Information processing method and information processing device

Abstract

The present invention provides an information processing method excellent in convenience, usefulness, and reliability. The information processing method includes: first processing for selecting an account title related to transaction information on the basis of an explanatory sentence of the account title; and second processing for correcting the explanatory sentence of the account title so that the account title can be correctly selected in the first processing when the selection of the account title is not appropriate. The first processing and the second processing are performed using a language model. When an account title cannot be selected in the first processing, the reason why the account title cannot be selected is presented, and after the additional information is received, the first processing is performed again.

Description

Information processing method and information processing device

[0001] One aspect of the present invention relates to an information processing method, an information processing device, and an information processing system. One aspect of the present invention relates to an information processing method and an information processing system using a language model. Specifically, one aspect of the present invention relates to an accounting work support method and an accounting work support system using 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 accounting work, account items are used to classify transaction details and prepare financial statements. Patent Document 1 discloses a method for assigning account items to transactions based on the contents of transaction receipts and journal entry rules.

[0004] In recent years, the development of language models using neural networks has been actively carried out, and large-scale language models (LLMs) in particular have attracted 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 a dialogue model that responds to user instructions, for example. Non-Patent Document 1 discloses GPT-4 (Generative Pre-trained Transformer 4) (registered trademark) as a large-scale language model, and also discloses 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 2020-16914 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] Account titles cannot be determined solely based on the name of the traded product, but must also be determined based on factors such as the purpose of the transaction. For this reason, even if the assignment of account titles is automated, there are cases where an account title cannot be selected due to insufficient information, for example. Furthermore, even if an account title can be selected, the selected account title may not be appropriate.

[0009] In view of the above, an object of one embodiment of the present invention is to provide an information processing method, an information processing device, and an information processing system that can present missing information when an account cannot be selected due to a lack of information. Another object of one embodiment of the present invention is to provide an information processing method, an information processing device, and an information processing system that can increase the probability of selecting a valid account in subsequent processing when an automatically selected account is invalid. Another object of one embodiment of the present invention is to provide an information processing method, an information processing device, and an information processing system that are highly convenient, useful, and reliable. Another object of one embodiment of the present invention is to provide a novel information processing method, an information processing device, and an information processing system.

[0010] 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.

[0011] In one aspect of the present invention, a first step involves accepting transaction information; a second step involves creating a first prompt having the transaction information, a list, and a first instruction sentence, the list having a plurality of first account items and a first explanation that is an explanation of the first account items, the first instruction sentence being an instruction sentence for selecting a second account item that is an account item related to the transaction information from the plurality of first account items based on the transaction information and the first explanation sentence; a third step involves creating a first response sentence using a language model based on the first prompt; a fourth step involves determining whether or not a second account item has been selected in the first response sentence; if the second account item has not been selected, a fifth step involves prompting the user to input additional information by presenting a reason for the second account item not being selected that is included in the first response sentence; a sixth step involves accepting the additional information; and after performing the sixth step, performing the second to fourth steps again; if the second account item has been selected, performing the seventh step. In step 8, the user is prompted to determine whether the second account item is valid, and if the second account item is valid, in step 9, the second account item is registered, and if the second account item is not valid, in step 9, the second account item is corrected by the user and then registered as a third account item, and in step 10, the transaction information, the second account item, the third account item, the second explanation which is an explanation of the second account item, the third explanation which is an explanation of the third account item, and and a second instruction sentence, wherein the second explanation sentence and the third explanation sentence are selected from the first explanation sentence, and the second instruction sentence is an instruction sentence for creating one or both of a revision suggestion for the second explanation sentence and a revision suggestion for the third explanation sentence based on the transaction information, the second account item, and the third account item, and in an eleventh step, an information processing method is used to create a second response sentence using a language model based on the second prompt, the second response sentence including one or both of the revision suggestion for the second explanation sentence and the third explanation sentence.

[0012] Alternatively, in the above aspect, an information processing method using a database may be provided, in which a data table including a list is registered in the database, and in an eighth step, a second account item is registered in the database as an account item related to transaction information, and in a ninth step, a third account item is registered in the database as an account item related to transaction information, and in a twelfth step, the first explanation registered in the database may be corrected based on one or both of the suggested corrections to the second explanation and the suggested corrections to the third explanation.

[0013] Alternatively, one aspect of the present invention includes a database, a reception unit, and a processing unit, wherein the database includes a data table, the data table includes a plurality of first account items and a first explanation that is an explanation of the first account items, the reception unit has a function of receiving transaction information, additional information, and a correction of a second account item related to the transaction information, and supplying the same to the processing unit, and the processing unit has a function of creating a first prompt including the transaction information, a list that is at least a part of the data table, and a first instruction statement, and the first instruction statement includes the transaction information, the first account item, and the first explanation statement. and an instruction sentence for selecting from a plurality of first account items based on the explanation sentence, and the processing unit has a function of determining whether or not a second account item is selected in a first response sentence acquired by transmitting a first prompt to a language model via a network, and the processing unit has a function of presenting to the user, if the second account item is not selected, the reason why the second account item is not selected, which is included in the first response sentence, and creating a first prompt based on the additional information, and if the second account item is selected, the processing unit determines whether or not the second account item is appropriate. The processing unit has a function of registering the second account item in the database as an account item related to the transaction information when it receives data indicating that the second account item is valid, and a function of registering the second account item in the database as an account item related to the transaction information when it receives data indicating that the second account item is invalid and that the second account item is to be corrected, making the correction made by the user to the second account item a third account item and registering the third account item in the database as an account item related to the transaction information, and the processing unit has a function of registering the transaction information, the second account item, the third account item, and a second explanation which is an explanation of the second account item The information processing device has a function of creating a second prompt having a clear sentence, a third explanation sentence which is an explanation sentence for a third account item, and a second instruction sentence, wherein the second explanation sentence and the third explanation sentence are selected from the first explanation sentence, and the second instruction sentence is an instruction sentence for creating one or both of a revision proposal for the second explanation sentence and a revision proposal for the third explanation sentence based on the transaction information, the second account item, and the third account item, and the processing unit has a function of modifying the first explanation sentence based on a second response sentence obtained by sending the second prompt to a language model via a network.

[0014] Alternatively, in the above aspect, if the second account item is selected, the first response sentence may include the reason for selecting the second account item.

[0015] Alternatively, in the above aspect, the transaction information may include the product name and the purpose of the transaction.

[0016] According to one aspect of the present invention, it is possible to provide an information processing method, an information processing device, and an information processing system that can present missing information when an account cannot be selected due to insufficient information. Furthermore, according to one aspect of the present invention, it is possible to provide an information processing method, an information processing device, and an information processing system that can increase the probability of selecting an appropriate account in subsequent processing when an automatically selected account is inappropriate. Furthermore, according to one aspect of the present invention, it is possible to provide an information processing method, an information processing device, and an information processing system that are highly convenient, useful, and reliable. Furthermore, according to one aspect of the present invention, it is possible to provide a novel information processing method, an information processing device, and an information processing system.

[0017] 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.

[0018] FIG. 1 is a schematic diagram showing an example of an information processing system. FIG. 2 is a flowchart showing an example of an information processing method. FIG. 3 is a flowchart showing an example of an information processing method. FIG. 4A is a schematic diagram showing an example of transaction information. FIG. 4B is a schematic diagram showing an example of a data table. FIG. 5A is a schematic diagram showing an example of a prompt. FIGS. 5B1 and 5B2 are schematic diagrams showing examples of response sentences. FIGS. 6A and 6B are schematic diagrams showing examples of a screen layout. FIG. 7 is a schematic diagram showing an example of a screen layout. FIG. 8A is a schematic diagram showing an example of a prompt. FIG. 8B is a schematic diagram showing an example of a response sentence. FIG. 9 is a flowchart showing an example of an information processing method. FIG. 10A is a schematic diagram showing an example of a prompt. FIG. 10B is a schematic diagram showing an example of a response sentence. FIG. 11 is a block diagram showing an example of an information processing system. FIG. 12 is a schematic diagram showing an example of an information processing system.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] In this specification, the ordinal numbers such as "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, an ordinal number assigned to a component in one part of this specification may not match an ordinal number assigned to the same component in another part of this specification or in the claims.

[0023] 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.

[0024] 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.

[0025] In this embodiment, an information processing method, an information processing device, and an information processing system according to one embodiment of the present invention will be described with reference to the drawings. Specifically, the information processing method, the information processing device, and the information processing system can be used for accounting work.

[0026] In this specification and the like, an information processing method, an information processing device, and an information processing system that can be used for accounting work are referred to as an accounting work support method, an accounting work support device, and an accounting work support system, respectively.

[0027] An information processing system according to one embodiment of the present invention includes an information terminal, a first information processing device, and a second information processing device, which are connected to each other via a network. The second information processing device has a function of performing processing using a language model. An information processing method according to one embodiment of the present invention is an information processing method using the information processing system.

[0028] In accounting work, account items are used to classify transactions. Account items are registered in a database along with their descriptions. In other words, a data table containing account items and their descriptions is registered in the database.

[0029] In one aspect of the information processing method of the present invention, a first process is performed to select an account item from the data table based on transaction information and a description of the account item. The transaction information includes the name of the product or goods involved in the transaction, the purpose of the transaction, etc. The first process can be performed using a first information processing device and a second information processing device. Here, the first information processing device can extract a portion of the data included in the data table to create a list and provide the list to the second information processing device. This allows the second information processing device to select an account item related to the transaction information from the list.

[0030] In the first process, if an account item cannot be selected due to, for example, insufficient information, the reason for the selection can be presented to the user of the information terminal. This can prompt the user of the information terminal to input additional information. After the first information processing device accepts the additional information, the first information processing device and the second information processing device can perform the first process again. Furthermore, in the first process, if an account item is selected, the reason for the selection can be presented to the user of the information terminal. This makes it easier for the user of the information terminal to determine whether the selected account item is appropriate.

[0031] After the first process, the first information processing device can prompt the user of the information terminal to determine whether the selected account item is appropriate. If the selected account item is not appropriate and the user of the information terminal corrects the account item, a second process is performed. In the second process, the description of the account item is corrected so that the account item can be correctly selected in the first process. The second process can be performed using the first information processing device and the second information processing device. Here, the first information processing device can supply the second information processing device with transaction information, the account item and its description before correction, the corrected account item, and the description related to the account item. This allows the second information processing device to present a correction proposal for the description of the account item before correction and a correction proposal for the description of the corrected account item.

[0032] As described above, the information processing method of one embodiment of the present invention can automate the assignment of account items. Furthermore, when the second information processing device is unable to select an account item due to insufficient information, the missing information can be presented using the first information processing device. Furthermore, when the second information processing device selects an account item, the reason for the selection can be presented using the first information processing device. Furthermore, when the selected account item is inappropriate, the user of the information terminal can correct the account item, and the second information processing device can increase the probability of selecting an appropriate account item in subsequent processing. As described above, the information processing method of one embodiment of the present invention can be an information processing method that is highly convenient, useful, and reliable. Furthermore, the information processing device and information processing system of one embodiment of the present invention can be an information processing device and information processing system that are highly convenient, useful, and reliable.

[0033] <Configuration Example 1 of Information Processing System> Fig. 1 is a schematic diagram showing an example of an information processing system 10 that is an information processing system according to one aspect of the present invention. The information processing system 10 includes an information processing device 40, an information terminal 20, and an information processing device 50. In the information processing system 10, the information processing device 40, the information terminal 20, and the information processing device 50 are connected via a network 30. Fig. 1 shows information terminal 20a, information terminal 20b, information terminal 20c, and information terminal 20d as examples of the information terminal 20.

[0034] A user of the information processing system 10 can access the information processing device 40 from the information terminal 20. The user can then receive a service using an information processing method according to one aspect of the present invention, i.e., an information processing method using the information processing system 10. Specifically, the user of the information processing system 10 can be a user of the information terminal 20.

[0035] The user can use the information processing device 40, for example, by using dedicated software or an application installed on the information terminal 20. Alternatively, the user can use the information processing device 40, for example, from a web browser on the information terminal 20.

[0036] An organization such as a company that provides a service using an information processing 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 40 .

[0037] The information processing device 40 is a device that can execute processing using an information processing method according to one aspect of the present invention. In FIG. 1, a server computer is illustrated as an example.

[0038] The information processing device 40 has a reception unit 41, a processing unit 43, and a database 45. The reception unit 41 has a function of receiving data from outside the information processing device 40. The reception unit 41 receives data from the information terminal 20, for example, via the network 30. The reception unit 41 also receives data from the information processing device 50, for example, via the network 30. The data supplied to the reception unit 41 can be supplied to the processing unit 43, for example.

[0039] The processing unit 43 has a function of performing processes such as calculation, analysis, and inference, etc. The processing unit 43 has a function of performing processes such as calculation, analysis, and inference, etc., using data supplied from, for example, one or both of the receiving unit 41 and the database 45.

[0040] The information processing device 40, specifically the processing unit 43, may use artificial intelligence (AI) for at least some of the processing. In this case, it is preferable that the information processing device 40 particularly uses an artificial neural network (ANN, hereinafter simply referred to as a neural network) model. The neural network model is realized by a circuit (hardware) or a program (software).

[0041] In this specification, a neural network model refers to a general model that mimics the neural network of a living organism, determines the connection strength between neurons through learning, and has problem-solving capabilities. A neural network model has an input layer, an intermediate layer (hidden layer), and an output layer.

[0042] In this specification and the like, when describing a neural network model, determining the connection strengths (also called weighting coefficients) between neurons from existing information may be referred to as "learning."

[0043] In this specification and the like, constructing a neural network model using connection strengths obtained by learning and deriving a new conclusion from it may be referred to as "inference."

[0044] 1 shows an example in which the database 45 is provided in the information processing device 40, but the database 45 may be provided outside the information processing device 40, or outside the information processing system 10. Alternatively, the database may be provided in two or more of the following locations: inside the information processing device 40, outside the information processing device 40 and inside the information processing system 10, and outside the information processing system 10. For example, the database 45 may be provided in a distributed manner in two or more of the following locations: inside the information processing device 40, outside the information processing device 40 and inside the information processing system 10, and outside the information processing system 10.

[0045] The information processing device 40 can perform information processing such as calculations using data input from the information terminal 20 via the network 30. The information processing device 40 can transmit the results of the information processing to the information terminal 20 via the network 30. This can reduce the calculation load on the information terminal 20.

[0046] The information processing device 40 can input data to the information processing device 50 via the network 30. The information processing device 50 can perform information processing such as calculations using the input data. The information processing device 50 can transmit the results of the information processing to the information processing device 40 via the network 30. This reduces the calculation load on the information processing device 40. The information processing device 40 can then transmit the data received from the information processing device 50 or data created based on the data to the information terminal 20 via the network 30.

[0047] The information terminal 20 is an information terminal such as a computer used by a user, and may also be referred to as a client computer. FIG. 1 illustrates, as examples of the information terminal 20, information terminal 20a, which is a desktop computer; information terminal 20b, which is a notebook computer; information terminal 20c, which is a smartphone; and information terminal 20d, which is a tablet computer. Note that 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 40 is not particularly limited. While FIG. 1 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.

[0048] The information terminal 20 may have a display unit. This allows the information terminal 20 to present information to the user. The information terminal 20 may present, for example, the result of information processing by the information processing device 40 to the user. The information terminal 20 may also present the result of information processing by the information processing device 50 to the user.

[0049] 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 can 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. Furthermore, the information processing device 40 may be capable of performing processing using a natural language processing model using AI.

[0050] The information processing device 50 is a large computer such as a server computer or a supercomputer. The information processing device 50 is a computer with higher processing power than the information terminal 20. The information processing device 50 is also a computer with higher processing power than the information processing device 40. The information processing device 50 preferably has a function as a parallel computer. By using the information processing device 50 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 40 and the information processing device 50 have a function as a parallel computer, the information processing device 50 has higher processing power and can perform large-scale calculations than the information processing device 40.

[0051] The information processing device 50 can perform processing using a natural language processing model that uses AI. Examples of natural language processing models that use AI include BERT (Bidirectional Encoder Representations from Transformers) and T5 (Text-to-Text Transfer Transformer). The information processing device 50 can also perform processing using a model that uses 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).

[0052] The information processing device 50 can execute processing using a general-purpose AI model that can perform various tasks. When the information processing device 40 and the information processing device 50 can each execute processing using a model that uses a large-scale language model, the information processing device 50 can execute processing using a larger-scale AI model than the information processing device 40.

[0053] An intranet or an extranet can be used as the network 30. The network 30 can be a computer network such as a personal area network (PAN), a local area network (LAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), or a global area network (GAN).

[0054] 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.

[0055] 2 and 3 are flowcharts illustrating an example of an information processing method according to one embodiment of the present invention. In FIG. 3, each process is performed by the information terminal 20, the information processing device 40, or the information processing device 50.

[0056] Before the process shown in step S11 in Fig. 2 is performed, step S31 shown in Fig. 3 is performed. In step S31, the user of the information terminal 20 inputs transaction information to the information processing device 40. In other words, in step S31, the information terminal 20 accepts the transaction information and transmits the accepted transaction information to the information processing device 40. The user of the information terminal 20 may be, for example, a person in charge of the transaction. Furthermore, if there are multiple people in charge of the transaction, the user of the information terminal 20 may be one of the people in charge. Furthermore, the user of the information terminal 20 may be different from the person in charge of the transaction.

[0057] 4A is a schematic diagram showing an example of transaction information 100, which is transaction information input to information processing device 40. Transaction information 100 includes item 101[1] and item 101[2]. In the example shown in FIG. 4A, item 101[1] is the product name, and item 101[2] is the purpose of the transaction. The user of information terminal 20 inputs data for each item 101.

[0058] 4A , items other than the items 101[1] and 101[2] may be included as items 101. Examples of the items 101 include the transaction number, the amount, the customer, the name of the sales representative, discount information, the quotation expiration date, and an email address.

[0059] In the example shown in FIG. 4A, item 101[1] is "aaa" and item 101[2] is "for bb." In other words, in the transaction information 100 shown in FIG. 4A, the product name is "aaa" and the transaction purpose is "for bb." Here, the product name is not limited to the name of a product, but can be, for example, the name of a service. Furthermore, the transaction purpose can be, for example, the use of the product represented by the product name, the reason for purchasing the product, the content of the service, or the reason for receiving the service. Note that transaction information 100 is not limited to information related to a transaction. In this case, for example, item 101[1] can be the name, and item 101[2] can be the purpose.

[0060] 4B is a schematic diagram illustrating an example of a data table 110 used in an information processing method according to one embodiment of the present invention. The data table 110 includes items 111, 112, and 113. The first row of the data table 110 represents an item name, and each row from the second row onward represents a record 115. FIG. 4B illustrates examples of record 115[1], record 115[2], and record 115[3]. The data table 110 can be registered in the database 45.

[0061] In this specification, a record refers to one of the units that make up a data table. One record can store one row of data.

[0062] 4B shows an example in which item 111 is a department, item 112 is an account, and item 113 is a description. Note that data table 110 may include items other than items 111, 112, and 113. Data table 110 may also include, for example, a code number assigned to each record.

[0063] Item 111 indicates, for example, the department in which a transaction is conducted. A department can be, for example, a department, branch office, sales office, or subsidiary within a company. Item 111 can also be a classification of an account item. In this case, item 111 included in record 115 can be, for example, any of assets, liabilities, net assets, revenue, and expenses. Item 111 can also be a classification of an account item. For example, the classification of an account item classified as "expenses" can be cost of sales, selling, general and administrative expenses, non-operating expenses, and extraordinary losses. Note that while FIG. 4B shows an example in which there is one type of item 111, data table 110 may include two or more types of item 111.

[0064] The item 112 can be an account title, as described above. The account title is assigned to the transaction information 100 and can be used to classify the transaction information 100. In an information processing method according to one aspect of the present invention, any of the items 112 included in the data table 110 is assigned to the transaction information 100. Note that the item 112 is not limited to an account title, as long as it can be used to classify the transaction information 100.

[0065] Item 113 may be a description of item 112. Item 113 may be a specific example of a transaction to which the account item, which is item 112, is assigned. In an information processing method according to one aspect of the present invention, item 112 is assigned to transaction information 100 based on transaction information 100 and item 113.

[0066] In the example shown in FIG. 4B, items 111, 112, and 113 of record 115[1] are respectively set to "CC1 portion," "DD1 expense," and "EE purchase price, expenses for FF, etc.". Also, items 111, 112, and 113 of record 115[2] are respectively set to "CC1 portion," "DD2 expense," and "GG participation fee, etc.". Furthermore, items 111, 112, and 113 of record 115[3] are respectively set to "CC2 portion," "DD3 expense," and "HH, etc.".

[0067] The following describes the processing from step S11 onward shown in Fig. 2. In Fig. 2 and the accompanying description, transaction information can be appropriately replaced with terms such as information, and account items can be appropriately replaced with terms such as classification.

[0068] [Step S11] In step S11, the information processing device 40 accepts the transaction information 100. Specifically, the reception unit 41 of the information processing device 40 accepts the transaction information 100 transmitted from the information terminal 20. In the example shown in Fig. 4A, the reception unit 41 accepts the transaction information 100 in which the item 101[1], which is the product name, is "aaa" and the item 101[2], which is the transaction purpose, is "for bb."

[0069] [Step S12] In step S12, the information processing device 40 generates a prompt 210. Specifically, the processing unit 43 of the information processing device 40 generates the prompt 210 for assigning an account item to the transaction information 100.

[0070] In this specification, a prompt corresponds to an input sentence that causes a language model to perform a desired operation. When a prompt is given, the language model generates a response sentence based on the prompt. The prompt includes an instruction sentence. The 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.

[0071] 5A is a schematic diagram showing an example of a prompt 210. The prompt 210 includes an instruction statement 211, an item 101[1], an item 101[2], and a list 120.

[0072] The instruction statement 211 includes an instruction to select an account item to be assigned to the transaction information 100. The instruction statement 211 also includes an instruction to present the reason if an account item cannot be selected. Furthermore, the instruction statement 211 may include an instruction to present the reason for the selection when selecting an account item. The instruction statement 211 can be, for example, "For the following item name and transaction object, please select an account item from the list of account items and explain the reason. Also, if there is insufficient information to make a selection, please explain the missing information."

[0073] The items 101[1] and 101[2] may be the items 101[1] and 101[2] included in the transaction information 100. In other words, the prompt 210 includes at least a portion of the transaction information 100. Note that the prompt 210 may include items 101 other than the item 101[1] that can be the product name and the item 101[2] that can be the transaction object.

[0074] The list 120 includes at least some of the information (records 115 and items) included in the data table 110 shown in Fig. 4B. In the example shown in Fig. 4B, the items included in the list 120 may be account items and descriptions of the account items.

[0075] In this specification, the account items and their explanations included in list 120 may be referred to as the first account items and the first explanations, respectively. That is, when item 112 is an account item and item 113 is an explanation of the account item, item 112 may be referred to as the first account item and item 113 as the first explanation.

[0076] The list 120 can include at least some of the records 115 included in the data table 110. For example, records 115 whose item 111 has desired content can be included in the list 120. For example, the user of the information terminal 20 can specify a department to specify the records 115 to be included in the list 120. Furthermore, records 115 of the department to which the user of the information terminal 20 belongs can be included in the list 120. The department to which the user of the information terminal 20 belongs can be determined, for example, based on login information. Note that the records 115 to be included in the list 120 may also be specified based on a predetermined search expression. For example, an item 111 may be specified based on a predetermined search expression, and records 115 including the specified item 111 may be included in the list 120.

[0077] In Fig. 5A, record 115 in which item 111 shown in Fig. 4B is the cc1 part, specifically items 112 and 113, are included in list 120. Specifically, items 112 and 113 included in record 115[1] shown in Fig. 4B and items 112 and 113 included in record 115[2] are included in list 120. Note that in Fig. 5A, items 112 and 113 included in record 115[1] are referred to as items 112[1] and 113[1], respectively. Also, in Fig. 5A, items 112 and 113 included in record 115[2] are referred to as items 112[2] and 113[2], respectively.

[0078] By narrowing down the items 112 and 113 to be included in the list 120 based on, for example, the item 111, it is possible to reduce the number of candidate account items to be assigned to the transaction information 100. This reduces the processing time required to assign account items while ensuring the accuracy of the account items assigned to the transaction information 100. It also prevents the amount of text in the prompt 210 from increasing, making it easier to obtain a response sentence from a language model in later processing. It also reduces the time required to generate a response sentence using a language model.

[0079] The list 120 preferably contains text. Figure 5A shows the list 120 in a non-tabular format. For example, the name of the item 112, "Account Title," is entered as a string in the prompt 210 for each item 112.

[0080] The instruction 211 is an instruction for selecting an item 112 related to the transaction information 100 from the items 112 included in the list 120. The item 112 can be selected based on the transaction information 100 and the item 113 included in the list 120.

[0081] [Step S13] In step S13, the information processing device 40 transmits the prompt 210 to the language model of the information processing device 50 via the network 30 (step S13a shown in FIG. 3). As a result, the information processing device 50 creates a response sentence 220 using the language model based on the prompt 210. The response sentence 220 includes, for example, an account item selected based on the information included in the prompt 210. The information processing device 50 transmits the response sentence 220 to the information processing device 40 via the network 30 (step S13b shown in FIG. 3). The reception unit 41 of the information processing device 40 receives the response sentence 220 transmitted from the information processing device 50. As a result, the information processing device 40 acquires the response sentence 220 (step S13c shown in FIG. 3).

[0082] 5B1 and 5B2 are schematic diagrams showing an example of a response sentence 220. The response sentence 220 includes an item 223 and an item 225. The item 223 can be the same type of item as the item 112. Therefore, the item 223 can be, for example, an account item. The item 225 can be, for example, the reason why an account item was not selected. Furthermore, the item 225 can be the reason for the selection of an account item if the account item was selected.

[0083] 5B1 is an example of a response sentence 220 when the language model is unable to select an account item, for example, due to a lack of information contained in one or both of items 101[1] and 101[2].

[0084] In the example shown in Figure 5B1, item 223 is marked "unselectable." Item 225 indicates the reason why the account item could not be selected, stating, "The account item cannot be identified because it is unclear what bb specifically refers to." As shown in Figure 5A, "bb" is included in item 101[2], which can be used as the transaction purpose, for example. Item 225 may also present, for example, multiple candidate account items along with the reasons.

[0085] 5B2 shows an example of a response sentence 220 when the language model selects an account item, specifically, when one account item is selected. In this case, item 223 can be the selected account item. Item 225 can be the reason for the selection.

[0086] In this specification and the like, the account item selected by the language model and its explanation may be referred to as the second account item and the second explanation, respectively.

[0087] In the example shown in Figure 5B2, item 223 is set to "dd1 expense." That is, this shows an example in which item 112[1] shown in Figure 5A is selected as the account item. Also, in the example shown in Figure 5B2, item 225 is set to "Due to ii, dd1 expense is the appropriate account item."

[0088] [Steps S14 to S21] In steps S14 and S15, the information processing device 40 determines whether an account item has been selected in the response sentence 220. Specifically, the processing unit 43 of the information processing device 40 determines whether an account item has been selected based on the contents of the response sentence 220. If it is determined that an account item has not been selected, in step S16, the information processing device 40 presents the reason why the account item has not been selected to the user of the information terminal 20. This causes the information processing device 40 to prompt the user of the information terminal 20 to input additional information. For example, the processing unit 43 of the information processing device 40 transmits the response sentence 220 and the additional information to the information terminal 20. This causes the response sentence 220 and the additional information to be displayed on the display unit of the information terminal 20.

[0089] FIG. 6A is a schematic diagram showing an example of a screen layout displayed on the display unit 23 of the information terminal 20 by performing step S15. FIG. 6A can also be considered an example of a graphical user interface (GUI) related to the information processing system 10. Forms, icons, tables, etc. in the drawings related to the GUI exemplified in this embodiment, such as FIG. 6A, are merely examples and are not particularly limited. The GUI can be configured as a web page that a user of the information terminal 20 accesses via a network. Alternatively, the GUI can be configured as a screen of a program application executed on the information terminal 20 used by the user.

[0090] As shown in FIG. 6A , the display unit 23 displays a response sentence 220 including items 223 and 225. The display unit 23 can also display transaction information 100 including items 101[1] and 101[2]. FIG. 6A illustrates an example in which an account item is not selected because it is unclear what the transaction purpose "bb" specifically refers to. In other words, the example illustrates an example in which an account item is not selected because there is insufficient information included in the transaction purpose. Therefore, in the example illustrated in FIG. 6A , a form 130 for inputting additional information is displayed in the area where item 101[2] is displayed. The form 130 may also be referred to as an input form, area, field, or input field. A sentence specifically indicating the additional information to be input into the form 130, such as "Please indicate what the transaction purpose "bb" specifically refers to," may also be displayed on the display unit 23.

[0091] 2 is performed, followed by step S32 shown in Fig. 3. In step S32, the user of the information terminal 20 inputs additional information 131 into, for example, the form 130. In other words, in step S32, the information terminal 20 accepts the additional information 131 and transmits the accepted additional information 131 to the information processing device 40.

[0092] 6B is a schematic diagram showing an example in which additional information 131 is input into the form 130 in step S32. FIG. 6B shows an example in which "cd1 and cd2 are listed as bb" is input as the additional information 131 into the form 130. Note that while FIG. 6B shows an example in which the additional information 131 is input as a sentence, one aspect of the present invention is not limited to this. The additional information 131 may be input in bullet points, for example. Furthermore, for example, only nouns may be input into the form 130.

[0093] After step S32 is performed, step S17 is performed. In step S17, the information processing device 40 receives additional information 131. Specifically, the reception unit 41 included in the information processing device 40 receives the additional information 131 transmitted from the information terminal 20. In the example shown in FIG. 6B , the reception unit 41 receives, as the additional information 131, "cd1 and cd2 are listed as bb."

[0094] After step S17 is performed, steps S12 to S15 are performed again. Specifically, steps S12 to S15 are performed again using the information processing device 40 and the information processing device 50.

[0095] If it is determined in steps S14 and S15 that an account item has been selected, in step S18, the information processing device 40 prompts the user of the information terminal 20 to determine whether the selected account item is appropriate. For example, the processing unit 43 of the information processing device 40 sends the response sentence 220 shown in FIG. 5B2 to the information terminal 20. As a result, the response sentence 220 shown in FIG. 5B2 is displayed on the display unit of the information terminal 20. Here, it is preferable that the response sentence 220 includes a reason for selecting the account item, such as item 225 shown in FIG. 5B2. This makes it easier for the user of the information terminal 20 to determine whether the selected account item is appropriate.

[0096] 7 is a schematic diagram showing an example of a screen layout displayed on the display unit 23 by performing step S18. As shown in FIG. 7, the display unit 23 displays transaction information 100 having items 101[1] and 101[2]. The display unit 23 also displays a response sentence 220 having items 223 and 225. The display unit 23 also displays a list 120 having items 112 and 113. The list 120 may be the list 120 included in the prompt 210 shown in FIG. 5A.

[0097] 7 shows an example in which the list 120 is displayed in a table format on the display unit 23. In the example shown in FIG. 7, the first line of the list 120 shows the item name.

[0098] 7, buttons 117 are displayed on each of the second and subsequent rows of the list 120. The buttons 117 are used to select an account item and may be, for example, radio buttons. Note that, instead of the buttons 117, check boxes may be displayed.

[0099] Furthermore, the example shown in Fig. 7 shows buttons 141 and 142. Fig. 7 shows an example in which button 141 displays "OK" and button 142 displays "Edit." The example shown in Fig. 7 shows an example in which transaction information 100, response sentence 220, button 141, list 120, and button 142 are displayed on display unit 23 from top to bottom.

[0100] After step S18 shown in FIG. 2 is performed, steps S33 and S19 shown in FIG. 3 are performed. In steps S33 and S19, the user of the information terminal 20 determines whether the selected account item is appropriate. For example, the user of the information terminal 20 determines whether the selected account item, item 223, is appropriate by examining the transaction information 100 and response sentence 220 shown in FIG. 7. If the selected account item is appropriate, the user of the information terminal 20 selects button 141. As a result, data indicating that the selected account item is appropriate is transmitted to the information processing device 40. In other words, the information terminal 20 receives data indicating that the selected account item is appropriate and transmits the received data to the information processing device 40. Here, the reason for selecting an account item such as item 225 is displayed on the display unit 23, making it easier for the user of the information terminal 20 to determine whether the selected account item is appropriate.

[0101] If data indicating that the selected account item is valid is sent to the information processing device 40, in step S20 the information processing device 40 registers the account item as an account item related to the transaction information 100. Specifically, the information processing device 40 registers the account item of the response statement 220, i.e., the account item selected by the information processing device 50 in step S13, in the database 45 as an account item related to the transaction information 100.

[0102] If the selected account item is not valid, in step S34, the user of information terminal 20 selects button 117 on list 120 and then selects button 142. By selecting button 142, data indicating that the selected account item is not valid and that it should be corrected to the account item corresponding to the selected button 117 is sent to information processing device 40. In other words, information terminal 20 accepts this data and transmits the accepted data to information processing device 40. Here, an explanation of the account item such as item 113 is displayed on display unit 23, making it easier for the user of information terminal 20 to select a valid account item.

[0103] In the example shown in FIG. 7 , the button 117 corresponding to the account item "dd1 expense" is button 117[1], and the button 117 corresponding to the account item "dd2 expense" is button 117[2]. Also, in the example shown in FIG. 7 , button 117[2] is selected. When button 142 is selected in this state, the account item assigned to the transaction information 100 is corrected from the account item "dd1 expense" in the response statement 220 to the account item "dd2 expense" corresponding to button 117[2]. Note that account items identical to the account items in the response statement 220 do not need to be included in the list 120 displayed on the display unit 23. In other words, a list obtained by removing the account items and their explanations in the response statement 220 from the list 120 shown in FIG. 5A may be displayed on the display unit 23 as the list 120 shown in FIG. 7. Also, account items identical to the account items selected by the information processing device 50 in step S13 may be included in the list 120 displayed on the display unit 23, but the account items may not be selectable. For example, button 117[1] shown in FIG. 7 may be made unselectable. In FIG. 7, button 117[1] and the corresponding account item and its description are hatched. This distinguishes the account item in response sentence 220 from other account items. Note that the list 120 displayed on display unit 23 may include items 111 included in data table 110 shown in FIG. 4B. Here, if data table 110 includes two or more types of items 111, some of the items 111 may be displayed on display unit 23. In other words, some of the items 111 may be included in list 120 displayed on display unit 23.

[0104] The user of the information terminal 20 can select buttons 141 and 142 by clicking, touching, or using a keyboard. It is not necessary to display buttons 141 and 142 on the display unit 23. For example, a form may be displayed on the display unit 23. In this case, the user of the information terminal 20 can input into the form, for example, a statement indicating that the account item selected by the information processing device 50 in step S13 is appropriate, or a statement indicating that the account item is not appropriate and an appropriate account item. This allows data indicating the above to be transmitted to the information processing device 40.

[0105] The information processing device 40 or the information processing device 50 may determine whether the account title in the response statement 220 is appropriate. For example, the information processing device 40 or the information processing device 50 can use a neural network model to determine whether the account title in the response statement 220 is appropriate. For example, by inputting the transaction information 100, the response statement 220, and the list 120 shown in FIG. 7 into the neural network model, the information processing device 40 or the information processing device 50 can determine whether the account title in the response statement 220 is appropriate.

[0106] Furthermore, if the account item in the response sentence 220 is invalid, the information processing device 40 or the information processing device 50 may present a proposed amendment to the account item to the user of the information terminal 20. For example, the information processing device 40 may present the proposed amendment using a neural network model. Alternatively, the information processing device 50 may present the proposed amendment using a language model. For example, by inputting the transaction information 100, the response sentence 220, and the list 120 shown in FIG. 7 into the neural network model of the information processing device 40, the information processing device 40 can present one or more proposed amendments.

[0107] Alternatively, the information processing device 50 can present one or more suggested corrections by inputting a prompt having an instruction sentence, the transaction information 100, the response sentence 220, and the list 120 shown in FIG. 7 into the language model of the information processing device 50. The instruction sentence includes an instruction to select an account item from the list 120 that is different from the account item in the response sentence 220. The instruction sentence can be, for example, "The selected account item needs to be corrected. Please reselect an account item from the list of accounts and explain why."

[0108] If the account item selected by the information processing device 50 in step S13 is invalid and data indicating that the account item should be corrected is sent to the information processing device 40, then in step S21 the information processing device 40 registers the corrected account item as an account item related to the transaction information 100. Specifically, the information processing device 40 registers the account item selected by the user of the information terminal 20 in step S34 in the database 45 as an account item related to the transaction information 100. In step S21, the information processing device 40 accepts the user's correction of the account item selected by the information processing device 50 in step S13. Subsequently, the information processing device 40 registers the corrected account item in the database 45 as an account item related to the transaction information 100. In step S21, specifically, the reception unit 41 of the information processing device 40 accepts the user's correction of the account item.

[0109] In this specification, the corrected account item may be referred to as the third account item, and the explanation for the third account item may be referred to as the third explanation.

[0110] [Step S22] If the processing of step S21 has been performed, in step S22 the information processing device 40 creates a prompt 230. Specifically, the processing unit 43 of the information processing device 40 creates the prompt 230 for correcting the description of the account item. More specifically, the processing unit 43 of the information processing device 40 creates the prompt 230 for correcting the description of the account item so that the information processing device 50 can correctly select the account item in step S13.

[0111] 8A is a schematic diagram showing an example of a prompt 230. The prompt 230 includes an instruction statement 231, an item 101[1], an item 101[2], an item 223, an item 224, an item 227, and an item 228.

[0112] The instruction statement 231 includes a command to present a suggested revision to the description of the account item. Specifically, the instruction statement 231 includes a command to present a suggested revision to the description of the account item so that the information processing device 50 can select the account item selected by the user of the information terminal 20 in step S34 shown in FIG. 3 in step S13. The instruction statement 231 can be, for example, "The account item presented based on the product name and transaction purpose has been revised by the user. Please present a suggested revision to the description of the account item so that the revised account item can be selected."

[0113] Item 101[1] and item 101[2] may be the items 101[1] and 101[2] included in the transaction information 100. In other words, the prompt 230 includes at least a portion of the transaction information 100, similar to the prompt 210. Note that items other than the product name and transaction purpose may also be included in the prompt 230 as items 101.

[0114] The item 223 can be the account item selected by the information processing device 50 in step S13. In the example shown in Fig. 8A, the item 223 is "dd1 expense."

[0115] Item 224 can be the account item after correction, i.e., the account item selected by the user of information terminal 20 in step S34 shown in Fig. 3. In the example shown in Fig. 8A, item 224 is "dd2 expense." As described above, Fig. 8A shows an example in which the account item related to transaction information 100 is corrected from "dd1 expense" to "dd2 expense."

[0116] Item 227 can be the description of the account item selected by the information processing device 50 in step S13. That is, item 227 can be the description of item 223. When item 223 is the account item and item 227 is the description of that account item, item 223 can be called the second account item and item 227 can be called the second description.

[0117] Item 228 can be the description of the account item after the correction. That is, item 228 can be the description of item 224. When item 224 is the account item and item 228 is the description of that account item, item 224 can be called the third account item and item 228 can be called the third description.

[0118] In the example shown in FIG. 8A, items 223, 224, 227, and 228 may be items 112[1], 112[2], 113[1], and 113[2], respectively, included in prompt 210 shown in FIG. 5A.

[0119] Instruction 231 is an instruction for creating one or both of the proposed revisions for item 227 and item 228. The proposed revisions can be created based on transaction information 100, item 223, and item 224.

[0120] [Step S23] In step S23, the information processing device 40 transmits the prompt 230 to the language model of the information processing device 50 via the network 30 (step S23a shown in FIG. 3). As a result, the information processing device 50 creates a response sentence 240 using the language model based on the prompt 230. The response sentence 240 includes, for example, an account item whose explanation has been revised based on the information included in the prompt 230, and the explanation. Specifically, the response sentence 240 includes a revision proposal for the explanation of the account item selected by the information processing device 50 in step S13 and a revision proposal for the explanation of the account item selected by the user of the information terminal 20 in step S34 shown in FIG. 3. The information processing device 50 transmits the response sentence 240 to the information processing device 40 via the network 30 (step S23b shown in FIG. 3). The reception unit 41 of the information processing device 40 receives the response sentence 240 transmitted from the information processing device 50. As a result, the information processing device 40 acquires the response sentence 240 (step S23c shown in FIG. 3).

[0121] FIG. 8B is a schematic diagram showing an example of a response sentence 240. The response sentence 240 includes items 223, 224, 227a, and 228a. Items 223 and 224 may be the same as those in FIG. 8A. Items 227a and 228a may be suggested revisions to the explanations of items 223 and 224, respectively. That is, items 227a and 228a may be suggested revisions to items 227 and 228, respectively. Specifically, items 227a and 228a may be suggested revisions to the second explanation and the third explanation, respectively. Note that item 227a or 228a may not be included in the response sentence 240. For example, if item 227a is the same as item 227 shown in FIG. 8A, i.e., if item 227 is not revised, item 227a may not be included in the response sentence 240. Similarly, if item 228a is identical to item 228 shown in FIG. 8A, that is, if no modification of item 228 is made, item 228a may not be included in response sentence 240.

[0122] 8A and 8B show an example in which the description of item 223 is corrected from "Purchase price of ee, expenses for ff, etc." to "Purchase price of ee." Also, an example is shown in which the description of item 224 is corrected from "Participation fee for gg, etc." to "Participation fee for gg, cost of purchasing teaching materials, etc."

[0123] In an information processing method according to one aspect of the present invention, a series of processes are performed using transaction information 100. Therefore, it is preferable that the language model used in the process of step S13 is the same as the language model used in the process of step S23. In other words, it is preferable that prompt 230 be sent to the language model that accepts prompt 210. With this configuration, the accuracy of response sentence 220 and the accuracy of response sentence 240 can be maintained constant.

[0124] Note that the language model used in the processing of step S13 may be different from the language model used in the processing of step S23. That is, a configuration may be adopted in which prompt 210 is sent to a first language model and prompt 230 is sent to a second language model different from the first language model. With such a configuration, processing using the first language model and processing using the second language model can be performed using different information processing devices, specifically different information processing devices 50. This reduces the computational load per information processing device.

[0125] [Step S24] In step S24, the information processing device 40 modifies the explanation of the account item based on the response sentence 240. For example, the processing unit 43 of the information processing device 40 modifies item 113 based on one or both of items 227a and 228a. For example, the explanation of the account item in the data table 110 shown in FIG. 4 is modified based on the proposed modification of the explanation included in the response sentence 240. For example, the explanation of the account item "dd1 expense" in the data table 110 is modified to "Purchase price of ee" as in item 227a. Furthermore, the explanation of the account item "dd2 expense" in the data table 110 is modified to "Participation fee for gg, cost of purchasing teaching materials, etc." as in item 228a. In this way, the data table 110 registered in the database 45 is updated.

[0126] For example, a user of the information processing system 10 shown in FIG. 1 or an administrator of the information processing system 10 may review the response statement 240 and then modify the description of the account item in the data table 110. For example, a suggested modification to the description included in the response statement 240 may be reflected in the description of the account item in the data table 110. The user or administrator may not modify the description of the account item in the data table 110, or may use a description different from the suggested modification included in the response statement 240. If the information processing system 10 is an accounting support system, the administrator of the information processing system 10 may be, for example, an accountant. In this case, the user of the information processing system 10 is not limited to an accountant, but may also be, for example, a person who performed a transaction related to the transaction information 100. The administrator of the information processing system 10 may be, for example, an administrator of the information processing device 40.

[0127] Furthermore, before the information processing device 40 performs the processing in step S24, it may create the prompt 230 in step S22 and obtain the response sentence 240 in step S23 again. The user or administrator can decide whether to create the prompt 230 and obtain the response sentence 240 again. For example, if the user or administrator checks the proposed revision of the explanatory sentence included in the response sentence 240 and determines that the content is inappropriate, the user or administrator can create the prompt 230 and obtain the response sentence 240 again. When creating the prompt 230 and obtaining the response sentence 240 again, the user or administrator can specify at least a portion of the information to be included in the prompt 230.

[0128] As described above, the information processing method of one embodiment of the present invention can automate the assignment of account items to transaction information by using a language model. Furthermore, if the language model is unable to select an account item due to, for example, insufficient information, the missing information can be presented to the user of the information terminal. Furthermore, if the language model selects an account item, the reason for the selection can be presented to the user of the information terminal. Furthermore, if the selected account item is inappropriate, the user of the information terminal can correct the account item, and the probability that the information processing device 50 will select an appropriate account item in subsequent processing can be increased. As described above, the information processing method of one embodiment of the present invention can be an information processing method that is highly convenient, useful, and reliable. Furthermore, the information processing device and information processing system of one embodiment of the present invention can be an information processing device and information processing system that are highly convenient, useful, and reliable.

[0129] <Example 2 of Information Processing Method> An information processing method different from the above-described information processing method 1 will be described below. Fig. 9 is a flowchart showing an example of the information processing method. In Fig. 9, steps that perform the same processes as the steps shown in Fig. 3 are assigned the same reference numerals.

[0130] In the information processing method shown in Figure 9, the information processing device 40 performs the process in step S21, i.e., the process of registering the corrected account items as account items related to the transaction information 100, and then accepts the reason for correcting the account items. Then, after the information processing device 40 accepts n reasons for correction (n is an integer greater than or equal to 1), it corrects the explanations of the account items, specifically the explanations of the account items held in the data table 110, based on the n reasons for correction. Specifically, after the information processing device 40 accepts reasons for correcting the n account items in the transaction information 100, it corrects the explanations of the account items held in the data table 110 based on these reasons for correction. Note that n can be specified, for example, by an administrator of the information processing system 10.

[0131] 9, the process in step S10 is performed before the process in step S31. In step S10, the information processing device 40, specifically the processing unit 43, prepares a variable i and inputs, for example, 1 as an initial value to the variable i.

[0132] After the processing in step S10 is performed, the processing in step S31 and steps S11 to S15 is performed. That is, after the user of the information terminal 20 inputs the transaction information 100 to the information processing device 40, a response sentence 220 is created using a language model. The information processing device 40 then determines whether an account item has been selected in the response sentence 220. If it is determined that an account item has not been selected, the processing in steps S16, S32, and S17 is performed, and then the processing in steps S12 to S15 is performed again.

[0133] If it is determined that an account item has been selected, in step S18, the information processing device 40 prompts the user of the information terminal 20 to determine whether the selected account item is appropriate. Subsequently, in steps S33 and S19, the user of the information terminal 20 determines whether the account item selected in step S13 is appropriate. If it is appropriate, the process in step S20 is performed, and then the process in step S31 is performed again. That is, the information processing device 40 enters a state of waiting for the user of the information terminal 20 to input transaction information 100. If it is not appropriate, in step S34, the user of the information terminal 20 corrects the account item. Subsequently, in step S21, the information processing device 40 accepts the correction of the account item. Thereafter, the corrected account item is registered in the database 45 as an account item related to the transaction information 100.

[0134] In the information processing method shown in Fig. 9, the information processing device 40, specifically the reception unit 41, receives the reason why the user of the information terminal 20 revised the account item in step S41. The reason for revision received by the reception unit 41 is referred to as revision reason [i]. For example, on the display unit 23 shown in Fig. 7, a form is displayed between the list 120 and the button 142. The user of the information terminal 20 then inputs the reason for revision into the form. Note that the user of the information terminal 20 does not need to input the reason for revision. In this case, the information processing device 40 does not need to perform the process of step S41.

[0135] After the information processing device 40 performs the process in step S41, in step S42, the value of the variable i is incremented by 1. If the information processing device 40 does not perform the process in step S41, the process in step S42 does not need to be performed either.

[0136] Next, the processes of steps S22 to S24 are performed. Specifically, first, a response sentence 240 is created using a language model. Next, the information processing device 40 modifies the description of the account item in the data table 110 as shown in FIG. 4B based on the response sentence 240. After that, in step S43, the information processing device 40, specifically the processing unit 43, determines whether i is greater than n. If i is equal to or less than n, the process of step S31 is performed again. That is, the information processing device 40 enters a state of waiting for the user of the information terminal 20 to input transaction information 100.

[0137] If i is greater than n, in step S44, the information processing device 40 creates a prompt 250. Specifically, the processing unit 43 of the information processing device 40 creates a prompt 250 for the information processing device 50 to correct the description of the account item in step S13. More specifically, the processing unit 43 of the information processing device 40 creates a prompt 250 for correcting the description of the account item so that the information processing device 50 can correctly select the account item in step S13.

[0138] 10A is a schematic diagram showing an example of a prompt 250. The prompt 250 includes an instruction statement 251, items 253[1] through 253[n], and a list 120.

[0139] The instruction statement 251 includes a command to present a suggested revision to the description of the account item. Specifically, the instruction statement 251 includes a command to present a suggested revision to the description of the account item so that the information processing device 50 can select the account item selected by the user of the information terminal 20 in step S34 shown in FIG. 3 in step S13. The instruction statement 251 can be, for example, "In the following transaction, the account item presented based on the product name and transaction purpose has been revised by the user. Please present a suggested revision to the description of the account item so that the revised account item can be selected."

[0140] 10A, items 253[1] through 253[n] each include item 101[1], item 101[2], item 223, item 224, and item 229. Therefore, item 253 can be obtained by omitting instruction statement 231, items 227, and item 228 from prompt 230 shown in FIG. 8A and adding item 229.

[0141] Item 229 can be the reason why the user of information terminal 20 revised the account item, which was accepted by acceptance unit 41 in step S41. Fig. 10A shows an example in which the reason for revising the account item related to transaction information 100 from "dd1 expense" to "dd2 expense" is "for kk."

[0142] As described above, in the example shown in FIG. 10A, the prompt 250 includes n pieces of transaction information 100 in which the account item has been corrected, the account item before and after the correction, and the reason for the correction.

[0143] Like the list 120 shown in FIG. 5A , the list 120 shown in FIG. 10A includes at least some of the information included in the data table 110 shown in FIG. 4B . In the example shown in FIG. 10A , the items included in the list 120 can be account items and descriptions of those account items. Similarly to the list 120 shown in FIG. 5A , the list 120 shown in FIG. 10A can also include at least some of the records 115 included in the data table 110. For example, an account item included in any of items 253[1] through 253[n] can be included in the list 120. Specifically, an account item included in at least one of items 223 of items 253[1] through 253[n] and items 224 of items 253[1] through 253[n] can be included in the list 120.

[0144] Here, for example, multiple variables may be prepared in step S10. For example, it is preferable to prepare multiple variables corresponding to the type of item 111. Specifically, if the list 120 is created using a record 115 whose item 111 is "cc1 section," 1 is added to the variable i1 in step S42. Also, if the list 120 is created using a record 115 whose item 111 is "cc2 section," 1 is added to the variable i2 in step S42. Then, in step S43, the information processing device 40 determines whether each of the variables i1 and i2 exceeds n. If n is exceeded, processing is performed in step S44. In step S44, the processing history associated with the variable determined to exceed n in step S43 is used as the item 253. This prevents the text volume of the prompt 250 shown in FIG. 10A from increasing, making it easier to obtain a response sentence from a language model in subsequent processing. Furthermore, the time required for generating a response sentence using a language model can be shortened. Different variables may be prepared for each account item. For example, if the data table 110 shown in FIG. 4B has m rows (m is an integer equal to or greater than 1) of records 115, m types of variables may be prepared in step S10.

[0145] In step S45, the information processing device 40 transmits the prompt 250 to the language model of the information processing device 50 via the network 30. As a result, the information processing device 50 creates a response sentence 260 using the language model based on the prompt 250. The response sentence 260 includes, for example, an account item whose explanation has been modified based on the information included in the prompt 250, and the explanation itself. The information processing device 50 transmits the response sentence 260 to the information processing device 40 via the network 30. The reception unit 41 of the information processing device 40 receives the response sentence 260 transmitted from the information processing device 50. As a result of the above, the information processing device 40 acquires the response sentence 260.

[0146] FIG. 10B is a schematic diagram showing an example of a response sentence 260. The response sentence 260 includes items 263 and 264. Item 263 can be an account item for which a proposed revision to the explanation has been presented. Item 264 can be a proposed revision to the explanation of the account item. In the example shown in FIG. 10B, items 263[1] and 263[2] are shown as items 263, and items 264[1] and 264[2] are shown as items 264. In addition, in the example shown in FIG. 10B, item 263[1] is "pp1 expense," item 263[2] is "pp2 expense," item 264[1] is "expense for qq1," and item 264[2] is "expense for qq2."

[0147] 9, the language model used in step S13, the language model used in step S23, and the language model used in step S45 are preferably the same. That is, prompt 210, prompt 230, and prompt 250 are preferably sent to the same language model. With this configuration, the accuracy of response sentence 220, response sentence 240, and response sentence 260 can be maintained constant.

[0148] Note that the language model used in the processing of step S13, the language model used in the processing of step S23, and the language model used in the processing of step S45 do not all need to be the same language model. For example, the language model used in the processing of step S13 may be different from one or both of the language models used in the processing of step S23 and the processing of step S45. That is, a configuration may be adopted in which the prompt 210 is sent to a first language model, and the prompt 230 or the prompt 250 is sent to a second language model different from the first language model. Alternatively, a configuration may be adopted in which the prompt 210 is sent to a first language model, the prompt 230 is sent to a second language model different from the first language model, and the prompt 250 is sent to a third language model different from the first and second language models. By using such a configuration, the information processing device used in the information processing method shown in FIG. 9 can be different for each language model. This reduces the computational load per information processing device.

[0149] In step S46, the information processing device 40 corrects the explanation of the account item based on the response statement 260. For example, the processing unit 43 of the information processing device 40 corrects item 113 based on item 264. For example, the explanation of the account item in the data table 110 shown in FIG. 4B is corrected based on the proposed correction of the explanation included in the response statement 260. For the correction of item 113, see the explanation of step S24.

[0150] 9, the explanation of the account item can be corrected by taking into consideration not only the account item selected by the information processing device 50 and the account item corrected by the user of the information terminal 20, but also the reason why the user corrected the account item. Therefore, for example, the information processing device 50 can assign an appropriate account item to the transaction information with high accuracy.

[0151] The above is an example of the information processing method according to one embodiment of the present invention.

[0152] <Configuration Example 2 of Information Processing System> Fig. 11 is a block diagram showing a configuration example of the information processing system 10. In Fig. 11, the configuration example of the information processing device 40 will be mainly described.

[0153] The information processing device 40 includes a receiving unit 41, a storage unit 44, a processing unit 43, an output unit 47, and a transmission path 49. The information processing device 40 may include a database 45.

[0154] In the drawings attached to 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 43 may function as the reception unit 41. Also, one function may be involved in multiple components. For example, the processing performed by the processing unit 43 may be executed by different servers depending on the processing.

[0155] [Accepting Unit 41] The accepting unit 41 has a function of accepting data from outside the information processing device 40. The data accepted by the accepting unit 41 is preferably text data.

[0156] For example, the reception unit 41 has a function of receiving data from the information terminal 20 via the network 30. Examples of the data include transaction information 100 and additional information 131. Another example of the data is data indicating whether the account item selected by the language model is appropriate. Another example of the data is data indicating a user's modification of the account item related to the transaction information 100. Another example of the data is data indicating a user's modification of the account item related to the transaction information 100. Another example of the data is data indicating a response sentence 220, a response sentence 240, a response sentence 260, etc. The data received by the reception unit 41 is supplied to one or both of the memory unit 44 and the processing unit 43 via the transmission path 49.

[0157] [Storage Unit 44] The storage unit 44 has a function of storing a program executed by the processing unit 43. The storage unit 44 may also have a function of storing data created by the processing unit 43 (e.g., calculation results, analysis results, inference results), data input to the reception unit 41, etc.

[0158] The storage unit 44 preferably has a function of storing an instruction sentence in the prompt. For details of the instruction sentence in the prompt, please refer to the description of the information processing method according to one aspect of the present invention.

[0159] The storage unit 44 preferably has a database 45. The database 45 has a data table 110 shown in FIG. 4B. As described above, the data table 110 can have, for example, a plurality of account items and descriptions of the account items. The information processing device 40 has a function to retrieve data from the database 45 and a function to record data in the database 45. Furthermore, the database 45 registers transaction information and account items related to the transaction information.

[0160] The information processing device 40 may have a database 45 outside the storage unit 44. The information processing device 40 may have a function to retrieve data from a database 45 that exists outside the storage unit 44, outside the information processing device 40, or outside the information processing system 10. The information processing device 40 may also have a function to record data in a database 45 that exists outside the storage unit 44, outside the information processing device 40, or outside the information processing system 10. The information processing device 40 may also have a function to retrieve data from both its own database and an external database. The information processing device 40 may also have a function to record data in both its own database and an external database.

[0161] 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.

[0162] Furthermore, the information processing device 40 may have a function of extracting at least one piece of data from data managed by the information terminal 20 or an IMAP server or the like.

[0163] The storage unit 44 includes at least one of a volatile memory and a nonvolatile memory. Examples of the volatile memory include a dynamic random access memory (DRAM) and a static random access memory (SRAM). Examples of the nonvolatile 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 44 may also include at least one of NOSRAM (registered trademark) and DOSRAM (registered trademark). The storage unit 44 may also include a recording media drive. Examples of recording media drives include a hard disk drive (HDD) and a solid state drive (SSD).

[0164] 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.

[0165] 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.

[0166] 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.

[0167] 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.

[0168] 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.

[0169] [Processing Unit 43] The processing unit 43 has a function of performing processes such as calculation, analysis, and inference using data supplied from at least one of the reception unit 41, the storage unit 44, and the database 45. The processing unit 43 can supply the created data (e.g., calculation results, analysis results, and inference results) to at least one of the storage unit 44, the database 45, and the output unit 47.

[0170] The processing unit 43 has a function of acquiring data from one or both of the storage unit 44 and the database 45. The processing unit 43 also has a function of recording or registering data in one or both of the storage unit 44 and the database 45.

[0171] The processing unit 43 has a function of creating a prompt 210 having transaction information 100, a list 120, and an instruction statement 211. The instruction statement 211 is an instruction statement for selecting an account item related to the transaction information 100 from the account items included in the list 120. The account item can be selected based on the transaction information 100 and the description of the account item. As described above, the instruction statement 211 is an instruction statement for selecting an item 112 related to the transaction information 100 from the items 112 included in the list 120. As shown in FIG. 4A , the transaction information 100 can have an item 101[1] that is a product name and an item 101[2] that is a transaction purpose.

[0172] The processing unit 43 has a function of determining whether an account item has been selected in a response sentence 220 acquired by sending a prompt 210 to the language model of the information processing device 50 via the network 30. If an account item has not been selected, the processing unit 43 has a function of presenting to the user of the information terminal 20 the reason why the account item has not been selected, which is included in the response sentence 220 shown in Fig. 5B1. The processing unit 43 also has a function of re-creating the prompt 210 based on the additional information 131 shown in Fig. 6B.

[0173] When an account item is selected, the processing unit 43 has a function of prompting the user of the information terminal 20 to determine whether the account item is appropriate. Here, as shown in Fig. 5B2, when an account item is selected, the response sentence 220 can include the reason for selecting the account item.

[0174] When the user of the information terminal 20 judges that the account item selected by the language model is appropriate, the processing unit 43 has a function of registering the account item in the database 45 as an account item related to the transaction information 100. In other words, when the processing unit 43 receives data from the information terminal 20 via the reception unit 41 indicating that the account item selected by the language model is appropriate, the processing unit 43 has a function of registering the account item in the database 45 as an account item related to the transaction information 100.

[0175] When the user of the information terminal 20 determines that the account item selected by the language model is not appropriate and corrects the account item, the processing unit 43 has a function of registering the corrected account item in the database 45 as an account item related to the transaction information 100. In other words, when the processing unit 43 receives data from the information terminal 20 via the reception unit 41 indicating that the account item selected by the language model is not appropriate and that the account item will be corrected, the processing unit 43 has a function of registering the corrected account item in the database 45 as an account item related to the transaction information 100.

[0176] The processing unit 43 has a function of generating a prompt 230, as shown in FIG. 8A , that includes the transaction information 100, an account title selected by the language model, a description of the account title selected by the language model, an account title corrected by the user of the information terminal 20, the description of the corrected account title, and an instruction 231. These explanations are selected from the data table 110 shown in FIG. 4B . Specifically, they are selected from the items 113 of the data table 110. The instruction 231 is an instruction for generating a suggested revision to the description of the account title selected by the language model and a suggested revision to the description of the account title corrected by the user of the information terminal 20. The above-mentioned revision suggestions can be generated based on the transaction information 100, the account title selected by the language model, and the account title corrected by the user of the information terminal 20. Furthermore, the instruction statement 231 may be an instruction statement for creating one or both of the proposed revisions for item 227 and item 228 based on the transaction information 100, item 223, and item 224, as described above.

[0177] The processing unit 43 has a function of correcting the description of the account item, i.e., the item 113, of the data table 110 shown in Fig. 4B based on the response sentence 240 shown in Fig. 8B. The processing unit 43 can obtain the response sentence 240 by sending a prompt 230 to the language model of the information processing device 50 via the network 30.

[0178] The processing unit 43 may include, for example, an arithmetic circuit, a central processing unit (CPU), and a graphics processing unit (GPU).

[0179] The processing unit 43 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 43 may also have a quantum processor. The processing unit 43 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 44.

[0180] The processing unit 43 may have 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.

[0181] 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 43. The operating system, application programs, program modules, program data, lookup tables, and the like stored in the storage unit 44 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 43.

[0182] 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.

[0183] The processing unit 43 can have 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.

[0184] The processing unit 43 preferably includes an OS transistor. Because an OS transistor has an extremely small 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, thereby enabling low power consumption of the information processing system 10.

[0185] [Output Unit 47] The output unit 47 has a function of outputting data based on the processing result of the processing unit 43. The output unit 47 can supply at least one of the calculation result, analysis result, and inference result of the processing unit 43 to the outside of the information processing device 40.

[0186] The output unit 47 has a function of outputting data to the outside of the information processing device 40. For example, the output unit 47 has a function of outputting data to the information processing device 50 via the network 30. The output unit 47 has a function of outputting data created by the processing unit 43 to the information processing device 50. The data can be created based on data received from the information terminal 20, for example. Examples of data output to the information processing device 50 include prompt 210, prompt 230, and prompt 250.

[0187] Furthermore, for example, the output unit 47 has a function of outputting data to the information terminal 20 via the network 30. The output unit 47 can output data created by the processing unit 43 to the information terminal 20. The data can be created based on data received from the information processing device 50, for example. Alternatively, the output unit 47 can output the data itself received from the information processing device 50 to the information terminal 20. An example of the data to be output to the information terminal 20 is a response sentence 220.

[0188] 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.

[0189] [Transmission Path 49] The transmission path 49 has a function of transmitting data. Data can be transmitted and received between the reception unit 41, the storage unit 44, the processing unit 43, and the output unit 47 via the transmission path 49.

[0190] <Configuration Example 3 of Information Processing System> Fig. 12 is a schematic diagram showing an example of the information processing system 10 that is different from Fig. 1. In the information processing system 10 shown in Fig. 12, an information processing device 40 and a plurality of information terminals 20 are connected via a network 31. Furthermore, the information processing device 40 is connected to an information processing device 50 via a network 33. In Fig. 12, the above-mentioned information terminals 20a to 20d are shown as examples of the information terminals 20.

[0191] A user of the information processing system 10 can access the information processing device 40 from the information terminal 20. The user can then receive a service using the information processing method according to one aspect of the present invention.

[0192] In the information processing system 10 shown in Figure 12, the data flow between the information terminal 20 and the information processing device 40, and the data flow between the information processing device 40 and the information processing device 50 can be understood by referring to the explanation in Figure 1 by replacing network 31 and network 33 with network 30, respectively.

[0193] An organization such as a company that provides a service using the information processing method of one embodiment of the present invention (hereinafter also referred to as a service provider) can provide the service using the information processing device 40. The service provider does not need to prepare the information processing device 50 on its own, but can use the information processing device 50 via the network 33.

[0194] Network 31 is a computer network smaller than network 33. Typically, network 31 is a local network, and network 33 is a global network. It is preferable to use an intranet or an extranet as network 31. It is preferable to use the Internet, which is the foundation of the World Wide Web (WWW), as network 33. In this embodiment, an example will be mainly described in which an intranet is used as network 31 and the Internet is used as network 33.

[0195] Alternatively, the network 31 and the network 33 may be computer networks such as PAN, LAN, CAN, MAN, WAN, and GAN.

[0196] The information processing system 10 shown in FIG. 12 is suitable for a case where a user who uses an information terminal 20 and a service provider belong to the same organization such as a company.

[0197] For example, it is preferable that data is transmitted and received between the information terminal 20 and the information processing device 40 using a network 31 established within an organization such as a company. It is also preferable that data is transmitted and received between the information processing device 40 and the information processing device 50 using a network 33 (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 50 via the Internet.

[0198] A plurality of configuration examples shown in this embodiment mode can be combined as appropriate.

[0199] 10: Information processing system, 20: Information terminal, 20a: Information terminal, 20b: Information terminal, 20c: Information terminal, 20d: Information terminal, 21: Housing, 23: Display unit, 30: Network, 31: Network, 33: Network, 40: Information processing device, 41: Reception unit, 43: Processing unit, 44: Storage unit, 45: Database, 47: Output unit, 49: Transmission path, 50: Information processing device, 100: Transaction information, 101[1]: Item, 101[2]: Item, 101: Item, 110: Data table, 111: Item, 112[1]: Item, 112[2]: Item, 112: Item, 113[1]: Item, 113[2]: Item, 113: Item, 115[1]: Record, 115[2]: Record, 115[3] : Record, 115: Record, 117[1]: Button, 117[2]: Button, 117: Button, 120: List, 130: Form, 131: Additional Information, 141: Button, 142: Button, 210: Prompt, 211: Instruction, 220: Response, 223: Item, 224: Item, 225: Item, 227: Item, 227a: Item, 22 8: item, 228a: item, 229: item, 230: prompt, 231: instruction sentence, 240: response sentence, 250: prompt, 251: instruction sentence, 253[1]: item, 253[n]: item, 253: item, 260: response sentence, 263[1]: item, 263[2]: item, 263: item, 264[1]: item, 264[2]: item, 264: item

Claims

In a first step, transaction information is accepted; In a second step, a first prompt is created having the transaction information, a list, and a first instruction; The list includes a plurality of first account items and a first description that is a description of the first account items, the first instruction is an instruction for selecting a second account item, which is an account item related to the transaction information, from a plurality of the first account items based on the transaction information and the first explanation; In a third step, a first response sentence is generated based on the first prompt using a language model; In a fourth step, it is determined whether or not the second account item is selected in the first response sentence; If the second account item is not selected, in a fifth step, the reason why the second account item is not selected, which is included in the first response sentence, is presented to the user to prompt for input of additional information; In a sixth step, the additional information is accepted; After performing the sixth step, the second to fourth steps are performed again; If the second account title is selected, in a seventh step, the user is prompted to determine whether the second account title is valid; If the second account title is valid, in an eighth step, the second account title is registered; If the second account title is not valid, in a ninth step, after accepting a correction of the second account title by the user, registering it as a third account title; In a tenth step, a second prompt is created having the transaction information, the second account item, the third account item, a second explanation that is an explanation of the second account item, a third explanation that is an explanation of the third account item, and a second instruction; the second explanatory text and the third explanatory text are selected from the first explanatory text; the second instruction is an instruction for creating one or both of a revision proposal for the second explanation and a revision proposal for the third explanation based on the transaction information, the second account item, and the third account item; In an eleventh step, an information processing method is provided in which a second response sentence including one or both of a suggested revision to the second explanatory sentence and a suggested revision to the third explanatory sentence is created using the language model based on the second prompt. 　 In claim 1, An information processing method using a database, a data table including the list is registered in the database; In the eighth step, the second account item is registered in the database as an account item related to the transaction information; In the ninth step, the third account item is registered in the database as an account item related to the transaction information; In a twelfth step, the information processing method includes correcting the first explanatory sentence registered in the database based on one or both of the suggested revisions to the second explanatory sentence and the suggested revisions to the third explanatory sentence. 　 In claim 1, An information processing method, wherein, in the third step, if the second account item is selected, the first response sentence includes a reason for selecting the second account item. 　 In any one of claims 1 to 3, An information processing method, wherein the transaction information includes a product name and a transaction purpose. 　 The apparatus includes a database, a reception unit, and a processing unit, the database includes a data table; The data table includes a plurality of first account items and a first explanation that is an explanation of the first account items, the reception unit has a function of receiving transaction information, additional information, and a correction of a second account item related to the transaction information, and supplying the same to the processing unit; the processing unit has a function of generating a first prompt having the transaction information, a list that is at least a portion of the data table, and a first instruction sentence; the first instruction is an instruction for selecting the second account item from a plurality of the first account items based on the transaction information and the first explanation; the processing unit has a function of determining whether or not the second account item is selected in a first response sentence acquired by transmitting the first prompt to a language model via a network; the processing unit has a function of, if the second account item is not selected, presenting to the user a reason why the second account item is not selected, the reason being included in the first response sentence, and creating the first prompt based on the additional information; The processing unit has a function of, when the second account item is selected, prompting the user to determine whether the second account item is appropriate; the processing unit has a function of registering the second account item in the database as an account item related to the transaction information when data indicating that the second account item is valid is received; the processing unit has a function of, when the second account title is not valid and data is received indicating that the second account title is to be corrected, changing the correction of the second account title made by the user to a third account title, and registering the third account title in the database as an account title related to the transaction information; the processing unit has a function of creating a second prompt having the transaction information, the second account title, the third account title, a second explanation that is an explanation of the second account title, a third explanation that is an explanation of the third account title, and a second instruction sentence; the second explanatory text and the third explanatory text are selected from the first explanatory text; the second instruction is an instruction for creating one or both of a revision proposal for the second explanation and a revision proposal for the third explanation based on the transaction information, the second account item, and the third account item; The information processing device has a function of modifying the first explanatory sentence based on a second response sentence obtained by sending the second prompt to the language model via the network. 　 In claim 5, An information processing device in which the first response sentence includes, when the second account item is selected, a reason for selecting the second account item. 　 In claim 5 or 6, The transaction information includes a product name and a transaction purpose.

Images