１５６により会計システム３に登録された仕訳データの少なくとも一部と、当該仕訳データに対応する過去テキスト情報の少なくとも一部とを含む教師データにより、勘定科目決定モデル、按分判定モデル、按分率決定モデルの学習を行う。例えば、学習部１５７は、登録部１５６が仕訳データベースに情報を格納したことに応じて、新たに格納された情報を教師データとして、勘定科目決定モデル、按分判定モデル、按分率決定モデルの学習を行う。このようにすることで、データ処理装置１においては、仕訳データが会計システム３に登録されればされるほどモデルが学習することができるので、モデルから出力される勘定科目、按分するか否かの判定結果、及び按分率の精度を高めることができる。 The learning unit 157 corresponds to the apportionment rate output by the journal data creation unit 155, and at least a part of the journal data registered in the accounting system 3 by the registration unit 156 and at least one of the past text information corresponding to the journal data. The account item determination model, the apportionment judgment model, and the apportionment rate determination model are learned from the teacher data including the department. For example, the learning unit 157 learns the account item determination model, the apportionment determination model, and the apportionment rate determination model using the newly stored information as teacher data in response to the information stored in the journal entry database by the registration unit 156. conduct. By doing so, in the data processing device 1, the more the journal data is registered in the accounting system 3, the more the model can be learned. Therefore, the account output from the model is apportioned or not. It is possible to improve the accuracy of the determination result and the proportional division rate.
［データ処理装置１における処理の流れ］
　図６は、データ処理装置１における処理の流れの一例を示すフローチャートである。
　まず、データ取得部１５１は、読取装置２が請求書を読み取ることによって生成した請求書データを取得する（Ｓ１）。
　続いて、抽出部１５２は、データ取得部１５１が取得した請求書データからテキスト情報を抽出する（Ｓ２）。 [Process flow in data processing device 1]
FIG. 6 is a flowchart showing an example of the processing flow in the data processing apparatus 1.
First, the data acquisition unit 151 acquires the invoice data generated by the reading device 2 reading the invoice (S1).
Subsequently, the extraction unit 152 extracts text information from the invoice data acquired by the data acquisition unit 151 (S2).
　続いて、勘定科目決定部１５３は、抽出部１５２が抽出したテキスト情報に含まれる複数の商品又はサービスの中から、一度も選択されていない商品名又はサービス名を選択する（Ｓ３）。また、勘定科目決定部１５３は、抽出部１５２が抽出したテキスト情報に含まれる複数の取引金額の中から、選択した商品名又はサービス名に対応する取引金額を特定する（Ｓ４）。 Subsequently, the account item determination unit 153 selects a product name or service name that has never been selected from a plurality of products or services included in the text information extracted by the extraction unit 152 (S3). Further, the account item determination unit 153 specifies the transaction amount corresponding to the selected product name or service name from the plurality of transaction amounts included in the text information extracted by the extraction unit 152 (S4).
　続いて、勘定科目決定部１５３は、請求書の日付情報と、選択した商品名又はサービス名と、特定した取引金額とを勘定科目決定モデルに入力し、勘定科目決定モデルから出力された勘定科目を取得することにより、勘定科目を決定する（Ｓ５）。 Subsequently, the account determination unit 153 inputs the date information of the invoice, the selected product name or service name, and the specified transaction amount into the account determination model, and the account output from the account determination model. The account item is determined by acquiring (S5).
　続いて、按分率決定部１５４は、請求書の日付情報と、選択した商品名又はサービス名と、特定した取引金額とに対応する按分率を決定する（Ｓ６）。具体的には、按分率決定部１５４は、請求書の日付情報と、選択した商品名又はサービス名と、特定した取引金額とを按分判定モデルに入力し、按分判定モデルから出力された判定結果が按分を行うか否かを判定する。按分率決定部１５４は、按分を行うと判定すると、請求書の日付情報と、選択した商品名又はサービス名と、特定した取引金額とを按分率決定モデルに入力し、按分率決定モデルから出力された複数の部署へのそれぞれの按分率を取得することにより、複数の部署のそれぞれへの按分率を決定する。 Subsequently, the apportionment rate determination unit 154 determines the apportionment rate corresponding to the date information of the invoice, the selected product name or service name, and the specified transaction amount (S6). Specifically, the apportionment rate determination unit 154 inputs the date information of the invoice, the selected product name or service name, and the specified transaction amount into the apportionment determination model, and the determination result output from the apportionment determination model. Determines whether or not to perform proportional division. When the apportionment rate determination unit 154 determines that the apportionment is to be performed, the date information of the invoice, the selected product name or service name, and the specified transaction amount are input to the apportionment rate determination model and output from the apportionment rate determination model. By acquiring the apportionment rate for each of the plurality of departments, the apportionment rate for each of the plurality of departments is determined.
　続いて、制御部１５は、抽出部１５２が抽出したテキスト情報に含まれる複数の商品又はサービスを全て選択したか否かを判定する（Ｓ７）。制御部１５は、複数の商品又はサービスを全て選択したと判定すると、Ｓ８に処理を移し、複数の商品又はサービスを全て選択していないと判定すると、Ｓ３に処理を移す。 Subsequently, the control unit 15 determines whether or not all of the plurality of products or services included in the text information extracted by the extraction unit 152 have been selected (S7). When the control unit 15 determines that all the plurality of products or services have been selected, the process is transferred to S8, and when it is determined that all the plurality of products or services have not been selected, the control unit 15 transfers the process to S3.
　続いて、仕訳データ作成部１５５は、抽出部１５２が抽出した一以上の商品又はサービスのそれぞれに対応して、抽出部１５２が抽出した日付情報と、勘定科目決定部１５３が決定した勘定科目と、按分率決定部１５４が決定した按分率と、按分先の部署名とを含む仕訳データを作成し、当該仕訳データの会計システムへの登録を受け付ける登録画面を表示部１３に出力する（Ｓ８）。 Subsequently, the journal data creation unit 155 includes the date information extracted by the extraction unit 152 and the account items determined by the account item determination unit 153, corresponding to each of the one or more products or services extracted by the extraction unit 152. , Creates journal data including the apportionment rate determined by the apportionment rate determination unit 154 and the department name of the apportionment destination, and outputs a registration screen for accepting registration of the journal data to the accounting system to the display unit 13 (S8). ..
　続いて、登録部１５６は、登録画面に表示された仕訳データの修正をユーザから適宜受け付けた後、登録画面に示されている仕訳データを会計システムに登録する（Ｓ９）。
　続いて、学習部１５７は、登録部１５６により会計システム３に登録された仕訳データの少なくとも一部と、当該仕訳データに対応する過去テキスト情報の少なくとも一部とを含む教師データにより、勘定科目決定モデル、按分判定モデル、按分率決定モデルのさらなる学習を行う（Ｓ１０）。 Subsequently, the registration unit 156 appropriately accepts corrections of the journal data displayed on the registration screen from the user, and then registers the journal data displayed on the registration screen in the accounting system (S9).
Subsequently, the learning unit 157 determines the account item based on the teacher data including at least a part of the journal data registered in the accounting system 3 by the registration unit 156 and at least a part of the past text information corresponding to the journal data. Further learning of the model, the apportionment determination model, and the apportionment rate determination model is performed (S10).
［データ処理装置１による効果］
　以上説明したように、データ処理装置１は、過去に取得された証憑としての請求書の請求書データから抽出された少なくとも一部の過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された、一以上の部署それぞれへの金額の按分率が指定された仕訳データとに基づいて、抽出部１５２が抽出したテキスト情報の少なくとも一部に対応する金額の一以上の部署それぞれへの按分率を決定し、決定した按分率を出力する。このようにすることで、データ処理装置１は、新規の請求書データに対し、複数の部署への按分に対応した仕訳データを作成することができる。 [Effect of data processing device 1]
As described above, the data processing device 1 has created at least a part of the past text information extracted from the invoice data of the invoice as a voucher acquired in the past and the past text information corresponding to the past text information. Based on the journal data for which the apportionment rate of the amount to each of one or more departments is specified, the apportionment of the amount corresponding to at least a part of the text information extracted by the extraction unit 152 to each of one or more departments. Determine the rate and output the determined apportionment rate. By doing so, the data processing device 1 can create journal data corresponding to the apportionment to a plurality of departments for the new invoice data.
　以上、本発明を実施の形態を用いて説明したが、本発明の技術的範囲は上記実施の形態に記載の範囲には限定されず、その要旨の範囲内で種々の変形及び変更が可能である。例えば、上記実施の形態では、按分率決定部１５４は、按分判定モデルと按分率決定モデルとを用いて一以上の部署のそれぞれへの按分率を決定したが、これに限らない。例えば、按分率決定部１５４は、抽出部１５２が抽出したテキスト情報の入力に対して一以上の部署への按分率を出力する按分率決定モデルを用いて、一以上の部署のそれぞれへの按分率を決定してもよい。 Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes can be made within the scope of the gist. be. For example, in the above embodiment, the apportionment rate determination unit 154 determines the apportionment rate for each of one or more departments using the apportionment determination model and the apportionment rate determination model, but the present invention is not limited to this. For example, the apportionment rate determination unit 154 uses an apportionment rate determination model that outputs the apportionment rate to one or more departments in response to the input of the text information extracted by the extraction unit 152, and apportionments to each of the one or more departments. The rate may be determined.
　また、例えば、装置の全部又は一部は、任意の単位で機能的又は物理的に分散・統合して構成することができる。また、複数の実施の形態の任意の組み合わせによって生じる新たな実施の形態も、本発明の実施の形態に含まれる。組み合わせによって生じる新たな実施の形態の効果は、もとの実施の形態の効果を併せ持つ。 Further, for example, all or a part of the device can be functionally or physically distributed / integrated in any unit. Also included in the embodiments of the present invention are new embodiments resulting from any combination of the plurality of embodiments. The effect of the new embodiment produced by the combination has the effect of the original embodiment together.
１　データ処理装置
２　読取装置
３　会計システム
１１　通信部
１２　操作部
１３　表示部
１４　記憶部
１５　制御部
１５１　データ取得部
１５２　抽出部
１５３　勘定科目決定部
１５４　按分率決定部
１５５　仕訳データ作成部
１５６　登録部
１５７　学習部
  1 Data processing device 2 Reading device 3 Accounting system 11 Communication unit 12 Operation unit 13 Display unit 14 Storage unit 15 Control unit 151 Data acquisition unit 152 Extraction unit 153 Account item determination unit 154 Proportional division rate determination unit 155 Journal data creation unit 156 Registration unit 157 Learning Department

Claims (11)

  1. 　証憑データを取得するデータ取得部と、
    　前記証憑データから、テキスト情報を抽出する抽出部と、
    　過去に取得された証憑の証憑データから抽出された少なくとも一部の過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された、一以上の部署それぞれへの金額の按分率が指定された仕訳データとに基づいて、前記テキスト情報の少なくとも一部に対応する金額の前記一以上の部署それぞれへの按分率を決定する按分率決定部と、
    　決定された前記按分率を出力する出力部と、
    　を有するデータ処理装置。     The data acquisition department that acquires voucher data, and
    An extraction unit that extracts text information from the voucher data,
    At least a part of the past text information extracted from the voucher data of the voucher acquired in the past and the apportionment rate of the amount to one or more departments created in the past corresponding to the past text information are specified. Based on the journal data, the apportionment rate determination unit that determines the apportionment rate of the amount corresponding to at least a part of the text information to each of the one or more departments, and
    An output unit that outputs the determined apportionment rate,
    Data processing device with.
  2. 　前記過去テキスト情報には、商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額とが含まれており、
    　前記按分率決定部は、前記テキスト情報に含まれる商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額とに対応する前記一以上の部署それぞれへの按分率を決定する、
    　請求項１に記載のデータ処理装置。     The past text information includes a product name or a service name and a transaction amount corresponding to the product name or the service name.
    The apportionment rate determination unit determines the apportionment rate for each of the one or more departments corresponding to the product name or service name included in the text information and the transaction amount corresponding to the product name or service name.
    The data processing apparatus according to claim 1.
  3. 　前記過去テキスト情報には、日付情報が含まれており、
    　前記按分率決定部は、前記テキスト情報に含まれる日付に対応する前記一以上の部署それぞれへの按分率を決定する、
    　請求項１又は２に記載のデータ処理装置。     The past text information includes date information.
    The apportionment rate determination unit determines the apportionment rate for each of the one or more departments corresponding to the dates included in the text information.
    The data processing apparatus according to claim 1 or 2.
  4. 　前記按分率決定部は、前記過去テキスト情報に対応する取引先に関する情報と、前記仕訳データとに基づいて、前記テキスト情報の少なくとも一部と、前記テキスト情報に対応する取引先に関する情報とに対応する前記一以上の部署それぞれへの按分率を決定する、
    　請求項１から３のいずれか一項に記載のデータ処理装置。     The apportionment rate determination unit corresponds to at least a part of the text information and information about the business partner corresponding to the text information based on the information about the business partner corresponding to the past text information and the journal data. Determine the apportionment rate for each of the above-mentioned one or more departments,
    The data processing apparatus according to any one of claims 1 to 3.
  5. 　前記按分率決定部は、前記過去テキスト情報と、前記仕訳データの入力者に関する情報と、前記仕訳データとに基づいて、前記テキスト情報の少なくとも一部と、前記テキスト情報に対応する前記仕訳データの入力者とに対応する前記一以上の部署それぞれへの按分率を決定する、
    　請求項１から４のいずれか一項に記載のデータ処理装置。     Based on the past text information, the information about the person who entered the journal data, and the journal data, the apportionment rate determination unit has at least a part of the text information and the journal data corresponding to the text information. Determine the apportionment rate for each of the above-mentioned one or more departments corresponding to the input person,
    The data processing apparatus according to any one of claims 1 to 4.
  6. 　前記按分率決定部は、前記過去テキスト情報と、当該過去テキスト情報に対応して前記データ処理装置に過去に登録された前記仕訳データとを含む教師データにより学習した結果に基づいて、前記テキスト情報に含まれる商品名又はサービス名と、当該商品名又はサービス名に対応する取引金額とに対応する前記一以上の部署それぞれへの按分率を決定する、
    　請求項１から５のいずれか一項に記載のデータ処理装置。     The apportionment rate determination unit is based on the result of learning from the teacher data including the past text information and the journal data registered in the data processing device in the past corresponding to the past text information. Determine the apportionment rate to each of the above-mentioned one or more departments corresponding to the product name or service name included in the above and the transaction amount corresponding to the product name or service name.
    The data processing apparatus according to any one of claims 1 to 5.
  7. 　前記出力部が出力した前記按分率に対応して会計システムに登録された仕訳データの少なくとも一部と、当該仕訳データに対応する前記過去テキスト情報の少なくとも一部とを含む教師データにより前記学習を行う学習部をさらに有する、
    　請求項６のいずれか一項に記載のデータ処理装置。     The learning is performed by the teacher data including at least a part of the journal data registered in the accounting system corresponding to the proportional division rate output by the output unit and at least a part of the past text information corresponding to the journal data. Has more learning departments to do,
    The data processing apparatus according to any one of claims 6.
  8. 　前記按分率決定部は、抽出された前記テキスト情報が、予め設定されている按分率を決定するためのルールに対応していない場合、教師データにより学習した結果に基づいて、前記テキスト情報の少なくとも一部に対応する前記一以上の部署それぞれへの按分率を決定し、抽出された前記テキスト情報が前記ルールに対応している場合、前記ルールに基づいて前記一以上の部署それぞれへの按分率を決定する、
    　請求項６又は７に記載のデータ処理装置。     When the extracted text information does not correspond to a preset rule for determining the apportionment rate, the apportionment rate determination unit at least the text information based on the result learned from the teacher data. When the apportionment rate to each of the one or more departments corresponding to a part is determined and the extracted text information corresponds to the rule, the apportionment rate to each of the one or more departments is based on the rule. To decide,
    The data processing apparatus according to claim 6 or 7.
  9. 　前記按分率決定部は、前記過去テキスト情報と、前記過去テキスト情報を複数の部署に按分するか否かを示す情報とを含む第２教師データにより学習した結果に基づいて、前記テキスト情報の少なくとも一部に対して前記テキスト情報を複数の部署に按分するか否かを判定し、前記テキスト情報を複数の部署に按分すると判定した場合に、前記教師データにより学習した結果に基づいて、前記テキスト情報の複数の部署それぞれへの按分率を決定する、
    　請求項６から８のいずれか一項に記載のデータ処理装置。     The apportionment rate determination unit has at least the text information based on the result of learning from the second teacher data including the past text information and information indicating whether or not the past text information is apportioned to a plurality of departments. When it is determined whether or not to apportion the text information to a plurality of departments for a part and it is determined to apportion the text information to a plurality of departments, the text is based on the result learned from the teacher data. Determine the proportion of information to each of multiple departments,
    The data processing apparatus according to any one of claims 6 to 8.
  10. 　コンピュータが実行する、
    　証憑データを取得するステップと、
    　前記証憑データから、テキスト情報を抽出するステップと、
    　過去に取得された証憑の証憑データから抽出された少なくとも一部の過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された、一以上の部署それぞれへの金額の按分率が指定された仕訳データとに基づいて、前記テキスト情報の少なくとも一部に対応する金額の前記一以上の部署それぞれへの按分率を決定するステップと、
    　決定された前記按分率を出力するステップと、
    　を有するデータ処理方法。     Computer runs,
    Steps to get voucher data and
    Steps to extract text information from the voucher data,
    At least a part of the past text information extracted from the voucher data of the voucher acquired in the past and the apportionment rate of the amount to one or more departments created in the past corresponding to the past text information are specified. A step of determining the apportionment rate of the amount corresponding to at least a part of the text information to each of the one or more departments based on the journal data.
    The step of outputting the determined apportionment rate and
    Data processing method.
  11. 　コンピュータを、
    　証憑データを取得するデータ取得部、
    　前記証憑データから、テキスト情報を抽出する抽出部、
    　過去に取得された証憑の証憑データから抽出された少なくとも一部の過去テキスト情報と、当該過去テキスト情報に対応して過去に作成された、一以上の部署それぞれへの金額の按分率が指定された仕訳データとに基づいて、前記テキスト情報の少なくとも一部に対応する金額の前記一以上の部署それぞれへの按分率を決定する按分率決定部、及び、
    　決定された前記按分率を出力する出力部、
    　として機能させるプログラム。
          Computer,
    Data acquisition department to acquire voucher data,
    Extractor that extracts text information from the voucher data,
    At least a part of the past text information extracted from the voucher data of the voucher acquired in the past and the apportionment rate of the amount to one or more departments created in the past corresponding to the past text information are specified. Based on the journal data, the apportionment rate determination unit that determines the apportionment rate of the amount corresponding to at least a part of the text information to each of the one or more departments, and the apportionment rate determination unit.
    An output unit that outputs the determined apportionment rate,
    A program that functions as.
PCT/JP2020/034103 2020-09-09 2020-09-09 Data processing device, data processing method, and program WO2022054165A1 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
PCT/JP2020/034103 WO2022054165A1 (en) 2020-09-09 2020-09-09 Data processing device, data processing method, and program
JP2020547247A JP6810306B1 (en) 2020-09-09 2020-09-09 Data processing equipment, data processing methods and programs
JP2020205042A JP2022045868A (en) 2020-09-09 2020-12-10 Data processing apparatus, data processing method, and program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/034103 WO2022054165A1 (en) 2020-09-09 2020-09-09 Data processing device, data processing method, and program

Publications (1)

Publication Number Publication Date
WO2022054165A1 true WO2022054165A1 (en) 2022-03-17

Family

ID=73992972

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/034103 WO2022054165A1 (en) 2020-09-09 2020-09-09 Data processing device, data processing method, and program

Country Status (2)

Country Link
JP (2) JP6810306B1 (en)
WO (1) WO2022054165A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7233794B1 (en) * 2022-10-21 2023-03-07 ファーストアカウンティング株式会社 Information processing device, information processing method and program

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006195570A (en) * 2005-01-11 2006-07-27 Hitachi Information Systems Ltd Proportional division processing system, method and program of accounting slip
JP2016181254A (en) * 2015-03-23 2016-10-13 株式会社オービック Automatic journalizing processing apparatus, automatic journalizing processing method, and automatic journalizing processing program
WO2018179411A1 (en) * 2017-03-31 2018-10-04 ファーストアカウンティング株式会社 Accounting system and accounting method
JP2019174949A (en) * 2018-03-27 2019-10-10 株式会社オービック Common expenses management device, common expenses management method and common expenses management program
JP2020057186A (en) * 2018-10-02 2020-04-09 フードゲート株式会社 Accounting software and system
JP6712738B1 (en) * 2019-10-31 2020-06-24 株式会社日本デジタル研究所 Voucher judging device, accounting processor, voucher judging program, voucher judging system and voucher judging method
JP6732325B1 (en) * 2020-04-21 2020-07-29 ファーストアカウンティング株式会社 Accounting system, accounting method, accounting program

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006195570A (en) * 2005-01-11 2006-07-27 Hitachi Information Systems Ltd Proportional division processing system, method and program of accounting slip
JP2016181254A (en) * 2015-03-23 2016-10-13 株式会社オービック Automatic journalizing processing apparatus, automatic journalizing processing method, and automatic journalizing processing program
WO2018179411A1 (en) * 2017-03-31 2018-10-04 ファーストアカウンティング株式会社 Accounting system and accounting method
JP2019174949A (en) * 2018-03-27 2019-10-10 株式会社オービック Common expenses management device, common expenses management method and common expenses management program
JP2020057186A (en) * 2018-10-02 2020-04-09 フードゲート株式会社 Accounting software and system
JP6712738B1 (en) * 2019-10-31 2020-06-24 株式会社日本デジタル研究所 Voucher judging device, accounting processor, voucher judging program, voucher judging system and voucher judging method
JP6732325B1 (en) * 2020-04-21 2020-07-29 ファーストアカウンティング株式会社 Accounting system, accounting method, accounting program

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7233794B1 (en) * 2022-10-21 2023-03-07 ファーストアカウンティング株式会社 Information processing device, information processing method and program
WO2024084692A1 (en) * 2022-10-21 2024-04-25 ファーストアカウンティング株式会社 Information processing device, information processing method, and program

Also Published As

Publication number Publication date
JP2022045868A (en) 2022-03-22
JP6810306B1 (en) 2021-01-06
JPWO2022054165A1 (en) 2022-03-17

Similar Documents

Publication Publication Date Title
US7416131B2 (en) Electronic transaction processing server with automated transaction evaluation
JP5204208B2 (en) Receipt data collation support device and receipt data collation support program
JP6635563B1 (en) Journal element analysis device, accounting processing system, journal element analysis method, journal element analysis program
JPWO2019008766A1 (en) Voucher processing system and voucher processing program
JP6646308B1 (en) Voucher analysis device, accounting processing system, voucher analysis method, voucher analysis program
WO2022054165A1 (en) Data processing device, data processing method, and program
JP6464440B1 (en) Accounting processing apparatus, accounting processing system, accounting processing method and program
JP6732325B1 (en) Accounting system, accounting method, accounting program
CN109544134B (en) Convenient payment service method and system
JP6810303B1 (en) Data processing equipment, data processing method and data processing program
CN116051092A (en) Cross-system automatic account checking method and device
WO2002054307A1 (en) Electronic shop management system
JP6944603B1 (en) Data processing equipment, data processing methods and programs
JP2001236354A (en) Account system with automatic journal processing function
JP6844076B1 (en) Data processing equipment, data processing methods and programs
WO2022049689A1 (en) Data processing device, data processing method, and program
JP6967174B1 (en) Data processing equipment, data processing methods and programs
WO2022176100A1 (en) Data processing device, data processing method, and program
WO2022102039A1 (en) Data processing device, data processing method, and program
JP7021819B1 (en) Data processing equipment, data processing methods and programs
JP7063963B2 (en) Payment support system, payment support method and payment support program
WO2018221146A1 (en) Account title guessing program, account title guessing device, and account title guessing method
WO2020050391A1 (en) System and method for displaying journalization
JP2002230255A (en) Accounting processing system with support function for journal input
JP3066021U (en) Computer for calculating business performance by computer

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2020547247

Country of ref document: JP

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20953231

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20953231

Country of ref document: EP

Kind code of ref document: A1