Accounting business assist method and accounting business assist system

Abstract

Provided is an accounting business assist method using a language model. The accounting business assist method includes a first process for extracting an electronic mail related to an order, and a second process for extracting information from the electronic mail related to the order. The first process and the second process are executed by using the language model. The accounting business assist method also includes a process for narrowing down electronic mails by using a search expression before the first process, and a process for registering, in a database, the information extracted in the second process.

Description

Accounting work support method, accounting work support system

[0001] One aspect of the present invention relates to an accounting work support method and an accounting work support system, and more particularly to an accounting work support method and an accounting work support system that utilize a language model.

[0002] One embodiment of the present invention is not limited to the above technical field, and examples of the technical field of one embodiment of the present invention include semiconductor devices, display devices, light-emitting devices, power storage devices, memory devices, electronic devices, lighting devices, input devices (e.g., touch sensors), input / output devices (e.g., touch panels), driving methods thereof, and manufacturing methods thereof.

[0003] In recent years, negotiations or consultations with business partners regarding product transactions are often conducted via email. Furthermore, documents related to orders are sometimes attached to emails. Therefore, to find emails related to an order, it is necessary to search through emails that have been sent or received. Patent Document 1 discloses an email management system that displays sorting marks for sorting sent and received emails, allowing users to easily and conveniently find desired emails.

[0004] In recent years, the development of language models using neural networks has been actively pursued, with large-scale language models (LLMs) attracting particular attention. A large-scale language model is a natural language processing model trained using a large amount of data. A large-scale language model can realize, for example, a dialogue model that responds to user instructions. Non-Patent Document 1 discloses GPT-4 (registered trademark) (Generative Pre-trained Transformer 4) as a large-scale language model, and ChatGPT as a dialogue model.

[0005] The use of large-scale language models has significantly increased the capabilities of natural language processing models. However, as language models become larger, it is difficult to incorporate and operate language models in-house due to the equipment and cost involved. Therefore, one way to use language models is to use external services that provide language models.

[0006] JP 2017-204175 A

[0007] Summary of ChatGPT / GPT-4 Research and Perspective Towards the Future of Large Language Models, Yiheng Liu et al. (Submitted on 4 Apr 2023, [online], Internet <URL: https: / / arxiv.org / abs / 2304.01852>

[0008] In order processing, after negotiations or consultations regarding product transactions are conducted with a business partner, order information is created and registered in a database. Furthermore, negotiations or consultations with business partners are often conducted via email. In this case, if the order information registered in the database includes the details of the negotiations or consultations with the business partner, it can serve as a reference for future negotiations or consultations with the business partner. Therefore, it is preferable to register emails related to orders in a database. However, because order-related emails may be sent and received multiple times, the period during which order-related emails are sent and received varies for each order, and multiple people may be involved in a single order, it is difficult to appropriately extract order-related emails from a list of emails sent and received to date.

[0009] Furthermore, because accounting work is performed based on order information registered in a database, order information must be registered appropriately in the database. However, it takes a great deal of effort for a person in charge to select and select information from multiple order-related emails and register it in the database. Furthermore, since the input of information that is optional to register in the database is dependent on the person in charge, some information may not be registered in the database.

[0010] Therefore, an object of one aspect of the present invention is to provide an accounting work support method or accounting work support system that uses a language model.An object of one aspect of the present invention is to provide an accounting work support method or accounting work support system that uses a language model to extract emails related to orders.An object of one aspect of the present invention is to provide an accounting work support method or accounting work support system that uses a language model to select and register information in a database.An object of one aspect of the present invention is to provide an accounting work support method or accounting work support system that reduces variation in the amount of information registered in the database depending on the person in charge.An object of one aspect of the present invention is to provide a highly convenient accounting work support method or accounting work support system.An object of one aspect of the present invention is to provide a novel accounting work support method or accounting work support system.

[0011] Note that the description of these problems does not preclude the existence of other problems. One embodiment of the present invention does not necessarily have to solve all of these problems. Problems other than these can be extracted from the description in the specification, drawings, and claims.

[0012] One aspect of the present invention is an accounting work support method having first to tenth steps. In the first step, an information processing device accepts order information. In the second step, the information processing device registers the order information in a data table. In the third step, the information processing device searches emails based on the order information to obtain multiple first emails. In the fourth step, the information processing device creates a first prompt. The first prompt has a list and a first instruction. The list has first text included in each of the multiple first emails. The first instruction is an instruction for determining whether the first text is related to the order information. In the fifth step, the information processing device transmits the first prompt to a language model via a network to obtain a first response sentence including a result of the determination. In the sixth step, the information processing device extracts at least one second email from the multiple first emails based on the first response sentence. The second email is an email determined to be related to the order information based on the first text. In a seventh step, the information processing device adds at least one second email to a record in the data table in which the order information is registered. In an eighth step, the information processing device generates a second prompt. The second prompt has second text contained in the at least one second email and a second instruction. The second instruction is an instruction for extracting information corresponding to each item in the data table from the second text. In a ninth step, the information processing device transmits the second prompt to a language model via a network to obtain a second response sentence including the extracted information. In a tenth step, the information processing device updates the record using the extracted information based on the second response sentence.

[0013] In the accounting work support method, it is preferable that the first instruction sentence includes at least a part of the order information.

[0014] Preferably, the accounting support method includes a step between the fourth step and the fifth step in which, if a file is attached to the first email, the information processing device extracts third text from the file, and the list includes the third text.

[0015] In the above accounting work support method, if the order information registered in the record does not match the information extracted in the ninth step, it is preferable that the mismatched order information be highlighted and displayed on the display unit of the information terminal.

[0016] It is preferable that the accounting work support method includes, between the fourth step and the fifth step, a step in which the information processing device determines whether the first prompt contains a word that is personal information or confidential information, and replaces the word contained in the first prompt with another word that belongs to the same field as the word to which it belongs.

[0017] One aspect of the present invention is a program having a function of causing a processor to execute any one of the accounting work support methods described above.

[0018] One aspect of the present invention is an accounting information support system having a database, a reception unit, and a processing unit. The database has a data table. The reception unit has a function of receiving order information. The processing unit is configured to execute the following processes: registering the order information in the data table; retrieving multiple first emails by searching emails based on the order information; creating a first prompt; transmitting the first prompt to a language model via a network to retrieve a first response sentence; extracting at least one second email from the multiple first emails based on the first response sentence; adding the at least one second email to a record in the data table where the order information is registered; generating a second prompt; transmitting the second prompt to the language model via the network to retrieve the second response sentence; and updating the record based on the second response sentence. The first prompt has a list and a first instruction sentence. The list has first text included in each of the multiple first emails. The first instruction is an instruction for determining whether the first text is related to order information. The first response includes the result of the determination. The second email is an email determined to be related to order information based on the first text data. The second prompt includes at least one second text included in the second email and a second instruction. The second instruction is an instruction for extracting information corresponding to each item in the data table from the second text. The process of updating the record is performed using the extracted information included in the second response.

[0019] In the above accounting information support system, it is preferable that the first instruction sentence includes at least a part of the order information.

[0020] In the above-mentioned accounting information support system, it is preferable that the processing unit is configured to execute a process to extract third text from the file if the file is attached to the first email, and the list includes the third text.

[0021] It is preferable that the accounting information support system has an information processing device having a database, a reception unit, and a processing unit, and an information terminal, and that the order information registered in the record is displayed on the display unit of the information terminal.

[0022] In the above-mentioned accounting information support system, if the order information does not match the extracted information contained in the second response sentence, it is preferable that the mismatched order information be highlighted and displayed on the display unit of the information terminal.

[0023] One aspect of the present invention provides an accounting work support method or system that utilizes a language model. One aspect of the present invention provides an accounting work support method or system that utilizes a language model to extract emails related to orders. One aspect of the present invention provides an accounting work support method or system that utilizes a language model to select and register information in a database. One aspect of the present invention provides an accounting work support method or system that reduces variation in the amount of information registered in a database depending on the person in charge. One aspect of the present invention provides a highly convenient accounting work support method or system. One aspect of the present invention provides a novel accounting work support method or system.

[0024] Note that the description of these effects does not preclude the existence of other effects. One embodiment of the present invention does not necessarily have all of these effects. Effects other than these can be extracted from the description in the specification, drawings, and claims.

[0025] FIG. 1 is a diagram illustrating an example of an accounting work support method. FIG. 2 is a diagram illustrating an example of an accounting work support method. FIGS. 3A to 3D are diagrams illustrating an example of an accounting work support method. FIGS. 4A and 4B are diagrams illustrating an example of an accounting work support method. FIGS. 5A to 5C are diagrams illustrating an example of an accounting work support method. FIGS. 6A to 6C are diagrams illustrating an example of an accounting work support method. FIGS. 7A to 7D are diagrams illustrating an example of an accounting work support method. FIGS. 8A and 8B are diagrams illustrating an example of an accounting work support method. FIGS. 9A and 9B are diagrams illustrating an example of an accounting work support method. FIG. 10 is a diagram illustrating an example of an accounting work support method. FIGS. 11A and 11B are diagrams illustrating an example of an accounting work support method. FIGS. 12A to 12C are diagrams illustrating an example of an accounting work support method. FIG. 13 is a diagram illustrating an example of an accounting work support system. FIG. 14 is a diagram illustrating an example of an accounting work support system. FIG. 15 is a block diagram illustrating an example of an accounting work support system. FIG. 16 is a diagram illustrating an example of an accounting work support system. Fig. 17 is a diagram illustrating an example of an accounting work support system. Fig. 18 is a diagram illustrating an example of an accounting work support system. Fig. 19 is a block diagram illustrating an example of an accounting work support system.

[0026] The embodiments will be described in detail with reference to the drawings. However, the present invention is not limited to the following description, and it will be readily understood by those skilled in the art that various changes can be made in form and detail without departing from the spirit and scope of the present invention. Therefore, the present invention should not be interpreted as being limited to the description of the embodiments shown below.

[0027] In the configuration of the invention described below, the same parts or parts having similar functions are denoted by the same reference numerals in different drawings, and repeated explanations thereof will be omitted. Furthermore, when referring to similar functions, the same hatching pattern may be used and no particular reference numeral may be assigned.

[0028] Furthermore, for ease of understanding, the position, size, range, etc. of each component shown in the drawings may not represent the actual position, size, range, etc. Therefore, the disclosed invention is not necessarily limited to the position, size, range, etc. disclosed in the drawings.

[0029] In this specification, the ordinal numbers "first" and "second" are used for convenience and do not limit the number of components or the order of the components (for example, the order of processes or the order of stacking). Furthermore, the ordinal numbers assigned to components in one part of this specification may not match the ordinal numbers assigned to the same components in other parts of this specification or in the claims.

[0030] In this specification, when the same symbol is used for multiple elements, and particularly when it is necessary to distinguish between them, an identification symbol such as “_1”, “[n]”, or “[m, n]” may be added to the symbol.

[0031] In this specification, goods may be either tangible or intangible. For example, intangible goods are sometimes called services. Therefore, the goods described in this specification may sometimes be referred to as services.

[0032] Embodiment In this embodiment, an accounting work support method and an accounting work support system according to one embodiment of the present invention will be described with reference to FIGS.

[0033] Order information can be managed, for example, by registering it in a database. Accounting work is performed based on the order information registered in the database. Because order-related communication with business partners is often conducted via email, it is preferable to register order-related emails in a database. However, it is difficult to properly extract order-related emails. Furthermore, it takes a great deal of effort to select and select information from multiple order-related emails and register it in a database.

[0034] Therefore, an accounting work support method according to one aspect of the present invention includes a first process for extracting emails related to orders and a second process for extracting information from the emails related to orders. By performing the first process, emails related to orders can be extracted with high accuracy. Furthermore, by performing the second process, information to be registered in a database can be extracted from the emails related to orders. Note that the first and second processes are preferably performed using a language model.

[0035] The accounting work support method according to one aspect of the present invention further includes a step of registering the information extracted by the second step in a database. By extracting the information by the second step, the order information can be accurately registered in the database.

[0036] Furthermore, the accounting work support method according to one aspect of the present invention preferably further includes a process of narrowing down the emails using a search expression before the first process, thereby reducing the time required to extract emails related to orders.

[0037] The above-described configuration makes it possible to realize an accounting work support method using a language model. Note that, since the accounting work support method according to one aspect of the present invention can also be applied to an order processing operation, the accounting work support method according to the present invention can also be referred to as an order processing operation support method.

[0038] The accounting support method described above can be implemented using an accounting support system according to one aspect of the present invention. The accounting support system according to one aspect of the present invention includes a processing unit, a reception unit, and a database. The processing unit is configured to execute a first process for extracting emails related to orders and a second process for extracting information from the emails related to orders. The first and second processes are preferably performed using a language model.

[0039] Preferably, the processing unit is configured to execute a process of narrowing down emails using a search expression.

[0040] The receiving unit has a function of receiving order information. The processing unit is configured to execute a process of registering the received order information in a database and a process of registering information extracted by the second process in the database. By extracting information by the second process, the order information can be accurately registered in the database.

[0041] The above-described configuration makes it possible to realize an accounting support system that uses a language model. Note that the accounting support system according to one embodiment of the present invention can also be applied to ordering operations, and therefore the accounting support system according to the present invention can be rephrased as an ordering operation support system.

[0042] <Accounting Work Support Method 1> The accounting work support method 1 of this embodiment will be described with reference to FIGS. 1 to 7D.

[0043] The accounting work support method 1 of this embodiment includes the processes of steps S11 to S20 shown in Fig. 1. Note that the accounting work support method 1 of this embodiment preferably uses an accounting work support system according to one aspect of the present invention. In this case, the accounting work support system can be said to be a system that can execute processing using the accounting work support method 1 of this embodiment.

[0044] The accounting work support system preferably includes an information terminal 20 used by a user, an information processing device 10 having a processing unit and a database, and an information processing device 40 capable of performing processing using a language model. The information terminal 20, the information processing device 10, and the information processing device 40 are connected to each other via a network. Details of the information terminal 20, the information processing device 10, and the information processing device 40 will be described later in <Configuration Example 1 of Accounting Work Support System> (e.g., FIG. 13).

[0045] 1 to 7D are diagrams illustrating accounting work support method 1. FIGS. 3A, 3B, 7B, 7C, and 7D can be considered examples of a graphical user interface (GUI) for the accounting work support system. Forms, icons, tables, and the like in the drawings relating to the GUI exemplified in this embodiment, such as FIG. 3A, are merely examples and are not particularly limited. The GUI can be configured as a web page accessed by a user via a network. Alternatively, the GUI can be configured as a screen of a program application executed on an information terminal 20 used by the user. FIGS. 1 to 7D illustrate an example in which an instruction to extract an email and an instruction to extract information from the email are given to a language model.

[0046] Fig. 2 is a processing flowchart when using the information terminal 20, the information processing device 10, and the information processing device 40. In Fig. 2, the processing performed by the person in charge of ordering and the processing performed using the information terminal 20 are shown on the left side, the processing performed using the information processing device 40 is shown on the right side, and the processing performed using the information processing device 10 is shown between them. Furthermore, of the processing performed using the information processing device 10, the processing related to the processing unit of the information processing device 10 is shown on the left side, and the processing related to the database of the information processing device 10 is shown on the right side.

[0047] As shown in Fig. 2, before the process of step S11, the process of step S41 is performed. In step S41, a person in charge of ordering starts the process of purchasing a product. For example, the person in charge starts the process of purchasing a product by performing at least one of the following: requesting a quote for the product, obtaining a quote for the product, negotiating or consulting about a transaction for the product, issuing a purchase order for the product, and inspecting the product. The person in charge may be one person or multiple people.

[0048] 2, the process of step S41 is followed by the process of step S42. In step S42, the user uses the information terminal 20 to input order information for the product for which the purchase process has begun to the information processing device 10. In other words, in step S42, the information terminal 20 accepts the order information for the product and transmits the accepted order information to the information processing device 10. Hereinafter, the order information for the product for which the purchase process has begun will be simply referred to as order information.

[0049] The order information has at least one data item that corresponds one-to-one to an item in the data table described below. Examples of order information include order number, product name, price, customer, name of sales representative, discount information, quotation expiration date, email address, etc.

[0050] The user may be the person in charge of placing the order. If there are multiple people in charge of placing the order, the user may be one of the people in charge. The user may also be different from the person in charge of placing the order.

[0051] Area 600 shown in FIG. 3A is an area that the user can use to input order information. Area 600 is displayed, for example, on the display unit of information terminal 20. Area 600 has a plurality of pairs of items 511 and areas for inputting data. Examples of items 511 include order number, product name, price, customer, name of sales representative, discount information, quotation expiration date, and email address. In FIG. 3A, four items 511 (items 511[1] to 511[4]) are displayed in area 600. The user inputs data for each item 511.

[0052] 3B shows the area 600 after the user has entered part of the order information. In this example, the data "aa1" is entered in the area corresponding to item 511[1], and the data "bb1" is entered in the area corresponding to item 511[2]. In this case, the entered order information is the data "aa1" and "bb1."

[0053] [Step S11] In step S11, the information processing device 10 accepts order information. Specifically, a reception unit included in the information processing device 10 accepts the order information via the information terminal 20. In the example shown in FIG. 3B , the reception unit accepts data "aa1" and data "bb1" as the order information. Specifically, the reception unit accepts data "aa1" as item 511[1] and data "bb1" as item 511[2].

[0054] [Step S12] In step S12, the information processing device 10 registers the order information in a data table of the database. Specifically, the processing unit of the information processing device 10 transmits the order information received in step S11 to the database (step S12a shown in FIG. 2). The order information received by the database is recorded in the data table (step S12b shown in FIG. 2).

[0055] An example of a data table is shown in FIG. 3C . In the data table 510 shown in FIG. 3C , the first row is an item 511, and each row from the second row onward is a record 513. In this specification, a record is one of the units constituting a data table, and one row's worth of data can be stored per record. Each record 513 registers one piece of order information, and each field in the record 513 registers each piece of data included in the order information. In this specification, a field is the smallest unit constituting a data table, and can store data. For example, in the field in the second row and first column, data corresponding to item 511[1] of the first order information is registered, and in the field in the second row and second column, data corresponding to item 511[2] of the first order information is registered. Furthermore, in the field in the third row and first column, data corresponding to item 511[1] of second order information different from the first order information is registered.

[0056] 3D shows the data table 510 after the processing of step S12 has been performed. As shown in the example of FIG. 3B, when data "aa1" is received as item 511[1] and data "bb1" is received as item 511[2], the data "aa1" is registered in the field in the second row and first column of the data table 510, and the data "bb1" is registered in the field in the second row and second column of the data table 510.

[0057] [Step S13] In step S13, the information processing device 10 searches for emails and acquires the first emails. Specifically, the processing unit of the information processing device 10 searches for emails based on the order information received in step S11, thereby acquiring multiple first emails.

[0058] It is preferable to use a search expression to search for emails. For example, the search expression is preferably created using data included in the order information (specifically, product name, business partner, etc.), information such as the email address of the person in charge or sales representative, etc. Alternatively, it is preferably created using the date and time, such as the date of quotation acquisition or the date of purchase request, etc. Alternatively, it is preferably created using a combination of these. Note that it is possible to use a search expression generated by a language model to search for emails.

[0059] An example of the process of step S13 is shown in Fig. 4A. The multiple emails 210 shown in Fig. 4A are emails to be searched. Examples of the emails 210 include emails managed by the information terminal 20 and emails managed by an IMAP (Internet Message Access Protocol) server.

[0060] 4A are emails that are hit by the search query and may be related to the order information. The emails 220 correspond to the first emails described above.

[0061] In the accounting work support method 1 of this embodiment, n (n is an integer greater than or equal to 2) emails 220 (emails 220[1] to 220[n]) are obtained by searching emails.

[0062] Next, the information processing device 10 preferably extracts first text included in each of the multiple first emails. An example of the first text is shown in FIG. 4B . Text 321[i] (where i is an integer between 1 and n) shown in FIG. 4B is text extracted from email 220[i], and includes at least one of the email address described in the sender (From) field, the email address described in the destination (To) field, the subject (Subject), the text body (Text), and the like.

[0063] The process of step S13 can be thought of as a process of narrowing down the e-mails using a search expression. By performing this process, the number of e-mails that are judged to be related to the order information can be reduced. Therefore, the time required to extract e-mails related to the order information can be shortened.

[0064] The plurality of emails 220 are also used when creating the first prompt in the subsequent processing of step S14. By acquiring the plurality of emails 220, it is possible to narrow down the emails used when creating the first prompt. This prevents the amount of text in the first prompt from increasing, making it easier to obtain a response sentence from a language model. It is also possible to shorten the time it takes to generate a response sentence using a language model.

[0065] A prompt corresponds to an input sentence that causes a language model to perform a desired action. When a prompt is given to a language model, the language model generates a response sentence based on the prompt. The prompt includes an instruction sentence. An instruction sentence can be rephrased as a question sentence, an imperative sentence, or the like. It is preferable that the instruction sentence be set in advance by a developer or a service provider. Setting the instruction sentence included in the prompt in advance makes it easier to obtain a desired answer from the language model.

[0066] [Step S14] In step S14, the information processing device 10 creates a first prompt. Specifically, the processing unit of the information processing device 10 creates a first prompt for extracting emails related to order information. In other words, the processing unit of the information processing device 10 creates a first prompt for classifying a plurality of first emails.

[0067] An example of the first prompt is shown in Figure 5A. Prompt 410 shown in Figure 5A includes instruction statement 411 and list 412.

[0068] The instruction 411 includes an instruction for determining whether the email 220 is related to the order information. The instruction 411 also includes an instruction for determining whether the text 321 is related to the order information. For example, an instruction such as "Please answer whether each email included in the list is related to the order information under the constraints" may be prepared. Note that an example of the constraints may be "Exclude emails containing private information." The instruction 411 also preferably includes at least a portion of the order information. By including at least a portion of the order information in the instruction 411, it is possible to accurately determine whether the email 220 is related to the order information.

[0069] The list 412 preferably contains text. The list 412 is preferably created using a plurality of emails 220. For example, the list 412 is preferably created to include n pieces of text 321 (text 321[1] to text 321[n]). In Fig. 5A, the serial number "001" is assigned to the text 321[1], and the serial number "002" is assigned to the text 321[2].

[0070] As described above, for each of the plurality of first e-mails, it is possible to determine whether the first e-mail is related to the order information. The first prompt is a prompt for determining whether the first e-mail is related to the order information, and can be called a determination prompt.

[0071] The output format of the determination result may be specified by an instruction statement 411. For example, a statement such as "Indicate related emails with T (True) and unrelated emails with F (False)" may be added to the instruction statement 411. In this way, by outputting the determination result using the initial letters of words, abbreviations such as acronyms, or symbols, the number of characters in the first response sentence can be reduced. Therefore, the language model is less subject to limitations on the number of characters it can output, and the probability of obtaining the first response sentence can be increased.

[0072] [Step S15] In step S15, the information processing device 10 acquires a first response sentence by transmitting a first prompt to the language model via the network. Because the first prompt is a determination prompt, the first response sentence includes the result of the determination as to whether the first email is related to order information. In other words, the information processing device 10 acquires a first response sentence including the result of the determination.

[0073] For example, information processing device 10 transmits a first prompt to information processing device 40 via a network (step S15a shown in FIG. 2). Upon receiving the first prompt, information processing device 40 performs processing using a language model to generate a first response sentence corresponding to the first prompt and transmits the first response sentence to information processing device 10 via the network (step S15b shown in FIG. 2). Information processing device 10 then acquires the first response sentence (step S15c shown in FIG. 2).

[0074] An example of a first response sentence is shown in FIG. 5B . Response sentence 420 shown in FIG. 5B includes text 421. Text 421 includes the result of determining whether the first email is related to order information. FIG. 5B shows an example in which T (True) is output for a first email determined to be related to order information, and F (False) is output for a first email determined not to be related to order information. For example, because F (False) is output for text 321[1] with the serial number "001," email 220[1] is determined not to be related to order information. Furthermore, because T (True) is output for text 321[2] with the serial number "002," email 220[2] is determined to be related to order information.

[0075] In accounting work support method 1 of the present embodiment, list 412 includes n pieces of text 321, and therefore text 421 includes the results of determination for each of n pieces of email 220. In other words, a single communication between information processing device 10 and information processing device 40 can obtain the results of determination for each of n pieces of email 220. Therefore, the frequency of communication between information processing device 10 and information processing device 40 can be minimized.

[0076] [Step S16] In step S16, the information processing device 10 extracts at least one second email from the plurality of first emails based on the first response sentence. The second email is an email determined to be related to the order information.

[0077] An example of the processing of step S16 is shown in FIG. 5C. The email 230 shown in FIG. 5C is an email determined to be related to the order information and corresponds to the second email. In the accounting work support method 1 of this embodiment, at least one email 230 is extracted. Note that FIG. 5C shows multiple emails 230. For example, in the example shown in FIG. 5B, email 220[2] is determined to be related to the order information, and is therefore one of the multiple emails 230. In this case, the multiple emails 230 include email 220[2].

[0078] On the other hand, if the email 230 is not extracted, the subsequent processing may not be performed, or the process may return to step S13, where the search formula may be changed to search for emails.

[0079] [Step S17] In step S17, the information processing device 10 adds at least one second email to the data table. For example, in step S17, the information processing device 10 adds at least one second email to a record in which order information is registered. Here, the record in which order information is registered refers to a record in which the order information included in the instruction text of the first prompt is registered. In addition, in this specification, adding an email to a data table refers to adding an email to a record in which order information is registered.

[0080] For example, the information processing device 10 sends at least one second email to the database (step S17a shown in FIG. 2). The at least one second email received by the database is recorded in a record in which the order information is registered (step S17b shown in FIG. 2).

[0081] Next, the information processing device 10 preferably extracts second text contained in at least one second email from each of the second emails. Note that the second emails are emails extracted from multiple first emails. Therefore, the structure of the second text is the same as the structure of the first text. Therefore, the structure of the second text can refer to the description of the first text described above.

[0082] [Step S18] In step S18, the information processing device 10 creates a second prompt. Specifically, the processing unit of the information processing device 10 creates the second prompt for extracting information from at least one second email. The information is data corresponding to each item in the data table.

[0083] An example of the second prompt is shown in Figure 6A. Prompt 430 shown in Figure 6A has instruction sentence 431 and object sentence 432.

[0084] The instruction 431 includes an instruction to extract information corresponding to each item in the data table from the second text. For example, an instruction such as "Refer to the second text and the items in the data table and extract information corresponding to each item in the data table" may be prepared.

[0085] The target sentence 432 includes text 331. Preferably, the text 331 includes second text included in the second email. Note that, if multiple second emails are extracted, the text 331 may include the second text of each of the multiple second emails.

[0086] Furthermore, it is preferable that target sentence 432 includes text 433 in addition to text 331. For example, text 433 preferably includes items in a data table and order information registered in a record. In this case, it is preferable to prepare an instruction such as, "Refer to the second text and the items in the data table to supplement the order information not recorded in the record, and extract the order information corresponding to each item in the data table." This configuration makes it possible to efficiently extract order information not registered in a record.

[0087] The second prompt is a prompt for extracting data corresponding to each item in the data table, and the record is later updated using the extracted data. Therefore, it is preferable that a response statement including the extracted data be output in SQL (Structured Query Language). Therefore, it is preferable that instruction statement 431 includes a command for creating SQL, and text 433 includes information about the data table. For example, an instruction such as "Refer to the second text and the items in the data table to supplement the order information not recorded in the record, extract the order information corresponding to each item in the data table, and create SQL to register it in the database" may be prepared. This configuration allows for efficient extraction of order information not registered in the record, and also allows for the acquisition of a response statement including a description corresponding to the SQL.

[0088] As a result, information corresponding to each item in the data table can be extracted from at least one second email. The second prompt is a prompt for extracting information from the second email, and can be called an information extraction prompt.

[0089] [Step S19] In step S19, the information processing device 10 acquires a second response sentence by transmitting a second prompt to the language model via the network. Because the second prompt is an information extraction prompt, the second response sentence includes information extracted from the second email. In other words, the information processing device 10 acquires a second response sentence including the extracted information.

[0090] For example, the information processing device 10 transmits a second prompt to the information processing device 40 via a network (step S19a shown in FIG. 2). The information processing device 40, which has received the second prompt, generates a second response sentence corresponding to the second prompt by performing processing using a language model, and transmits the second response sentence to the information processing device 10 via the network (step S19b shown in FIG. 2). The information processing device 10 then acquires the second response sentence (step S19c shown in FIG. 2).

[0091] An example of the second response sentence is shown in Figure 6B. Response sentence 440 shown in Figure 6B has text 441. Text 441 includes information extracted from the second email. In Figure 6B, "aa1" is output for item "511[1]", "bb1" is output for item "511[2]", "cc1" is output for item "511[3]", and "dd1" is output for item "511[4]". In this case, it can be said that data "cc1" and data "dd1" have been extracted as order information that is not registered in the record.

[0092] Another example of the second response statement is shown in Fig. 6C. The response statement 440 shown in Fig. 6C has text 441. If the directive statement 431 includes an instruction to create SQL, the text 441 can include a description corresponding to the SQL, as shown in Fig. 6C.

[0093] In the accounting work support method 1 of this embodiment, a series of processes are performed using the received order information, so it is preferable that the language model used in the process of step S15 is the same as the language model used in the process of step S19. In other words, it is preferable that the second prompt be sent to the language model that receives the first prompt. With this configuration, the accuracy of the first response sentence and the accuracy of the second response sentence can be maintained constant.

[0094] The language model used in the processing of step S15 may be different from the language model used in the processing of step S19. That is, a first prompt may be sent to a first language model, and a second prompt may be sent to a second language model different from the first language model. With this configuration, the computational load in accounting work support method 1 of the present embodiment is distributed between a first information processing device capable of processing using the first language model and a second information processing device capable of processing using the second language model, thereby reducing the computational load on each of the first information processing device and the second information processing device.

[0095] [Step S20] In step S20, the information processing device 10 updates the data table using the extracted information based on the second response sentence. Furthermore, if the extracted information includes information for an item for which data is not registered, the information processing device 10 registers the information in the corresponding field. For example, in step S20, the information processing device 10 updates the record in which order information is registered using the extracted information based on the second response sentence. In this specification and the like, updating the data table refers to registering information in a field in which information is not registered in a record in which order information is registered, replacing information in a field in which information is registered, or the like.

[0096] For example, the information processing device 10 transmits the extracted information to a database (step S20a shown in FIG. 2). The information received by the database is recorded in a record in which order information is registered (step S20b shown in FIG. 2). By performing the processing of step S20, the order information registered in the record can be updated.

[0097] 7A shows the data table 510 after the processing of step S20 has been performed. When the data "cc1" and the data "dd1" are accepted as items 511[3] and 511[4], respectively, the data "cc1" is registered in the field in the second row and third column of the data table 510, and the data "dd1" is registered in the field in the second row and fourth column of the data table 510, as shown in FIG.

[0098] As described above, it is possible to register order information in a database using a language model, and to manage product purchasing operations using a language model.

[0099] In particular, by performing the processes in steps S18 to S20, it is possible to select information from emails related to orders and register that information in the database, thereby reducing the time, effort, and input errors required for registering information in the database.

[0100] Furthermore, by using a common instruction sentence included in the second prompt, the information extracted from the order-related email is less dependent on the user. Furthermore, information can be extracted to supplement order information not registered in the records. Therefore, order information can be accurately registered in the database.

[0101] The updated order information may be displayed on the display unit of the information terminal 20. Figure 7B shows an area 600 after the processing of step S20 has been performed. Figure 7B shows an example in which data "cc1" has been entered in the area corresponding to item 511[3], and data "dd1" has been entered in the area corresponding to item 511[4].

[0102] 2, after the process of step S20, the process of step S43 is performed. In step S43, the user checks the registered order information. For example, the user can check whether the order information displayed on the display unit of the information terminal 20 is correct.

[0103] It should be noted that the order information registered by the user in step S42 may differ from the information extracted in step S19. In other words, the order information registered in the record (the order information before being updated) may not match the information extracted in step S19. This is because, for example, an input error in step S42, an input error or mistranslation in the email, etc. may occur.

[0104] If the order information registered in the record (e.g., the order information registered in step S42) does not match the information extracted in step S19, it is preferable that the data of the mismatched items be highlighted. For example, if the order information registered in the record does not match the information extracted in step S19, it is preferable that the data of the mismatched items in the order information be highlighted on the display unit of the information terminal 20. For example, in Figure 7C, if the order information registered in step S42 as item 511[2] (the order information registered in the field of the data table as item 511[2]) does not match the information extracted in step S19, the mismatch between the order information and the information is indicated by hatching the area corresponding to item 511[2].

[0105] As described above, by highlighting the order information registered in the records of the data table that does not match the information extracted in step S19, the user can easily visually identify the order information that may be incorrect. Note that the highlighting method is not limited to hatching, and examples include changing the background color, changing the color of the characters, increasing the character size, changing the font, making the characters bold, and underlining the character string that is determined to need correction.

[0106] 7D, a speech bubble may be displayed near item 511[2], and the information extracted in step S19 may be output inside the speech bubble as a candidate for item 511[2]. This configuration allows the user to select the correct information from multiple candidates.

[0107] This concludes the explanation of accounting work support method 1.

[0108] <Modification 1> A modification of the accounting work support method 1 described above will be described with reference to FIGS. 8A, 9A, and 9B.

[0109] Variation 1 of this embodiment includes the processes of steps S11 to S20, step S31, and step S32. Variation 1 of this embodiment differs from the accounting work support method 1 described above mainly in that it includes the processes of steps S31 and S32.

[0110] The first modification of this embodiment is effective when a first prompt is created for each first email, or in other words, when a plurality of first prompts are created.

[0111] Steps S11 to S13 shown in FIG. 8A are the same as steps S11 to S13 shown in FIG. 2, respectively, and therefore the explanation of the accounting work support method 1 mentioned above can be referred to.

[0112] In step S14, the information processing device 10 creates a first prompt. Step S14 in the first modification of the present embodiment is similar to step S14 in the first accounting work support method described above in that a prompt for determination is created.

[0113] In the accounting work support method 1 described above, one first prompt is created using all emails 220. On the other hand, in the first modification of this embodiment, a first prompt is created for each email 220. In other words, the same number of first prompts as the number of emails 220 are created. For example, if n emails 220 are obtained, n first prompts are created.

[0114] An example of the first prompt is shown in Fig. 9A. Prompt 410[i] shown in Fig. 9A includes instruction statement 411 and text 321[i]. For instruction statement 411 and text 321[i], the explanation of accounting work support method 1 can be referred to.

[0115] In step S15, the information processing device 10 acquires a first response sentence by transmitting a first prompt to the language model via the network. Since n first prompts are created, n communications are performed between the information processing device 10 and the information processing device 40.

[0116] An example of the first response sentence is shown in Figure 9B. Response sentence 420[i] shown in Figure 9B has text 421[i]. Text 421[i] includes the result of determining whether email 220[i] is related to order information. In Figure 9B, T (True) is output for email 220[i]. Therefore, email 220[i] is determined to be related to order information.

[0117] The number of characters in text 321[i] is fewer than the number of characters in list 412. In other words, the number of characters in prompt 410[i] is fewer than the number of characters in prompt 410 described in accounting work support method 1. This reduces the limitations on the number of characters that can be input to the language model, and increases the probability of obtaining the first response sentence.

[0118] In step S31, the information processing device 10 determines whether all of the first e-mails have been determined to be related to order information. If it is determined that all of the first e-mails have been determined to be related to order information, the process proceeds to step S16. On the other hand, if it is determined that there is at least one first e-mail for which determination has not been performed, the process returns to step S14.

[0119] 9A illustrates an example in which the same number of first prompts as the number of emails 220 are created, but the present invention is not limited to this. For example, multiple emails 220 may be divided, and a first prompt may be created for each of the divided emails 220.

[0120] For example, text 321[i] may be created by combining first texts within a range in which the total number of characters in the text does not exceed a predetermined number of characters. This configuration allows the number of first prompts to be less than the number of first emails. Therefore, the frequency of communication between information processing device 10 and information processing device 40 can be reduced while reducing the impact of limitations on the number of characters that can be input to the language model.

[0121] As described above, in step S16, the information processing device 10 extracts at least one second email from the plurality of first emails based on the first response sentence. Therefore, it is preferable that the information processing device 10 has a function of storing all the generated first response sentences until the processing of step S16 is started.

[0122] Steps S16 and S17 shown in FIG. 8A are the same as steps S16 and S17 shown in FIG. 2, respectively, and therefore the explanation of the accounting work support method 1 above can be referred to.

[0123] In step S18, the information processing device 10 creates a second prompt. Step S18 in the first modification of the present embodiment is similar to step S18 in the first accounting work support method described above in that a prompt for information extraction is created.

[0124] In the accounting work support method 1 described above, the second prompt is created using all emails 220. On the other hand, in the first modification of this embodiment, the second prompt is created for each email 220. In other words, the same number of second prompts as the number of second emails are created. The second prompt can refer to the content described using FIG. 6A.

[0125] In step S19, the information processing device 10 transmits a second prompt to the language model via the network to obtain a second response sentence. The second response sentence can refer to the content described with reference to FIG. 6B.

[0126] The number of characters in the second prompt in the first modification of this embodiment is smaller than the number of characters in the second prompt described above in the accounting work support method 1. This reduces the limitations on the number of characters that can be input to the language model, and increases the probability of obtaining a second response sentence.

[0127] In step S32, the information processing device 10 determines whether information extraction has been performed for all of the second e-mails. If it is determined that information extraction has been performed for all of the second e-mails, the process proceeds to step S20. On the other hand, if it is determined that there is at least one second e-mail for which information extraction has not been performed, the process returns to step S18.

[0128] Although the present invention is exemplified by a configuration in which the same number of second prompts as the number of second emails are created, the present invention is not limited to this. For example, multiple second emails may be divided and a second prompt may be created for each divided second email.

[0129] For example, the text 331 may be created by combining second texts as long as the total number of characters in the text does not exceed a predetermined number of characters. This configuration allows the number of second prompts to be less than the number of second emails. Therefore, the frequency of communication between the information processing device 10 and the information processing device 40 can be reduced while reducing the impact of limitations on the number of characters that can be input to the language model.

[0130] As described above, in step S20, the information processing device 10 updates the record in which the order information is registered using the extracted information based on the second response sentence. Therefore, it is preferable that the information processing device 10 has a function of storing the extracted information until the processing of step S20 is started.

[0131] Step S20 shown in FIG. 8A is the same as step S20 shown in FIG. 2, and therefore the explanation of the accounting work support method 1 described above can be referred to.

[0132] The above is the description of the first modification.

[0133] <Modification 2> Another modification of the accounting work support method 1 described above will be described with reference to FIG. 8B.

[0134] Variation 2 of this embodiment includes the processes of steps S11 to S20, step S33, and step S34. Variation 2 of this embodiment differs from the above-described accounting work support method 1 mainly in that it includes the processes of steps S33 and S34. Variation 2 of this embodiment also differs from the above-described variation 1 mainly in that it does not include the processes of steps S31 and S32, but includes the processes of steps S33 and S34.

[0135] The second modification of this embodiment is effective when a first prompt is created for each first email, or in other words, when a plurality of first prompts are created.

[0136] Steps S11 to S13 shown in FIG. 8B are the same as steps S11 to S13 shown in FIG. 2, respectively, and therefore the explanation of the accounting work support method 1 mentioned above can be referred to.

[0137] Steps S14 and S15 shown in FIG. 8B are the same as steps S14 and S15 shown in FIG. 8A, respectively, and therefore the description of the first modified example above can be referred to.

[0138] In step S33, the information processing device 10 determines whether the first email is related to order information. If it is determined that the first email is related to order information, the process proceeds to step S16. On the other hand, if it is determined that the first email is not related to order information, the process returns to step S14.

[0139] Steps S16 to S19 shown in FIG. 8B are the same as steps S16 to S19 shown in FIG. 8A, respectively, and therefore the description of the first modified example above can be referred to.

[0140] In step S34, the information processing device 10 determines whether the series of processes (specifically, steps S14 to S19) have been performed on all primary emails. If it is determined that the series of processes have been performed on all primary emails, the process proceeds to step S20. On the other hand, if it is determined that there is at least one primary email for which the series of processes has not been performed, the process returns to step S14.

[0141] Step S20 shown in FIG. 8B is the same as step S20 shown in FIG. 2, and therefore the explanation of the accounting work support method 1 described above can be referred to.

[0142] As with the above-described variant example 1, the configuration of variant example 2 reduces the limitations on the number of characters that can be input to the language model, and increases the probability of obtaining the first response sentence and the second response sentence.

[0143] The above is the explanation of the second modification.

[0144] <Accounting Work Support Method 2> Accounting work support method 2 of this embodiment will be described with reference to FIGS. 10 to 12C.

[0145] 10 to 12C are diagrams illustrating accounting work support method 2. Fig. 10 to 12C show an example of a case where an instruction to extract an email and an instruction to extract information from the email are given to a language model.

[0146] The accounting work support method 2 of this embodiment includes the processing of steps S11 to S21 shown in Fig. 10. The accounting work support method 2 of this embodiment differs from the above-described accounting work support method 1 mainly in that it includes the processing of step S21.

[0147] An e-mail related to an order may have attached files such as a purchase order, a quotation, etc. The accounting work support method 2 of this embodiment is effective when an e-mail related to an order has attached files.

[0148] Steps S11 to S13 shown in FIG. 10 are the same as steps S11 to S13 shown in FIG. 1, respectively, and therefore the explanation of the accounting work support method 1 mentioned above can be referred to.

[0149] If a file is attached to at least one of the plurality of first e-mails, the process proceeds to step S21. If a file is not attached to any of the plurality of first e-mails, the process proceeds to step S14. In other words, if a file is not attached to any of the plurality of first e-mails, the above-described accounting work support method 1 may be used.

[0150] [Step S21] In step S21, the information processing device 10 extracts third text from files attached to the first emails. Note that the process of step S21 is preferably performed for all first emails to which files are attached.

[0151] An example of the first email is shown in Fig. 11A. Email 220[i] shown in Fig. 11A includes text 321[i] and file 221[i] attached to email 220[i]. For text 321[i], the description of accounting work support method 1 can be referenced.

[0152] 11B, by performing the processing of step S31, text 322[i] is extracted from file 221[i]. Text 322[i] corresponds to the third text described above. Note that the extracted text 322[i] may be all of the text contained in file 221[i], or may be only a portion of the text contained in file 221[i].

[0153] Furthermore, if the file 221[i] contains data other than text data, the information processing device 10 preferably converts the data into text data. When converting various data into text data, it is preferable to use AI (Artificial Intelligence) technology. For example, if the data contained in the file 221[i] is image data, the information processing device 10 can generate text data from the image data by performing processing using at least one AI technology, such as image recognition, handwritten character recognition, or caption generation. Furthermore, the information processing device 10 may have an optical character recognition (OCR) function.

[0154] For a first email that does not have a file attached, the third text may be left blank.

[0155] In step S14, the information processing device 10 creates a first prompt. Step S14 of the accounting work support method 2 of this embodiment is similar to step S14 of the above-mentioned accounting work support method 1 in that a prompt for determination is created.

[0156] An example of the first prompt is shown in Fig. 12A. The prompt 410 shown in Fig. 12A has an instruction statement 411 and a list 412. For the instruction statement 411, the explanation of the accounting work support method 1 described above can be referred to.

[0157] The list 412 is preferably created using a plurality of emails 220. For example, the list 412 preferably includes a first text and a third text. For example, the list 412 is created to include n texts 321 (text 321[1] to text 321[n]) and n texts 322 (text 322[1] to text 322[n]).

[0158] By including text 322 in the first prompt, the accuracy of the determination made in step S15 can be improved.

[0159] Steps S15 to S17 shown in FIG. 10 are the same as steps S15 to S17 shown in FIG. 1, respectively, and therefore the explanation of the accounting work support method 1 mentioned above can be referred to.

[0160] As with the accounting work support method 1 described above, by configuring the accounting work support method 2 of this embodiment, it is possible to obtain the result of judgment for each of the n emails 220 in one communication between the information processing device 10 and the information processing device 40. Therefore, the frequency of communication between the information processing device 10 and the information processing device 40 can be minimized.

[0161] In step S18, the information processing device 10 creates a second prompt. Step S18 of the accounting work support method 2 of this embodiment is similar to step S18 of the above-mentioned accounting work support method 1 in that a prompt for information extraction is created.

[0162] An example of the second prompt is shown in Fig. 12B. The prompt 430 shown in Fig. 12B has an instruction statement 431 and a target statement 432. For the instruction statement 431, the explanation of the accounting work support method 1 described above can be referred to.

[0163] Target sentence 432 includes text 331 and text 332. Text 331 preferably includes second text included in at least one second email. If multiple second emails are extracted, text 331 preferably includes second text for each of the multiple second emails. Furthermore, text 332 preferably includes third text extracted from a file attached to the second email. If multiple second emails with attached files are obtained, text 332 preferably includes third text extracted from each of the multiple files.

[0164] Furthermore, it is preferable that the target sentence 432 includes text 433 in addition to text 331 and text 332. For the text 433, the explanation of the accounting work support method 1 described above can be referred to.

[0165] Steps S19 and S20 shown in FIG. 10 are the same as steps S19 and S20 shown in FIG. 1, respectively, and therefore the explanation of the accounting work support method 1 above can be referred to.

[0166] Although accounting work support method 2 of this embodiment illustrates a case in which the first prompt is created using a plurality of emails 220, the present invention is not limited to this. As explained in the previous modified example 1 or modified example 2, the first prompt may be created using each of the plurality of emails 220. In other words, the same number of first prompts as the number of emails 220 may be created.

[0167] An example of the first prompt is shown in Figure 12C. Prompt 410[i] shown in Figure 12C includes instruction 411[i], text 321[i], and text 322[i]. Instruction 411[i] and text 321[i] can refer to the explanation of accounting work support method 1 described above. Text 322[i] corresponds to the third text described above.

[0168] Alternatively, multiple emails 220 may be divided, and a first prompt may be created for each divided email 220. In other words, a smaller number of first prompts may be created than the number of emails 220. With this configuration, it is possible to reduce the frequency of communication between the information processing device 10 and the information processing device 40 while being less subject to limitations on the number of characters that can be input to the language model.

[0169] This concludes the explanation of accounting work support method 2.

[0170] In this embodiment, a configuration is exemplified in which a language model is used to determine whether the first email is related to order information, but the present invention is not limited to this. For example, the determination may be made using a model of a different type from the language model. Here, the model used for the determination is referred to as a determination model. For example, the determination model may be a model optimized by machine learning or a rule-based model. By using an optimized model or a rule-based model, high accuracy can be achieved with less computational effort (fewer parameters) than a language model. Furthermore, the determination model may be a model that combines machine learning and rule-based models. Note that the determination model may be a configuration that combines a language model and a prompt.

[0171] In this embodiment, a configuration in which information is extracted from the second email using a language model is exemplified, but the present invention is not limited to this. For example, information may be extracted using a model of a different type from a language model. Here, a model used to extract information is called an extraction model. For example, the extraction model may be a model optimized by machine learning or a rule-based model. By using an optimized model or a rule-based model, high accuracy can be achieved with less computational effort (fewer parameters) than a language model. Furthermore, the extraction model may be a model that combines machine learning and rule-based models. Note that the extraction model may be a configuration in which a language model and a prompt are combined.

[0172] In this embodiment, an example is given of a configuration in which the first prompt created in step S14 is sent to an information processing device 40 that can perform processing using a language model without any processing, but the present invention is not limited to this.

[0173] For example, when implementing the accounting work support method of this embodiment using an external service that provides a language model, there is a concern that personal information or confidential information, etc., included in the first prompt may be leaked. Therefore, it is preferable to determine whether the first prompt contains a word containing personal information or confidential information. Then, based on the result of the determination, it is preferable to delete the word included in the first prompt. Alternatively, based on the result of the determination, it is preferable to replace the word included in the first prompt with another word belonging to the same field as the word. This prevents words such as personal information or confidential information from being input into the language model. Therefore, it is possible to prevent the leakage of personal information, confidential information, etc., and realize an accounting work support method that can be used safely by users. Furthermore, it is also possible for service providers using the accounting work support method to prevent unintended information from being disclosed to users, thereby realizing a highly secure accounting work support method.

[0174] The above determination and word deletion or replacement may be performed by the information processing device 10. The above determination and word deletion or replacement may be performed between step S14 and step S15. The above determination and word deletion or replacement may be performed for the second prompt.

[0175] <Configuration Example 1 of Accounting Work Support System> Fig. 13 shows an example of an accounting work support system of this embodiment. The accounting work support system of this embodiment has an information processing device 10, an information terminal 20, and an information processing device 40. In the accounting work support system shown in Fig. 13, two selected from the information processing device 10, the information terminal 20, and the information processing device 40 are connected via a network 30. In Fig. 13, information terminal 20a, information terminal 20b, information terminal 20c, and information terminal 20d are shown as examples of the information terminal 20.

[0176] A user of the accounting work support system can access the information processing device 10 from the information terminals 20a to 20d, etc. The user can then receive services using the accounting work support method according to one aspect of the present invention.

[0177] A user can use the information processing device 10, for example, by using dedicated software or an application installed on the information terminal 20. Alternatively, a user can use the information processing device 10, for example, from a web browser on the information terminal 20.

[0178] An organization such as a company that provides a service using an accounting work support method according to one aspect of the present invention (hereinafter also referred to as a service provider) can provide the service using the information processing device 10 .

[0179] The information processing device 10 is a device that can execute processing using an accounting work support method according to one aspect of the present invention. In Fig. 13, a server computer is illustrated as an example.

[0180] The information processing device 10 can perform information processing such as calculations using data input from the information terminal 20a via the network 30. The information processing device 10 can transmit the results of the information processing to the information terminal 20a via the network 30. This can reduce the calculation load on the information terminal 20a. Here, the information terminal 20a has been described as an example, but the same can be said for the information terminals 20b to 20d.

[0181] The information processing device 10 can input data to the information processing device 40 via the network 30. The information processing device 40 can perform information processing such as calculations using the input data. The information processing device 40 can transmit the results of the information processing to the information processing device 10 via the network 30. This can reduce the calculation load on the information processing device 10. The information processing device 10 can then transmit the data received from the information processing device 40 or data created based on the data to the information terminal 20a via the network 30.

[0182] The information terminals 20a to 20d are each an information terminal such as a computer used by a user, and may also be referred to as client computers. FIG. 13 illustrates, as an example, the information terminal 20a, which is a desktop computer; the information terminal 20b, which is a notebook computer; the information terminal 20c, which is a smartphone; and the information terminal 20d, which is a tablet computer. The information terminal 20d can also be used as a notebook computer by connecting it to a housing 21 having a keyboard. The number of information terminals connected to the information processing device 10 is not particularly limited. While FIG. 13 illustrates four information terminals, the number of information terminals may be one, two, three, five, or more. Examples of such information terminals include desktop information terminals, notebook information terminals, tablet information terminals, and mobile information terminals such as smartphones.

[0183] The information processing device 10 is a large computer such as a server computer or a supercomputer. The information processing device 10 is a computer with higher processing power than the information terminal 20. The information processing device 10 can function as a parallel computer. By using the information processing device 10 as a parallel computer, it is possible to perform large-scale calculations required for learning and inference of artificial intelligence (AI), for example. The information processing device 10 may also be capable of performing processing using a natural language processing model using AI.

[0184] The information processing device 40 is a large computer such as a server computer or a supercomputer. The information processing device 40 is a computer with higher processing power than the information terminal 20. The information processing device 40 is also a computer with higher processing power than the information processing device 10. It is preferable that the information processing device 40 has a function as a parallel computer. By using the information processing device 40 as a parallel computer, it is possible to perform large-scale calculations required for AI learning and inference, for example. Note that when both the information processing device 10 and the information processing device 40 have a function as a parallel computer, the information processing device 40 has a higher processing power and can perform large-scale calculations than the information processing device 10.

[0185] The information processing device 40 can perform processing using a natural language processing model that uses AI. Examples of the natural language processing model that uses AI include BERT (Bidirectional Encoder Representations from Transformers) and T5 (Text-to-Text Transfer Transformer). The information processing device 40 can also perform processing using a model that utilizes a large-scale language model. Examples of large-scale language models include GPT-3, GPT-3.5, GPT-4 (registered trademark), LaMDA (Language Model for Dialogue Applications), PaLM (Pathways Language Model), and PaLM2, and it is preferable to use GPT-4 (registered trademark).

[0186] The information processing device 40 can execute processing using a general-purpose AI model that can perform various tasks. When the information processing device 10 and the information processing device 40 can each execute processing using a model that uses a large-scale language model, the information processing device 40 can execute processing using a larger-scale AI model than the information processing device 10.

[0187] The network 30 is typically a local network. It is preferable to use an intranet or an extranet as the network 30. In this embodiment, the case where the network 30 is an intranet will be mainly described as an example.

[0188] Other computer networks that can be used as the network 30 include a PAN (Personal Area Network), a LAN (Local Area Network), a CAN (Campus Area Network), a MAN (Metropolitan Area Network), a WAN (Wide Area Network), and a GAN (Global Area Network).

[0189] Furthermore, when wireless communication is performed, communication standards such as the fourth generation mobile communication system (4G), fifth generation mobile communication system (5G), and sixth generation mobile communication system (6G), or specifications standardized by the IEEE such as Wi-Fi (registered trademark) and Bluetooth (registered trademark), can be used as communication protocols or communication technologies.

[0190] The accounting work support system shown in FIG. 13 is suitable for a case where the user who uses the information terminal 20 and the service provider belong to the same organization such as a company.

[0191] For example, it is preferable that data be transmitted and received between the information terminal 20 and the information processing device 10, and between the information processing device 10 and the information processing device 40, using a network 30 established within an organization such as a company. This allows data to be transmitted and received more securely than when the information terminal 20 or the information processing device 10 is connected to the information processing device 40 via a global network such as the Internet. Furthermore, it is possible to prevent confidential information within the organization from leaking to the outside.

[0192] The flow of data in the accounting work support system shown in FIG. 13 will be described with reference to FIG.

[0193] First, the user inputs data DA1 to the information processing device 10 using the information terminal 20. The data DA1 is transmitted from the information terminal 20 to the information processing device 10 via the network 30.

[0194] The data DA1 is, for example, order information for a product for which a purchase operation has begun. The format of the data DA1 is not particularly important, but it is preferably text data. For details of the data DA1, please refer to the explanation in the above-mentioned <Accounting Work Support Method 1>.

[0195] Next, the information processing device 10 creates data DA2 based on data DA1. Data DA2 is a prompt that the user wants to input to the language model, and corresponds to the first prompt and the second prompt described in the above <Accounting Work Support Method 1>. For details of data DA2, please refer to the description in the above <Accounting Work Support Method 1>.

[0196] The created data DA2 is transmitted from information processing device 10 to information processing device 40 via network 30.

[0197] Next, the information processing device 40 generates data DB1 based on the data DA2. Data DB1 is a response sentence corresponding to the data DA2 input to the information processing device 40, and corresponds to the first response sentence and the second response sentence described in the above <Accounting Work Support Method 1>. For details of data DB1, please refer to the description in the above <Accounting Work Support Method 1>, etc.

[0198] Next, the data DB1 is transmitted from the information processing device 40 to the information processing device 10 via the network 30. That is, the information processing device 10 acquires the data DB1 from the information processing device 40 via the network 30.

[0199] Next, the information processing device 10 creates a data DB2 based on the data DB1. The data DB2 is the data DA1 updated based on the data DB1, and corresponds to the order information to which the information extracted in step S19 has been added. For details of the data DB2, please refer to the explanation in the above-mentioned <Accounting Work Support Method 1>.

[0200] Next, the information processing device 10 outputs the data DB2 to the information terminal 20 via the network 30. As a result, the user can obtain the data DB2 in which the data DA1 has been updated.

[0201] Fig. 15 shows a block diagram of the accounting work support system shown in Fig. 13. In Fig. 15, an example of the configuration of the information processing device 10 will be mainly described.

[0202] The information processing device 10 includes a receiving unit 110, a storage unit 120, a processing unit 130, an output unit 140, and a transmission path 150. The information processing device 10 includes a database 121.

[0203] In the drawings accompanying this specification, the components are classified by function and shown as independent blocks in the block diagrams. However, in reality, it is difficult to completely separate the components by function, and one component may be involved in multiple functions. For example, part of the processing unit 130 may function as the reception unit 110. Also, one function may be involved in multiple components. For example, the processing performed by the processing unit 130 may be executed by different servers depending on the processing.

[0204] [Reception Unit 110] The reception unit 110 can receive data from outside the information processing device 10. Furthermore, the data received by the reception unit 110 is preferably text data.

[0205] For example, the reception unit 110 receives data from the information terminal 20 via the network 30. The data received from the information terminal 20 corresponds to the data DA1 described above. Specifically, the reception unit 110 has a function of receiving the order information described above in <Accounting Work Support Method 1>.

[0206] Furthermore, for example, the receiving unit 110 receives data from the information processing device 40 via the network 30. The data received from the information processing device 40 corresponds to the above-mentioned data DB1. Specifically, the receiving unit 110 has a function of receiving the first response sentence and the second response sentence described above in <Accounting Work Support Method 1>.

[0207] The data supplied to the reception unit 110 is supplied to one or both of the storage unit 120 and the processing unit 130 via the transmission path 150 .

[0208] [Storage Unit 120] The storage unit 120 has a function of storing a program executed by the processing unit 130. The storage unit 120 may also have a function of storing data created by the processing unit 130 (e.g., calculation results, analysis results, inference results), data input to the reception unit 110, etc.

[0209] The storage unit 120 preferably has a function of storing the instruction text in the prompt. For details of the instruction text in the prompt, please refer to the explanation in the above-mentioned <Accounting Work Support Method 1>.

[0210] The storage unit 120 preferably has a database 121. The database 121 has a data table in which order information is registered. The information processing device 10 preferably has a function of retrieving data from the database 121 and a function of recording data in the database 121.

[0211] The information processing device 10 may have a database 121 outside the storage unit 120. The information processing device 10 may have a function to retrieve data from a database 121 that exists outside the storage unit 120, outside the information processing device 10, or outside the accounting work support system. The information processing device 10 may also have a function to record data in a database 121 that exists outside the storage unit 120, outside the information processing device 10, or outside the accounting work support system. The information processing device 10 may also have a function to retrieve data from both its own database and an external database. The information processing device 10 may also have a function to record data in both its own database and an external database.

[0212] A file server may be used instead of the database. For example, when using files stored in a file server, it is preferable that the database has paths to files stored in the file server.

[0213] The information processing device 10 may also have a function of extracting at least one piece of electronic data from electronic data managed by the information terminal 20 or an IMAP server.

[0214] The storage unit 120 includes at least one of a volatile memory and a non-volatile memory. Examples of the volatile memory include a dynamic random access memory (DRAM) and a static random access memory (SRAM). Examples of the non-volatile memory include a resistive random access memory (ReRAM), a phase change random access memory (PRAM), a ferroelectric random access memory (FeRAM), a magnetoresistive random access memory (MRAM), and a flash memory. The storage unit 120 may also include at least one of NOSRAM (registered trademark) and DOSRAM (registered trademark). The storage unit 120 may also include a recording media drive. Examples of recording media drives include a hard disk drive (HDD) and a solid state drive (SSD).

[0215] NOSRAM is an abbreviation for "Nonvolatile Oxide Semiconductor Random Access Memory (RAM)." NOSRAM refers to a memory in which memory cells are two-transistor (2T) or three-transistor (3T) gain cells and transistors (also called OS transistors) that use metal oxide in their channel formation regions. OS transistors have an extremely small leakage current, i.e., a current that flows between the source and drain in an off state. NOSRAM can be used as a nonvolatile memory by retaining a charge corresponding to data in the memory cell using its extremely small leakage current characteristic. In particular, NOSRAM can read stored data without destroying it (nondestructive readout), making it suitable for arithmetic processing in which only data read operations are repeated a large number of times. NOSRAM can increase its data capacity by stacking layers, and therefore can be used as a large-scale cache memory, main memory, or storage memory to improve the performance of semiconductor devices.

[0216] DOSRAM is an abbreviation for "Dynamic Oxide Semiconductor RAM" and refers to a RAM having 1T (transistor) 1C (capacitor) type memory cells. DOSRAM is a DRAM formed using OS transistors, and is a memory that temporarily stores information sent from an external device. DOSRAM is a memory that takes advantage of the low off-state current of OS transistors.

[0217] In this specification and the like, a metal oxide refers to an oxide of a metal in a broad sense. Metal oxides are classified into oxide insulators, oxide conductors (including transparent oxide conductors), oxide semiconductors (also referred to as oxide semiconductors or simply as OSs), and the like. For example, when a metal oxide is used for a semiconductor layer of a transistor, the metal oxide may be referred to as an oxide semiconductor.

[0218] The metal oxide included in the channel formation region preferably contains indium (In). When the metal oxide included in the channel formation region contains indium, the carrier mobility (electron mobility) of the OS transistor is increased. Furthermore, the metal oxide included in the channel formation region is preferably an oxide semiconductor containing element M. The element M is preferably at least one of aluminum (Al), gallium (Ga), and tin (Sn). Other elements applicable to element M include boron (B), silicon (Si), titanium (Ti), iron (Fe), nickel (Ni), germanium (Ge), yttrium (Y), zirconium (Zr), molybdenum (Mo), lanthanum (La), cerium (Ce), neodymium (Nd), hafnium (Hf), tantalum (Ta), and tungsten (W). However, a combination of two or more of the above elements may be used as element M. The element M is, for example, an element having a high bond energy with oxygen. For example, it is an element having a higher bond energy with oxygen than indium. The metal oxide contained in the channel formation region is preferably a metal oxide containing zinc (Zn), since zinc-containing metal oxides may be easily crystallized.

[0219] The metal oxide contained in the channel formation region is not limited to a metal oxide containing indium, but may be, for example, a metal oxide containing zinc but not indium, such as zinc tin oxide or gallium tin oxide, a metal oxide containing gallium, or a metal oxide containing tin.

[0220] [Processing Unit 130] The processing unit 130 has a function of performing processes such as calculation, analysis, and inference using data supplied from at least one of the reception unit 110, the storage unit 120, and the database 121. The processing unit 130 can supply the created data (e.g., calculation results, analysis results, and inference results) to at least one of the storage unit 120, the database 121, and the output unit 140.

[0221] The processing unit 130 has a function of acquiring data from one or both of the storage unit 120 and the database 121. The processing unit 130 also has a function of recording or registering data in one or both of the storage unit 120 and the database 121. For example, the processing unit 130 has a function of registering order information in a data table in the database 121.

[0222] The processing unit 130 has a function of retrieving multiple first emails by searching emails based on order information. The order information is data received by the reception unit 110 and corresponds to data DA1. Data DA1 can also be considered order information registered in a record. For details of the first emails, please refer to the explanation in the above-mentioned <Accounting Work Support Method 1>. Furthermore, the processing unit 130 has a function of extracting third text from a file attached to the first email if the file is attached. For details of the third text, please refer to the explanation in the above-mentioned <Accounting Work Support Method 2>.

[0223] The processing unit 130 has a function of creating a prompt. For example, the processing unit 130 has a function of creating a first prompt and a second prompt. The created prompt corresponds to data DA2. For details of the first prompt and the second prompt, please refer to the explanation in the above-mentioned <Accounting Work Support Method 1>, etc.

[0224] The processing unit 130 has a function of acquiring a response sentence by sending a prompt to the language model via the transmission path 150, the output unit 140, and the network 30. For example, the processing unit 130 has a function of acquiring a first response sentence corresponding to a first prompt by sending a first prompt to the language model via the transmission path 150, the output unit 140, and the network 30. The processing unit 130 also has a function of acquiring a second response sentence corresponding to a second prompt by sending a second prompt to the language model via the transmission path 150, the output unit 140, and the network 30. The acquired response sentence corresponds to data DB1. For details of the first response sentence and the second response sentence, please refer to the explanations in the above-mentioned <Accounting Work Support Method 1>, etc.

[0225] The processing unit 130 has a function of extracting at least one second email from the plurality of first emails based on the first response sentence. For details of the second email, please refer to the explanation in the above-mentioned <Accounting Work Support Method 1>.

[0226] The processing unit 130 has a function of adding at least one second email to a record in which order information is registered. The processing unit 130 also has a function of updating the record in which order information is registered based on the second response. The record to which the second email has been added or the record updated based on the second response corresponds to data DB2. Data DB2 is the record to which the email has been added or the record updated based on the second response, so it can also be called updated data DA1.

[0227] The processing unit 130 may include, for example, an arithmetic circuit, a central processing unit (CPU), and a graphics processing unit (GPU).

[0228] The processing unit 130 may have a microprocessor such as a DSP (Digital Signal Processor). The microprocessor may be implemented by a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array) or an FPAA (Field Programmable Analog Array). The processing unit 130 may also have a quantum processor. The processing unit 130 can perform various data processing and program control by interpreting and executing instructions from various programs using the processor. Programs that can be executed by the processor are stored in at least one of the memory area of ​​the processor and the storage unit 120.

[0229] The processing unit 130 may include a main memory. The main memory may include at least one of a volatile memory such as a random access memory (RAM) and a non-volatile memory such as a read-only memory (ROM). The main memory may also include at least one of the above-mentioned NOSRAM and DOSRAM.

[0230] The RAM may be, for example, a DRAM or an SRAM, and a virtual memory space is allocated and used as a working space for the processing unit 130. The operating system, application programs, program modules, program data, lookup tables, and the like stored in the storage unit 120 are loaded into the RAM for execution. The data, programs, and program modules loaded into the RAM are each directly accessed and operated by the processing unit 130.

[0231] The ROM can store a BIOS (Basic Input / Output System), firmware, etc., which do not require rewriting. Examples of ROM include mask ROM, OTPROM (One Time Programmable Read Only Memory), and EPROM (Erasable Programmable Read Only Memory). Examples of EPROMs include UV-EPROMs (Ultra-Violet Erasable Programmable Read Only Memories), which allow stored data to be erased by exposure to ultraviolet light, EEPROMs (Electrically Erasable Programmable Read Only Memories), and flash memories.

[0232] The processing unit 130 can include at least one of an OS transistor, a transistor having silicon in a channel formation region (Si transistor), and a transistor having a layered two-dimensional material such as graphene or transition metal dichalcogenide (TMD) in a channel formation region.

[0233] The processing unit 130 preferably includes an OS transistor. Because an OS transistor has an extremely low off-state current, using the OS transistor as a switch for retaining charge (data) flowing into a capacitor functioning as a memory element can ensure a long data retention period. By utilizing this characteristic in at least one of the register and cache memory of the processing unit, the processing unit can be operated only when necessary and can be turned off at other times by saving information from the previous processing in the memory element. In other words, normally-off computing is possible, enabling low power consumption in the accounting support system.

[0234] It is preferable that the information processing device 10 uses AI for at least some of its processing.

[0235] In particular, it is preferable that the information processing device 10 uses an artificial neural network (ANN, hereinafter also simply referred to as a neural network). A neural network is realized by a circuit (hardware) or a program (software).

[0236] In this specification, a neural network refers to a general model that mimics the neural circuit network of a living organism, determines the connection strength between neurons through learning, and has problem-solving capabilities. A neural network has an input layer, an intermediate layer (hidden layer), and an output layer.

[0237] In this specification and the like, when discussing neural networks, determining the connection strengths (also called weighting coefficients) between neurons from existing information may be referred to as "learning."

[0238] In this specification and the like, the act of constructing a neural network using connection strengths obtained by learning and deriving a new conclusion from it may be referred to as "inference."

[0239] [Output Unit 140] The output unit 140 outputs data based on the processing result of the processing unit 130. The output unit 140 can supply at least one of the calculation result, analysis result, and inference result of the processing unit 130 to an external device of the information processing device 10.

[0240] The output unit 140 can output data to the outside of the information processing device 10. For example, the output unit 140 outputs data to the information processing device 40 via the network 30. The output unit 140 can output data created by the processing unit 130 based on data received from the information terminal 20 to the information processing device 40. The data output to the information processing device 40 corresponds to the data DA2 described above.

[0241] Furthermore, for example, the output unit 140 outputs data to the information terminal 20 via the network 30. The output unit 140 can output data created by the processing unit 130 based on data received from the information processing device 40 to the information terminal 20. Alternatively, the output unit 140 can output the data itself received from the information processing device 40 to the information terminal 20. The data output to the information terminal 20 corresponds to the data DB 2 described above.

[0242] The data can be output, for example, by displaying it on the display screen of the information terminal 20, or by outputting a file in a format such as txt, docx, xml, pdf, or csv. For example, the order information registered in the record is displayed on the display unit of the information terminal 20.

[0243] [Transmission Path 150] The transmission path 150 has a function of transmitting data. Data can be transmitted and received between the reception unit 110, the storage unit 120, the processing unit 130, and the output unit 140 via the transmission path 150.

[0244] <Configuration Example 2 of Accounting Work Support System> Fig. 16 shows another example of the accounting work support system of this embodiment. In the accounting work support system shown in Fig. 16, an information processing device 10 is connected to a plurality of information terminals 20 via a network 30. The information processing device 10 is also connected to an information processing device 40 via a network 31. In Fig. 16, the above-mentioned information terminals 20a to 20d are shown as examples of the information terminals 20.

[0245] A user of the accounting work support system can access the information processing device 10 from an information terminal 20. The user can then receive services using the accounting work support method according to one aspect of the present invention.

[0246] The flow of data between the information processing device 10 and the information processing device 40 in the accounting work support system shown in FIG. 16 can be understood by referring to the description of FIG. 13 by replacing the network 30 with the network 31.

[0247] An organization such as a company that provides a service using the accounting work support method according to one aspect of the present invention (hereinafter also referred to as a service provider) can provide the service using the information processing device 10. The service provider does not need to prepare the information processing device 40 on its own, but can use the information processing device 40 via the network 31.

[0248] Network 30 is a computer network smaller than network 31. Typically, network 31 is a global network, and network 30 is a local network. It is preferable to use the Internet, which is the foundation of the World Wide Web (WWW), as network 31. It is preferable to use an intranet or an extranet as network 30. In this embodiment, an example will be mainly described in which the Internet is used as network 31 and an intranet is used as network 30.

[0249] Alternatively, the network 30 may be a computer network such as a PAN, LAN, CAN, MAN, WAN, or GAN.

[0250] The accounting work support system shown in FIG. 16 is suitable for a case where the user who uses the information terminal 20 and the service provider belong to the same organization such as a company.

[0251] For example, it is preferable that data is transmitted and received between the information terminal 20 and the information processing device 10 using a network 30 established within an organization such as a company. It is also preferable that data is transmitted and received between the information processing device 10 and the information processing device 40 using a network 31 (typically the Internet). This allows applications, tools, services, etc. provided outside the organization to be used more safely than when the information terminal 20 is directly connected to the information processing device 40 via the Internet.

[0252] <Configuration Example 3 of Accounting Work Support System> Fig. 17 shows another example of the accounting work support system of this embodiment. In the accounting work support system shown in Fig. 17, an information processing device 10 is connected to an information processing device 40 and a plurality of information terminals 20 via a network 31. In Fig. 17, the above-mentioned information terminals 20a to 20d are shown as examples of the information terminals 20.

[0253] The flow of data in the accounting work support system shown in FIG. 17 can be understood by referring to the explanation of FIG. 13 by replacing network 30 with network 31.

[0254] A user of the accounting work support system can access the information processing device 10 from an information terminal 20. The user can then receive services using the accounting work support method according to one aspect of the present invention.

[0255] The accounting work support system shown in FIG. 17 is suitable for a case where the user who uses the information terminal 20 and the service provider belong to different organizations such as companies.

[0256] The information processing device 10 is connected to an information terminal 20 used by a user who receives a service via a network 31 (typically, the Internet). A service provider can provide the service using the information processing device 10. Data is also transmitted and received between the information processing device 10 and the information processing device 40 using the network 31 (typically, the Internet). This allows applications, tools, services, etc. using the information processing device 40 to be used more safely than when the information terminal 20 is directly connected to the information processing device 40 via the Internet.

[0257] <Configuration Example 4 of Accounting Work Support System> Figure 18 shows another example of the accounting work support system of this embodiment. The accounting work support system of this embodiment has an information processing device 11, an information processing device 15, an information terminal 20, and an information processing device 40. In the accounting work support system shown in Figure 18, the information processing device 11 is connected to the information processing device 15 and multiple information terminals 20 via a network 30. The information processing device 11 is also connected to the information processing device 40 via the network 30. In Figure 18, information terminals 20a and 20b are shown as examples of the information terminals 20.

[0258] 18 is mainly used to relay transmission and reception of data between the information terminal 20 and the information processing device 40. Then, in the information processing device 15, data DA2 is created using the data DA1.

[0259] First, the user inputs data DA1 to the information processing device 11 using the information terminal 20. The data DA1 is transmitted from the information terminal 20 to the information processing device 11 via the network 30.

[0260] Next, the information processing device 11 transmits the data DA1 to the information processing device 15 via the network 30 .

[0261] Next, the information processing device 15 creates data DA2 using the data DA1. The information processing device 15 is configured to execute the various processes performed by the information processing device 10 described above.

[0262] Next, the information processing device 15 transmits the data DA2 to the information processing device 11 via the network 30.

[0263] The information processing device 11 may be configured to create the data DA2. For example, the information processing device 15 may search for emails to obtain a first email, and then input the first email into the information processing device 11, which then creates the first prompt.

[0264] Next, the information processing device 11 transmits the data DA2 to the information processing device 40 via the network 30.

[0265] Next, the data DB1 is transmitted from the information processing device 40 to the information processing device 11 via the network 30. The data DB1 is a response sentence corresponding to the data DA2 input to the information processing device 40.

[0266] Next, the information processing device 11 transmits the data DB2 to the information terminal 20 via the network 30. The data DB2 is data based on the data DB1 input to the information processing device 11. As a result, the user can obtain the data DB2, which is the data DA1 updated based on the data DB1.

[0267] The data DB2 is created by the information processing device 11 or the information processing device 15. For example, the information processing device 11 may input the data DB1 input from the information processing device 40 to the information processing device 15. Then, the information processing device 15 may create the data DB2 based on the data DB1.

[0268] The accounting work support system shown in FIG. 18, like the accounting work support system shown in FIG. 13, is suitable when the user who uses the information terminal 20 and the service provider belong to the same organization such as a company.

[0269] For example, it is preferable that data be transmitted and received between the information terminal 20 and the information processing device 11 using a network 30 established within an organization such as a company. It is also preferable that data be transmitted and received between the information processing device 15 and the information processing device 11 using a network 30 established within an organization such as a company. It is also preferable that data be transmitted and received between the information processing device 11 and the information processing device 40 using the network 30. This allows data to be transmitted and received more securely than when the information terminal 20 is connected to the information processing device 40 via a global network such as the Internet. It is also possible to prevent confidential information within the organization from leaking to the outside.

[0270] Furthermore, connecting the information processing device 15 to the Internet may result in the risk of external cyber attacks, etc. In the accounting work support system shown in Fig. 18, the information processing device 15 does not need to be directly connected to the information processing device 40 via the Internet, which is preferable from the viewpoint of security.

[0271] Fig. 19 is a block diagram of the accounting work support system shown in Fig. 18. In Fig. 19, an example of the configuration of the information processing device 11 and the information processing device 15 will be mainly described.

[0272] The information processing device 11 includes a receiving unit 110, a storage unit 120, a processing unit 130, an output unit 140, and a transmission path 150. The storage unit 120 preferably includes a database 121.

[0273] For the reception unit 110, storage unit 120, output unit 140, and transmission path 150 of the information processing device 11, and the database 121 of the storage unit 120, the descriptions of the information processing device 10 can be referred to. The following mainly describes the differences from the information processing device 10.

[0274] The information processing device 15 includes a receiving unit 115 , a storage unit 125 , a processing unit 135 , an output unit 145 , and a transmission path 155 .

[0275] The receiving unit 110 receives data from the information terminal 20 , the information processing device 40 , and the output unit 145 via the network 30 .

[0276] The receiving unit 115 receives data from the output unit 140 via the network 30 .

[0277] The information supplied to the reception unit 115 is supplied to one or both of the storage unit 125 and the processing unit 135 via the transmission path 155 .

[0278] The storage unit 125 has a function of storing a program executed by the processing unit 135. The storage unit 125 may also have a function of storing data created by the processing unit 135 (e.g., calculation results, analysis results, inference results), data input to the reception unit 115, etc. The storage unit 125 may have various configurations applicable to the storage unit 120.

[0279] The processing unit 135 has a function of performing processes such as calculation, analysis, and inference using data supplied from one or both of the reception unit 115 and the storage unit 125. The processing unit 135 can supply the created data (e.g., calculation results, analysis results, and inference results) to one or both of the storage unit 125 and the output unit 145. The processing unit 135 can use various configurations applicable to the processing unit 130.

[0280] The processing unit 135 has a function of acquiring data from the storage unit 125 .

[0281] The information processing device 15 preferably uses AI for at least a part of the processing, and preferably uses a neural network.

[0282] The output unit 140 outputs data to the reception unit 115 , the information terminal 20 , and the information processing device 40 via the network 30 .

[0283] The output unit 145 outputs information based on the processing result of the processing unit 135. The output unit 145 can supply at least one of the calculation result, analysis result, and inference result of the processing unit 135 to an external device of the information processing device 15.

[0284] The output unit 145 outputs the data to the reception unit 110 via the network 30 .

[0285] The transmission path 155 has a function of transmitting data. Data can be transmitted and received between the reception unit 115, the storage unit 125, the processing unit 135, and the output unit 145 via the transmission path 155.

[0286] The processing unit 130 and the processing unit 135 can each include at least one of a transistor having a metal oxide in a channel formation region (OS transistor), a transistor having silicon in a channel formation region (Si transistor), and a transistor having a layered two-dimensional material such as graphene or TMD in a channel formation region.

[0287] As in the configuration of the above-mentioned <Configuration Example 2 of the Accounting Work Support System>, the information processing device 11 may be connected to the information processing device 40 via the network 31. With such a configuration, the service provider does not need to prepare the information processing device 40 on their own, but can use the information processing device 40 via the network 31. Also, this is preferable from the viewpoint of security, since the information processing device 15 does not need to be directly connected to the information processing device 40 via the Internet.

[0288] One aspect of the present invention provides an accounting work support method or system that utilizes a language model. One aspect of the present invention provides an accounting work support method or system that utilizes a language model to extract emails related to orders. One aspect of the present invention provides an accounting work support method or system that utilizes a language model to select and register information in a database. One aspect of the present invention provides an accounting work support method or system that reduces variation in the amount of information registered in a database depending on the person in charge. One aspect of the present invention provides a highly convenient accounting work support method or system. One aspect of the present invention provides a novel accounting work support method or system.

[0289] A plurality of configuration examples shown in this embodiment mode can be combined as appropriate.

[0290] 10: Information processing device, 11: Information processing device, 15: Information processing device, 20: Information terminal, 20a: Information terminal, 20b: Information terminal, 20c: Information terminal, 20d: Information terminal, 21: Housing, 30: Network, 31: Network, 40: Information processing device, 110: Reception unit, 115: Reception unit, 120: Storage unit, 121: Database, 125: Storage unit, 130: Processing unit, 135: Processing unit, 140: Output unit, 145: Output unit, 150: Transmission path, 155: Transmission path, 210: Email, 220[1]: Email, 220[2]: Email, 220[i]: Email, 220[n]: Email, 220: Email, 221[i]: File, 230: Email, 321[1]: Text, 321[2]: Text Text, 321[i]: Text, 321[n]: Text, 321: Text, 322[1]: Text, 322[i]: Text, 322[n]: Text, 322: Text, 331: Text, 332: Text, 410[i]: Prompt, 410: Prompt, 411[i]: Instruction, 411: Instruction, 412: List, 420[i]: Response, 420: Response, 421[i]: Text, 421: Text, 430: Prompt, 431: Instruction, 432: Object, 433: Text, 440: Response, 441: Text, 510: Data table, 511[1]: Item, 511[2]: Item, 511[3]: Item, 511[4]: Item, 511: Item, 513: Record, 600: Area

Claims

An accounting operation support method having first to tenth steps, In the first step, the information processing device accepts order information; In the second step, the information processing device registers the order information in a data table; In the third step, the information processing device retrieves a plurality of first e-mails by searching e-mails based on the order information; In the fourth step, the information processing device creates a first prompt; the first prompt comprises a list and a first instruction; the list includes a first text included in each of the plurality of first emails; the first instruction is an instruction for determining whether the first text is related to the order information; In the fifth step, the information processing device transmits the first prompt to a language model via a network to obtain a first response sentence including a result of the determination; In the sixth step, the information processing device extracts at least one second e-mail from the plurality of first e-mails based on the first response sentence, the second email is an email determined to be related to the order information based on the first text; In the seventh step, the information processing device adds the at least one second email to a record in the data table in which the order information is registered; In the eighth step, the information processing device generates a second prompt; the second prompt comprises second text included in the at least one second email and a second instruction; the second instruction statement is an instruction statement for extracting information corresponding to each item of the data table from the second text, In the ninth step, the information processing device transmits the second prompt to the language model via the network to obtain a second response sentence including the extracted information; In the tenth step, the information processing device updates the record using the extracted information based on the second response sentence. 　 In claim 1, An accounting work support method, wherein the first instruction sentence includes at least a portion of the order information. 　 In claim 1, a step between the fourth step and the fifth step, in which the information processing device extracts a third text from the file when the first email is attached; The accounting work support method, wherein the list includes the third text. 　 In claim 1, An accounting work support method, in which, if the order information registered in the record does not match the information extracted in the ninth step, the mismatched order information is highlighted and displayed on the display unit of the information terminal. 　 In claim 1, An accounting work support method comprising, between the fourth step and the fifth step, a step in which the information processing device determines whether the first prompt contains a word that is personal information or confidential information, and replaces the word contained in the first prompt with another word that belongs to the same field as the word belongs to. 　 The apparatus includes a database, a reception unit, and a processing unit, the database includes a data table; The reception unit has a function of receiving order information, The processing unit includes: A process of registering the order information in the data table; obtaining a plurality of first e-mails by searching e-mails based on the order information; creating a first prompt; transmitting the first prompt to a language model over a network to obtain a first response sentence; A process of extracting at least one second e-mail from the plurality of first e-mails based on the first response sentence; adding the at least one second email to a record in the data table in which the order information is registered; generating a second prompt; transmitting the second prompt to the language model over the network to obtain a second response sentence; and updating the record based on the second response sentence. the first prompt comprises a list and a first instruction; the list includes a first text included in each of the plurality of first emails; the first instruction is an instruction for determining whether the first text is related to the order information; the first response sentence includes a result of the determination; the second email is an email determined to be related to the order information based on the first text; the second prompt comprises second text included in the at least one second email and a second instruction; the second instruction statement is an instruction statement for extracting information corresponding to each item of the data table from the second text, An accounting work support system, wherein the process of updating the record is performed using extracted information contained in the second response sentence. 　 In claim 6, An accounting work support system, wherein the first instruction sentence includes at least a portion of the order information. 　 In claim 6, the processing unit is configured to perform a process of extracting third text from a file attached to the first email if the file is attached to the first email; The list includes the third text. 　 In claim 6, an information processing device having the database, the reception unit, and the processing unit, and an information terminal; An accounting work support system, wherein the order information registered in the record is displayed on a display unit of the information terminal. 　 In claim 9, An accounting work support system in which, if the order information does not match the extracted information contained in the second response sentence, the mismatched order information is displayed in a highlighted manner on the display unit of the information terminal.

Images