Support device, program, recording medium, and method for creating documents

Abstract

We support the creation of documents through co-creation between humans and artificial intelligence. [Solution] The system includes a display control means that displays a screen corresponding to any one of the steps of interest included in a plurality of steps, and the screen includes an activation component for activating a large-scale language model utilization means for causing a large-scale language model to generate a string containing one or more answers to a question posed to the user in the step of interest, based on user operation; a display area for displaying the string containing one or more answers generated by the large-scale language model; a text box for inputting individual answers from the user or the large-scale language model to a question posed to the user in the step of interest; a list for holding the individual answers entered in the text boxes together; and a switching component for switching the activation / deactivation of individual answers held in the list based on user operation.

Description

[Technical Field]

[0001] The present invention relates to a support apparatus, program, recording medium, and method for material preparation. [Background Art]

[0002] In the business field, materials for new planning, exemplified by new business plans and the like, are prepared. However, it is not easy to prepare such materials so that they are evaluated as business documents. Therefore, there is a demand for technology that supports the preparation of materials evaluated as business documents.

[0003] Regarding technologies for supporting preparation of materials evaluated as business documents among conventional services and conventional technologies, Patent Document 1 discloses a preparation support apparatus that supports preparation of a proposal including information on bid specifications corresponding to a new inquiry by accepting the new inquiry including information on required specifications for bidding in a project such as plant design and construction, the preparation support apparatus comprising: a calculation unit that calculates a similarity between content included in the new inquiry and content included in a past inquiry or a past proposal, and extracts content useful for preparing a proposal from historical data associated with the past inquiry using the similarity; and a providing unit that generates and provides auxiliary data to be used for preparing a proposal using a result of the calculation unit.

[0004] The technology described in Patent Document 1 can efficiently grasp the specification document presented at the time of a request for bidding, and can support bidders in preparing an accurate proposal based on the specification document. [Prior Art Documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2021-120803 [Summary of the Invention] [Problem to be Solved by the Invention]

[0006] Incidentally, in bidding for plant design and construction projects, it is expected that specifications similar to those of past projects will be submitted. Therefore, proposals in such bidding processes are expected to be more likely to succeed and will be more highly regarded as business documents the more similar their content is to reference documents for similar past specifications.

[0007] In contrast, new projects are required to be different from past projects. Therefore, if the materials for a new project are similar to those for past projects, it is unlikely that they will be evaluated as business documents.

[0008] The technology described in Patent Document 1 extracts content useful for creating proposals from historical data associated with past reference documents, using similarity. Therefore, the technology described in Patent Document 1 is only capable of assisting in the creation of proposals for bidding in projects such as plant design and construction, and it is presumed that there is room for further improvement in terms of assisting in the creation of materials for new projects.

[0009] The objective of this invention is to support the creation of documents through co-creation between humans and artificial intelligence (AI). [Means for solving the problem]

[0010] As a result of diligent research to solve the above problems, the inventors of the present invention have found that the above objectives can be achieved by providing an operation in the input screen for inputting the results of a new project, which includes in the input of candidate study results generated by a large language model (LLM) based on the basic data of the new project and the data of the study results that have been previously entered. Thus, the inventors of the present invention have been completed.

[0011] According to an embodiment of the present invention, an assistance device is provided for assisting a user in creating a document using multiple steps, comprising: a display control means for displaying a screen corresponding to any one of the steps of interest included in the multiple steps, wherein the screen comprises: an activation component for activating a large-scale language model utilization means for causing a large-scale language model to generate a string containing one or more answers to a question posed to the user in the step of interest, based on user operation; a display area for displaying the string generated by the large-scale language model, which contains the one or more answers; a text box for inputting individual answers by the user or the large-scale language model to a question posed to the user in the step of interest; a list for collectively holding the individual answers entered in the text box; and a switching component for switching the activation / deactivation of the individual answers held in the list based on user operation, wherein the individual answers displayed in the display area can be copied to the text box based on user operation.

[0012] The input screen display unit in this embodiment displays an input screen for entering the results of the consideration of the new project. The support device in this embodiment uses a document generation unit to generate documents for the new project using a large-scale language model based on the basic data of the new project acquired by the basic data acquisition unit and the consideration result data entered from this input screen. However, for users unfamiliar with creating documents for new projects, it is not easy to appropriately list the necessary consideration items and input them.

[0013] In this embodiment, the input screen display unit displays an input screen that provides an operation to include some or all of the candidate review results generated by the candidate review result generation unit, which inputs prompts including basic data and the aforementioned review result data into the large-scale language model, into the review result data. In other words, the support device in this embodiment shows the candidate review results generated by the large-scale language model to the user, enabling the user to add some or all of them to their own review results. This allows the user to combine the AI's review results with their own, enabling co-creation between humans and AI. The support device in this embodiment then generates new project materials through this co-creation. If the AI ​​alone were to generate the materials, there would be concerns that they would be similar to new project materials created in the past. However, by enabling co-creation between humans and AI, the support device in this embodiment can provide materials that reflect different review results from past materials input by humans.

[0014] Based on the above, this embodiment can support the efficient creation of materials and the improvement of their quality through co-creation between humans and AI.

[0015] Furthermore, the present invention can take various forms, as exemplified by: a form that promotes co-creation between humans and AI through questions to the inputter; a form that allows input of the results of consideration divided into three stages: advantages and prospects of the new project, concerns of the new project, and measures to overcome the concerns; a form that shows the AI's evaluation of the generated materials; and a form that enables co-creation with others in addition to co-creation with AI. Each of these forms, through the effects brought about by the addition of its unique configuration, contributes to supporting the efficient creation of materials for new projects and improving their quality through co-creation between humans and AI. [Effects of the Invention]

[0016] Based on the above, the present invention can support the efficient creation of materials and the improvement of their quality through co-creation between humans and AI. [Brief explanation of the drawing]

[0017] [Figure 1]It is a block diagram showing an example of the hardware configuration and software configuration of system S of the present embodiment. [Figure 2] It is a main flowchart showing an example of a preferred flow of support processing executed by the support device 1 of the present embodiment. [Figure 3] It is a diagram continuing from the preceding figure. [Figure 4] It is a flowchart of examination processing in the main flowchart. [Figure 5] It is a diagram continuing from the preceding figure. [Figure 6] It is a display example of an input screen E. [Figure 7] It is a display example of an evaluation screen A. [Figure 8] It is a diagram showing a partial configuration of a support device according to a second embodiment. [Figure 9] It is a diagram showing the remaining partial configuration of the support device according to the second embodiment. [Figure 10] It is a diagram showing a screen displayed by the support device according to the second embodiment. [Figure 11] It is a diagram showing a screen displayed by the support device according to the second embodiment. [Figure 12] It is a diagram showing a screen displayed by the support device according to the second embodiment. [Figure 13] It is a diagram showing a screen displayed by the support device according to the second embodiment. [Figure 14] It is a diagram showing a screen displayed by the support device according to the second embodiment. [Figure 15] It is a diagram showing a screen displayed by the support device according to the second embodiment. [Figure 16] It is a diagram showing the remaining partial configuration of the support device according to the second embodiment. [Figure 17] It is a diagram showing a screen displayed by the support device according to the second embodiment. [Figure 18] It is a diagram showing a screen displayed by the support device according to the second embodiment. [Figure 19] It is a diagram showing a screen displayed by the support device according to the second embodiment. [Figure 20] This figure shows the screen displayed by the support device according to the second embodiment. [Figure 21] This figure shows the screen displayed by the support device according to the second embodiment. [Figure 22] This figure shows the screen displayed by the support device according to the second embodiment. [Figure 23] This figure shows the screen displayed by the support device according to the second embodiment. [Figure 24] This figure shows the screen displayed by the support device according to the second embodiment. [Figure 25] This figure shows the screen displayed by the support device according to the second embodiment. [Figure 26] This figure shows the screen displayed by the support device according to the second embodiment. [Figure 27] This figure shows the screen displayed by the support device according to the second embodiment. [Figure 28] This figure shows the screen displayed by the support device according to the second embodiment. [Modes for carrying out the invention]

[0018] Firstly, although the following disclosures, figures, and / or claims are described either individually or in combination with one or more other aspects, the subject matter of the immediate disclosure is not intended to be limited in that way. That is, the immediate disclosures, figures, and claims are intended to encompass the various aspects described herein, either individually or in one or more combinations with each other. For example, even if the immediate disclosure describes and illustrates the first, second, and third embodiments in such a way that the first embodiment is described and illustrated particularly in relation to the second embodiment, or the second embodiment is described and illustrated only in relation to the third embodiment, the immediate disclosures and illustrations are not limited in that way and may include only the first embodiment, only the second embodiment, only the third embodiment, or one or more combinations of the first, second, and / or third embodiments, such as the first and second embodiments, the first and third embodiments, the second and third embodiments, or the first, second, and third embodiments.

[0019] In this text, the phrase "or" is used to mean a "non-exclusive" arrangement unless explicitly specified otherwise. For example, when we say "item x is A or B," it means either (1) item x is either A or B, or (2) item x is both A and B. In other words, the word "or" is not used to define an "exclusive" arrangement.

[0020] Furthermore, when the phrases "contain at least one" or "contain at least one of the following" are used in the text, they mean that the system or element contains one or more of the elements listed after the phrase. For example, if there are three types of elements, from element 1 to element 3, the phrases "contain at least one" or "contain at least one of the following" are interpreted as any of the following structural arrangements: a device containing element 1, a device containing element 2, a device containing element 3, a device containing element 1 and element 2, a device containing element 1 and element 3, a device containing element 2 and element 3, or a device containing element 1, element 2, and element 3.

[0021] The same interpretation is intended when the phrase "used in at least one of the following" is used in the text. Furthermore, "and / or" as used in the text is used as a linguistic conjunction to indicate that one or more of the listed elements or conditions are included or occur. For example, a device containing the first element, the second element, and / or the third element is interpreted as any of the following structural arrangements: a device containing the first element, a device containing the second element, a device containing the third element, a device containing the first and second elements, a device containing the first and third elements, a device containing the second and third elements, or a device containing the first, second, and third elements.

[0022] Furthermore, the use of the phrase "and / or" in this text signifies a "non-exclusive" arrangement, as stipulated in the Japanese Industrial Standard (JIS) "Format and Preparation Method of Standards Documents JIS Z 8301".

[0023] The following describes in detail an example of an embodiment of the present invention with reference to the drawings.

[0024] [First Embodiment] <System S>

[0025] Figure 1 is a block diagram showing an example of the hardware and software configuration of System S in this embodiment. The new project document creation support system (System S) according to this embodiment is configured to include a new project document creation support device 1 (hereinafter also simply referred to as "support device 1"). Support device 1 is configured to communicate with terminal T via network N. [Materials for a new project]

[0026] In this embodiment, "materials for new projects" include, for example, drafts or strategic planning materials related to new projects such as new business development or project development. In addition, "materials for new projects" may also include materials related to the consideration of future visions, problem identification, idea generation, etc. [Support device 1]

[0027] Support device 1 comprises various hardware components such as a control unit 11, a memory unit 13, and a communication unit 14. Support device 1 assists in the creation of materials for new projects through co-creation between humans and AI by providing a unique user interface and processing related to large-scale language models. The type of support device 1 is not particularly limited and may be, for example, a server device, a cloud server, etc. [Control Unit 11]

[0028] The control unit 11 includes a Central Processing Unit (CPU), Random Access Memory (RAM), and Read Only Memory (ROM), among other things.

[0029] The control unit 11 cooperates with at least one of the storage unit 13 and the communication unit 14 as needed. The control unit 11 implements the software components of the program of this embodiment executed by the support device 1, such as the basic data acquisition unit 111, input screen display unit 112, examination result acquisition unit 113, examination result candidate generation unit 114, information provision unit 115, collaborator examination result acquisition unit 116, data generation unit 117, evaluation generation unit 118, comparison display command unit 119, etc. The control unit 11 may also be such that its functions are realized by a processor such as a CPU executing a program recorded on a computer-readable recording medium.

[0030] The details of the support processing realized by the above-mentioned software components will be explained later using Figures 2 to 5. [Storage section 13]

[0031] The storage unit 13 is a device on which data and / or files are stored, and has a storage unit that stores data non-temporarily using a hard disk, semiconductor memory, recording medium, and memory card, etc. The storage unit 13 stores programs and the like that are executed by a microcomputer. (Large-scale language models)

[0032] The large-scale language model in this embodiment is not particularly limited. Examples of large-scale language models in this embodiment include GPT-4 and GPT-4o, which are used in ChatGPT®. In the following, the large-scale language model may be simply referred to as "AI". [Communications Section 14]

[0033] The communication unit 14 is not particularly limited as long as it connects the support device 1 to the network N and enables communication. Examples of the communication unit 14 include a network card compatible with the Ethernet standard and a communication device compatible with wireless LAN. [Network N]

[0034] The type of network N is not particularly limited as long as it enables communication with the support device 1, etc. Examples of network N include the internet, a mobile phone network, a wireless LAN, etc. [Terminal T]

[0035] Terminal T performs processes such as transmitting user input data to support device 1 and displaying various information based on display-related commands received from support device 1. The type of terminal T is not particularly limited and may be any type of terminal, such as a stationary terminal like a personal computer or a portable terminal like a tablet. [Main flowchart for support processing]

[0036] Figure 2 is a main flowchart showing an example of a preferred flow of support processing performed by the support device 1 of this embodiment. Figure 3 is a continuation of the previous figure. The following is an example of a preferred flow of support processing performed by the support device 1 of this embodiment, using Figures 2 to 3. [Step S1: Obtain basic data]

[0037] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the basic data acquisition unit 111. Then, the control unit 11 executes the process of acquiring basic data for a new project using the basic data acquisition unit 111 (basic data acquisition step). The control unit 11 then moves the process to step S2.

[0038] In the basic data acquisition step, the basic data acquisition unit 111 acquires basic data, for example, that has been input at terminal T. If some or all of the basic data is already stored in the storage unit 13 in order to enable pre-input or reuse of basic data, the basic data acquisition unit 111 may acquire that data. The basic data acquisition unit 111 may also acquire that data from the internet. In this case, the basic data acquisition unit 111 searches the internet using search words based on, for example, the heading of a new project input from terminal T. (Basic data for new projects)

[0039] The basic data for a new project acquired in the basic data acquisition step includes, for example, the headline, basic idea, target market, strengths of the project, and assets of the project.

[0040] After acquiring basic data, the support process executes a review process that involves various considerations related to the new project through co-creation between humans and AI. The following is an example of a case where the review process is divided into the first review process, the second review process, and the third review process. Note that the number of divisions in the review process and the content of each divided review process are not limited to the example below. [Step S2: First Review Process]

[0041] The control unit 11 works in cooperation with the memory unit 13 and the communication unit 14 to perform a first review process related to the merits and prospects of the new project (first review step). The control unit 11 then moves the process to step S3. In the first review process, co-creation between humans and AI regarding the merits and prospects of the new project is carried out by the input screen display unit 112, the review result acquisition unit 113, and the review result candidate generation unit 114, etc. Furthermore, the first review process can take the form of supporting co-creation with collaborators by the information provision unit 115 and the collaborator review result acquisition unit 116. Details of the first review process will be explained later using Figures 4 and 5. [Step S3: Second Review Process]

[0042] The control unit 11 works in cooperation with the memory unit 13 and the communication unit 14 to perform a second review process related to concerns regarding the new project (second review step). The control unit 11 then moves the process to step S4. In the second review process, co-creation between humans and AI regarding concerns about the new project is carried out by the input screen display unit 112, the review result acquisition unit 113, and the review result candidate generation unit 114, etc. Furthermore, the second review process can take the form of supporting co-creation with the further involvement of collaborators by the information provision unit 115 and the collaborator review result acquisition unit 116. Details of the second review process will be explained later using Figures 4 and 5. [Step S4: Third Review Process]

[0043] The control unit 11, in cooperation with the memory unit 13 and the communication unit 14, executes a third review process related to measures to overcome the above-mentioned concerns (third review step). The control unit 11 then moves the process to step S5. In the third review process, co-creation between humans and AI regarding measures to overcome the above-mentioned concerns is carried out by the input screen display unit 112, the review result acquisition unit 113, and the review result candidate generation unit 114, etc. Furthermore, the third review process may take the form of supporting co-creation with collaborators by the information provision unit 115 and the collaborator review result acquisition unit 116. Details of the third review process will be explained later using Figures 4 to 5. [Step S5: Generate the document]

[0044] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the data generation unit 117. The control unit 11 then uses the data generation unit 117 to input prompts containing basic data and examination result data related to the examination process into the large-scale language model and executes the process of generating data (data generation step). The control unit 11 then moves the process to step S6. The examination result data related to the examination process referred to here refers to data that summarizes the examination result data related to each of the first to third examination processes. [Editing Steps]

[0045] To facilitate the finalization of the document by human hands, the support process preferably further includes an editing step that edits the document generated based on the user's input. In this case, the support device 1 preferably includes a display that visualizes the degree of AI's contribution to the edited document. This allows the support device 1 to enhance the reliability of the generated and edited document and further encourage the user to adopt a workflow that does not rely solely on AI generation. [Work Time Display Step]

[0046] To make the review process transparent, it is preferable that the support process displays the time spent by the user for each of the following stages: the first review process, the second review process, the third review process, and the process related to the generation and editing of materials. This allows the user to refer to how much time they spent reviewing at each stage and to refine the review process. [Adding explanatory text step]

[0047] To demonstrate the logical nature of the AI-driven review process, it is preferable that the support process displays text explaining the data sources the AI ​​relied on and the logic it employed for each of the following stages: the first review process, the second review process, the third review process, and the process related to the generation and editing of materials. This allows users to refer to how the AI ​​reviewed the material at each stage and refine the review process.

[0048] The support process preferably includes a series of processes for displaying an evaluation of the generated material. Steps S6 to S9 are an example of such processes. [Step S6: Determine whether to evaluate]

[0049] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the evaluation generation unit 118. The control unit 11 then performs a process to determine whether to evaluate the material generated by the evaluation generation unit 118 (evaluation determination step). If the control unit 11 determines that it should evaluate the material, it moves the process to step S7; otherwise, it moves the process to step S9.

[0050] The evaluation generation unit 118 achieves the above-mentioned determination by, for example, determining whether to perform an evaluation if data requesting an evaluation is received from terminal T. [Step S7: Generate evaluation]

[0051] The control unit 11 inputs a prompt containing the data and evaluation indicators generated by the data generation unit 117 into the large-scale language model via the evaluation generation unit 118, and executes a process to generate an evaluation for the data (evaluation generation step). The control unit 11 then moves the process to step S8. (Evaluation metrics)

[0052] The evaluation indicators related to the evaluation generation step include, for example, text indicating the items to be evaluated and the score range. To ensure stable evaluations, it is preferable that the indicators include text indicating the correspondence between the content described in the document and the scores. To obtain evaluations from various perspectives, it is preferable that the evaluation indicators can be selected from multiple sets. [Step S8: Show rating]

[0053] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the comparison display command unit 119. The control unit 11 then executes a process to instruct the comparison display command unit 119 to display an evaluation of the material (display command step). The control unit 11 then moves the process to step S9. (Regarding comparative displays)

[0054] The comparative display command unit 119 preferably commands the system to display a comparative evaluation of the multiple generated documents. This allows the user to consider which documents would receive a higher evaluation based on the displayed comparative evaluations.

[0055] It is preferable that multiple documents include a history of documents generated by the same user. This allows the user to consider what kind of document would receive a higher rating based on differences in input in past generation or differences in input between past generation, and their respective ratings.

[0056] It is preferable that the multiple materials include materials from different users. This allows users to incorporate the insights of other users who are expected to have different background knowledge or perspectives. Furthermore, this can foster a sense of competition among users, which can serve as motivation. A specific example of comparative display will be explained later using Figure 7. [Step S9: Determine whether to return to the first consideration]

[0057] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to perform a process to determine whether to return to the first review (first review continuation determination step). If the control unit 11 determines to return, it moves the process to step S2; otherwise, it moves the process to step S10.

[0058] The control unit 11 achieves the above determination by, for example, determining whether to return if data is received from terminal T requesting to continue or restart the first examination. [Step S10: Determine whether to return to the second review]

[0059] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to perform a process to determine whether to return to the second review (second review continuation determination step). If the control unit 11 determines to return, it moves the process to step S3; otherwise, it moves the process to step S11.

[0060] The control unit 11 achieves the above determination by, for example, determining whether to return if data is received from terminal T requesting to continue or restart the second examination. [Step S11: Determine whether to return to the third consideration]

[0061] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to perform a process to determine whether to return to the third review (third review continuation determination step). If the control unit 11 determines to return, it moves the process to step S4; otherwise, it returns the process to step S1 and repeats the process from step S1 to step S11.

[0062] The control unit 11 achieves the above determination by, for example, determining to return if data is received from terminal T requesting to continue or restart the third review. [Flowchart of the review process]

[0063] Figure 4 is a flowchart of the review process in the main flowchart. Figure 5 is a continuation of the previous figure. The following is an example of a preferred flow of the review process using Figures 4 and 5. The review process described below corresponds to various review processes such as the first review process (step S2), the second review process (step S3), and the third review process (step S4). Therefore, in the explanation of each step below, the common parts of the first review process (step S2), the second review process (step S3), and the third review process (step S4) will be explained first, followed by the parts that differ in each process. [Step S21: Command to display the input screen]

[0064] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the input screen display unit 112. The control unit 11 then executes a process to instruct the input screen display unit 112 to display an input screen for inputting the review result data related to the new project (input screen display command step). The control unit 11 then moves the process to step S22. (If the step for generating candidate review results has already been performed)

[0065] If the candidate analysis result generation step (described later) has been completed, the input screen display unit 112 commands the display of an input screen that provides an operation to include some or all of the candidate analysis results generated in that step in the input of the analysis result data. Such an input screen is implemented, for example, by displaying a list of candidate analysis results and by user interface elements such as copy buttons and paste buttons that allow some or all of the candidates to be pasted into the field where the analysis results are entered. If candidate analysis results that include questions for the inputter have been generated, it is preferable that the input screen be configured to allow the input of analysis result data that includes answers to those questions.

[0066] The input screen preferably includes an operation to exclude some or all of the candidate results from the input of the result data. Furthermore, in order to reduce the effort required of users involved in co-creation, the operation to include some or all of the candidate results from the input of the result data preferably completes in a single basic operation procedure, such as a single click. This allows users to switch between data to be handed over to the AI ​​and data not to be handed over in a single operation.

[0067] In order to incorporate AI suggestions for each cohesive candidate, if the candidate results are divided into multiple groups, it is preferable for the input screen display unit 112 to instruct the display of an input screen that provides an operation to select one of the divided candidates and include part or all of it in the input of the candidate results data. (In the case of the first review process)

[0068] In the first review process, the input screen display unit 112 commands the display of the first input screen for inputting the first review result data, including the advantages and prospects of the new project. This allows the support device 1 to support co-creation between humans and AI regarding the advantages and prospects of the new project. (In the case of the second review process)

[0069] In the second review process, the input screen display unit 112 commands the display of the second input screen for inputting the second review result data, which includes concerns about the new project. This allows the support device 1 to support co-creation between humans and AI regarding concerns about the new project. (In the case of the third review process)

[0070] In the third review process, the input screen display unit 112 commands the display of the third input screen for inputting the third review result data, which includes measures to overcome the concerns. This allows the support device 1 to support co-creation between humans and AI regarding measures to overcome the concerns of the new project. [Step S22: Obtain the results of the review]

[0071] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the examination result acquisition unit 113. The control unit 11 then executes the process of acquiring the examination result data entered via the input screen described above by the examination result acquisition unit 113 (examination result acquisition step). The control unit 11 then moves the process to step S23. (In the case of the first review process)

[0072] In the first review process, the review result acquisition unit 113 acquires the first review result data described above. (In the case of the second review process)

[0073] In the second review process, the review result acquisition unit 113 acquires the second review result data described above. (In the case of the third review process)

[0074] In the third review process, the review result acquisition unit 113 acquires the above-mentioned third review result data. [Step S23: Determine whether to generate candidate results for consideration]

[0075] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the candidate analysis result generation unit 114. The control unit 11 then performs a process to determine whether to generate candidate analysis results, which are candidates for the aforementioned analysis results, using the candidate analysis result generation unit 114 (candidate analysis result generation determination step). If the control unit 11 determines that it should generate candidates, it moves the process to step S24; otherwise, it moves the process to step S25. [Step S24: Generate candidate results]

[0076] The control unit 11 uses the candidate analysis result generation unit 114 to input a prompt containing the above-mentioned basic data, or a prompt containing the above-mentioned basic data and the above-mentioned analysis result data, into the large-scale language model and executes a process to generate the above-mentioned candidate analysis result (candidate analysis result) (candidate analysis result generation step). The control unit 11 then moves the process to step S25.

[0077] In the candidate analysis result generation step, the candidate analysis result generation unit 114 inputs, for example, prompts containing already entered data from the basic data and various analysis result data into the large-scale language model. (Questions for the person who entered the information)

[0078] In the step of generating candidate review results, it is preferable that the candidate review result generation unit 114 generates candidate review results that include questions to the inputter regarding the data, including basic data or review result data. This allows the support device 1 to promote co-creation between humans and AI through questions to the inputter. (In the case of the first review process)

[0079] In the first review process, the candidate review result generation unit 114 inputs a prompt containing basic data into the large-scale language model to generate the above-mentioned candidate first review result (first review result candidate). This allows the candidate review result generation unit 114 to generate candidate review results related to the merits and prospects of the new project based on the basic data. If first review result data has already been input, the candidate review result generation unit 114 may input a prompt further containing the input first review result data into the large-scale language model. This allows the candidate review result generation unit 114 to generate candidate review results for the merits and prospects of the new project by utilizing the input first review result data. (In the case of the second review process)

[0080] In the second review process, the candidate review result generation unit 114 inputs a prompt containing basic data and first review result data into the large-scale language model to generate the above-mentioned candidate for the second review result (second review result candidate). As a result, the candidate review result generation unit 114 can generate candidate review results related to concerns of the new project based on the basic data and first review result data. If second review result data has already been input, the candidate review result generation unit 114 may input a prompt further containing the input second review result data into the large-scale language model. As a result, the candidate review result generation unit 114 can generate candidate review results related to concerns of the new project by utilizing the input second review result data. (In the case of the third review process)

[0081] In the third review process, the candidate review result generation unit 114 inputs a prompt containing basic data, first review result data, and second review result data into the large-scale language model to generate the above-mentioned candidate for the third review result (third review result candidate). This allows the candidate review result generation unit 114 to generate candidate review results related to measures to overcome concerns regarding the new project, based on the basic data, first review result data, and second review result data. If third review result data has already been input, the candidate review result generation unit 114 may input a prompt further containing the already inputted third review result data into the large-scale language model. This allows the candidate review result generation unit 114 to generate candidate review results related to measures to overcome concerns regarding the new project, utilizing the already inputted third review result data.

[0082] The review process preferably includes a series of processes related to co-creation with collaborators other than the data inputters, such as basic data and review result data. Steps S25 to S28 are an example of such a process. [Step S25: Determine whether to provide information to the collaborator]

[0083] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the information provision unit 115. The control unit 11 then performs a process to determine whether to provide the information to a collaborator different from the person who input the data (information provision determination step). If the control unit 11 determines that it should provide the information, it moves the process to step S26; otherwise, it moves the process to step S28.

[0084] The control unit 11 achieves the above determination by, for example, a procedure that determines whether to provide the service if a collaborator has been specified from terminal T. [Step S26: Provide to collaborators]

[0085] The control unit 11 performs the process of providing data including basic data or the aforementioned study result data to a collaborator different from the person who input the data, via the information provision unit 115 (information provision step). The control unit 11 then moves the process to step S27. [Step S27: Obtain the results of the collaborator review]

[0086] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the collaborator review result acquisition unit 116. The control unit 11 then executes the process of acquiring the collaborator review results from the collaborators using the collaborator review result acquisition unit 116 (collaborator review result acquisition step). The control unit 11 then moves the process to step S28. (Regarding the provision of collaborator review results on the input screen)

[0087] In the input screen, it is preferable that the results of the collaborator review are provided in a manner that allows for the inclusion of some or all of the results in the input of the review result data, similar to the candidate review results. [Step S28: Determine whether to continue the investigation]

[0088] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to perform a process to determine whether to continue the review (review continuation determination step). If the control unit 11 determines to continue, it returns to step S21; otherwise, it moves the process to the step following the step that transitioned to the review process. (In the case of the first review process)

[0089] In the case of the first review process, if it is not determined to continue, the next step is step S3. (In the case of the second review process)

[0090] In the case of the second review process, if it is not determined to continue, the next step is step S4. (In the case of the third review process)

[0091] In the case of the third review process, if it is not determined to continue, the next step is step S5. [Effects of support processing]

[0092] In the support process, the basic data acquisition unit 111 acquires basic data for the new project (step S1). The input screen display unit 112 also commands the display of the input screen E for inputting the results of the consideration of the new project (step S21).

[0093] However, for users who are not sufficiently proficient in creating materials for new projects, it is not easy to appropriately list the various points that need to be considered and input the results of their considerations for those points. On the other hand, AI can output a large amount of consideration results in a short time compared to humans. Furthermore, AI has a broader range of knowledge than humans. Therefore, if human ideas and selections can be effectively combined with the consideration results of AI, excellent materials for new projects can be created in a short time through co-creation between humans and AI.

[0094] To achieve this co-creation, the candidate result generation unit 114 inputs prompts containing basic data and result data into a large-scale language model to generate candidate results (steps S23 to S24). Then, the input screen display unit 112 displays an input screen that provides an operation to include some or all of these candidate results in the result data (step S21).

[0095] Then, the review result acquisition unit 113 acquires the review result data entered from the input screen (step S22). The support device 1, using the material generation unit 117, generates materials for the new project using a large-scale language model based on the basic data of the new project acquired by the basic data acquisition unit 111 and the review result data entered from this input screen (step S5).

[0096] In other words, support device 1 presents the user with candidate results for consideration generated by the large-scale language model, allowing the user to add some or all of these results to their own considerations. This enables co-creation between humans and AI, where the user combines the AI's considerations with their own. Through this co-creation, support device 1 generates materials for new projects.

[0097] If the materials were generated solely by AI, there would be concerns that they would be similar to those created for new projects in the past. However, the support device 1, which performs the aforementioned support processing, facilitates co-creation between humans and AI, enabling it to provide materials that reflect different considerations than those previously created by humans, and in a shorter time than if created solely by humans. Therefore, the support device 1, which performs the support processing, can support the efficient creation of materials for new projects and improve their quality through co-creation between humans and AI.

[0098] In other words, the support device 1 of this embodiment can realize Human in the Loop (HITL), an approach in which humans and AI cooperate to solve problems. In this approach, the AI ​​makes judgments and analyses, and humans evaluate and correct the results. As a result, more accurate and efficient results can be obtained.

[0099] Furthermore, the support process may take the form of providing candidate review results that include questions to the inputter (step S24). This allows the inputter to determine what should be entered as review result data through the questions. Thus, the support device 1 in this form can promote co-creation between humans and AI through questions to the inputter.

[0100] In addition, the support process can be configured to input the results of the examination in three stages: the advantages and prospects of the new project, the concerns of the new project, and measures to overcome the concerns (steps S2 to S4). This allows the support device 1 to sequentially assist in co-creating the examination results related to these important points in the new project materials.

[0101] However, for users like those described above, it is not easy to judge the quality of the generated materials. The support process provides the user with evaluation criteria for the generated materials by showing the AI's evaluation of the generated materials (steps S6 to S8).

[0102] It is expected that by collaborating with multiple people and AI, materials reflecting a broader range of knowledge will be created. The support process can take the form of providing various data to collaborators different from the data inputter and obtaining the collaborators' review results (steps S25 to S27). This allows the support device 1 to perform not only co-creation with AI but also co-creation involving other people. <Usage example>

[0103] The following is an example of using System S of this embodiment. [Entering basic data]

[0104] The user inputs basic data related to the creation of materials for a new project into terminal T. Terminal T transmits this data to support device 1. Support device 1 receives this data. [Input of First Review Results]

[0105] Support device 1 instructs terminal T to display input screen E, which prompts the user to input the results of the first review. Terminal T displays input screen E. The user inputs a draft of the first review results, including the advantages and prospects of the new project.

[0106] The user instructs the support device 1 to perform an AI-driven analysis via terminal T. The support device 1 instructs the display of input screen E, which includes candidate analysis results. Terminal T displays the input screen E.

[0107] Figure 6 shows an example of the display of input screen E. At the top of this example, there is a display indicating the flow of the process, which involves sequentially performing "Preparation" for inputting basic data, "Advantages and Outlook" for inputting the first set of study results data, "Concerns" for inputting the second set of study results data, "Measures to overcome" for inputting the third set of study results data, and "Draft proposal" for generating materials, as well as indicating that the current step is "Advantages and Outlook".

[0108] Below the display showing the flow, there is an input field with the heading "Highlight the advantages." The input field contains a draft written by the user: "Economic value: Efficiency will increase significantly, saving human resources and reducing costs. Functional value: High-quality proposals can be created in a short time, so work The initial review results data shows that "the process will become smoother..."

[0109] Further down, there is a heading that reads "AI Review Results List." Below that, two candidate review results are displayed, along with buttons to copy the candidate results to the input field mentioned above, to edit the candidate results, to print the candidate results or output them as a PDF file, and to hide the candidate results.

[0110] The first candidate for the review results begins with: "Social value: By promoting efficient work processes, it will contribute to increased productivity and strengthen the overall competitiveness of the company. Organizational culture transformation: By increasing work efficiency, more time will be freed up..."

[0111] The second one begins with the following: "• Time saving • User-friendly UI • Output tailored to needs..."

[0112] Further down, there is a heading that reads "List of other users' ideas." Below that, the results of two collaborative discussions are displayed, along with buttons to copy them to the input field mentioned above, to edit them, to print or output them as a PDF file, and to hide them.

[0113] The first result of the collaborator consideration, under the heading "User 1," presents the result of the collaborator consideration, stating, "Why not just have AI evaluate the project proposal as well, and then collaboratively refine it by repeatedly rewriting it while looking at the AI's evaluation?"

[0114] The second point is that, along with a headline identifying the collaborator with "User 2," it presents the results of the collaborator consideration, stating, "Wouldn't readers feel more at ease if we visualized the degree of AI's contribution? I think there was a study on this at Duke University's Department of Psychiatry and Neurology."

[0115] In this example, the user can refine the initial evaluation result data by incorporating standardized evaluation result candidates based on broad knowledge generated by AI, as well as evaluation results from collaborators that offer perspectives not available to the AI. [Input of the results of the second review]

[0116] Support device 1 instructs terminal T to display input screen E, which prompts the user to input the second review results. Terminal T displays input screen E. The user inputs a draft of the second review results, including concerns about the new project. In addition, similar to the input of the first review results, the user refines the second review result data by incorporating standardized review result candidates based on broad knowledge generated by the AI, and collaborator review results presented from perspectives not available to the AI. [Input of the results of the third review]

[0117] Support device 1 instructs terminal T to display input screen E, which prompts the input of the third review results. Terminal T displays input screen E. The user inputs a draft of the third review results, which includes measures to overcome concerns regarding the new project. In addition, similar to the input of the first review results, the user refines the third review result data by incorporating standardized review result candidates based on broad knowledge generated by the AI, and collaborator review results presented from perspectives not available to the AI. [Generating documents]

[0118] The user, via terminal T, instructs support device 1 to generate materials for a new project based on the input described above. Support device 1 generates the materials and displays them on terminal T. [Refinement using evaluation]

[0119] The user instructs the support device 1 to evaluate the generated materials via terminal T. The support device 1 instructs terminal T to display a comparison of evaluations for multiple materials. Terminal T displays evaluation screen A.

[0120] Figure 7 shows an example of the display of evaluation screen A. In this example, following the heading "AI Evaluation," the scores and reasons for the evaluation for items 1 through 7 are shown. Further below, it is shown that the total score is 25 out of 35 points.

[0121] Further down, there are overall comments beginning with "This proposal utilizes new technology..." and advice for improving the score beginning with "The customer's problems...". Users can refer to these to refine their documents.

[0122] At the bottom of evaluation screen A, under the heading "Draft History," a summary of previously generated documents is displayed. The first entry in the history shows a heading indicating a document generated on "2024 / 10 / 01 12:00:00" with a "Total Score of 26 / 35," along with a button to view that document. The second entry in the history shows a heading indicating a document generated on "2024 / 10 / 01 11:50:00" with a "Total Score of 20 / 35," along with a button to view that document. Users can refer to these to refine their documents and achieve a higher score.

[0123] Further down, under the heading "Other Users' Proposals," a list of documents generated by other users is displayed. The first item in the list is a document created by a user identified as "User 3," titled "Iterative Processing Using AI," with a "Total Score of 27 / 35," and is indicated by a heading and a button to view the document. The second item in the list is a document created by a user identified as "User 4," with no title, also with a "Total Score of 27 / 35," and is indicated by a heading and a button to view the document. Users can refer to these documents to improve their own and achieve higher scores.

[0124] In the above embodiment, if the process has completed up to S8, and the result is NO at branch S9, S10, or S11, the process is returned to S2, S3, or S4 and restarted from there. However, this is not always the case, and the process may be returned even before reaching S8. For example, in the screen shown in Figure 6, if the user selects the "Preparation" tab, the "Advantages / Outlook" tab, the "Concerns" tab, the "Strategies to Overcome" tab, the "Draft Proposal" tab, or the "Proposal F" tab, the process may be returned to the step corresponding to the selected tab and restarted from there. For example, if the user selects the "Advantages / Outlook" tab while S4, which corresponds to the "Concerns" tab, is being executed, the process may be returned to S3 and restarted from there.

[0125] [Second Embodiment]

[0126] Figures 9 and 10 are block diagrams showing the configuration of the support device 900 according to the second embodiment. As shown in Figures 9 and 10, the support device 900 basically has a group of blocks sandwiched between six LLM utilization units and two adjacent LLM utilization units. The six LLM utilization units include a multi-output LLM utilization unit 902, a multi-output LLM utilization unit 909, a multi-output LLM utilization unit 916, a single-output LLM utilization unit 923, a single-output LLM utilization unit 926, and a single-output LLM utilization unit 929. The six LLM utilization units utilize an externally provided LLM (not shown). The multi-output LLM utilization units 902, 909, and 915 output multiple datasets using the LLM. The single-output LLM utilization units 923, 926, and 929 output a single dataset using the LLM.

[0127] As will be explained in more detail later, the multiple output LLM unit 902 outputs three datasets related to market segments per output. The multiple output LLM unit 909 outputs three datasets related to problems per output. The multiple output LLM unit 915 outputs three datasets related to solutions per output. Note that the number "3" is just an example; any integer greater than or equal to 2 is acceptable.

[0128] The single-output LLM utilization unit 923 outputs one dataset per output, consisting of a concept and its evaluation.

[0129] The single-output LLM utilization unit 926 outputs one dataset related to a specific concept for each output. Here, the dataset output by the single-output LLM utilization unit 923 includes questions related to the concept, and the dataset output by the single-output LLM utilization unit 926 includes the answers to those questions as a specific concept.

[0130] The single-output LLM utilization unit 929 outputs one dataset that constitutes a business plan for each output.

[0131] There are multiple blocks between each LLM utilization unit and the next LLM utilization unit, and these blocks basically operate in response to user operations. Therefore, the user intervenes between LLMs, and this repeated process enables a Human in the Loop (HITL) configuration.

[0132] The holding units 903-1 to 903-3 hold each of the datasets output by the multiple output LLM utilization unit 902.

[0133] The replication unit 904 replicates, for example, a dataset held in the storage unit 903-1 to the editing unit 906 based on user operations. Similarly, the replication unit 904 replicates, for example, datasets held in the storage units 903-2 and 903-3 to the editing unit 906 based on user operations. The input unit 905 also supplies the data entered by the user to the editing unit 906 based on user operations. Therefore, the editing unit 906 can input either the dataset output by the replication output LLM utilization unit 302 or the dataset entered by the user, and edit it.

[0134] Based on user operations, the storage units 907-1 to 907-n1 store the dataset currently stored in the editing unit 906. Therefore, for example, storage unit 907-1 can store the dataset that was stored in storage unit 903-1, storage unit 907-2 can store the dataset that was stored in storage unit 903-3, and storage unit 907-3 can store the dataset entered by the user.

[0135] The selection unit 908 basically supplies the datasets held in the storage unit 907-n1 to the multiple output LLM utilization unit 909 selectively based on user operations, from the datasets held in the storage unit 907-1. However, the selection unit 908 can also selectively supply any number of datasets from the datasets held in the storage unit 907-n1 to the multiple output LLM utilization unit 909 selectively based on user operations.

[0136] Next, since each functional block from the multiple output LLM utilization unit 909 to the selection unit 915 is the same as each functional block from the multiple output LLM utilization unit 902 to the selection unit 918, redundant explanations will be omitted.

[0137] Each functional block from the multiple output LLM utilization unit 916 to the selection unit 922 is the same as each functional block from the multiple output LLM utilization unit 902 to the selection unit 918, so redundant explanations will be omitted.

[0138] As described above, the single-output LLM utilization unit 923 outputs one dataset per output, containing a concept and its evaluation. The selection unit 922 sequentially selects datasets held in the holding units 921-1 to 921-n3 and supplies them to the single-output LLM utilization unit 923, so that the datasets sequentially output by the single-output LLM utilization unit 923 can be held in multiple holding units 924. The number of holding units 924 is represented by n4 instead of n3 because there are various ways to use them, and the number varies depending on the usage.

[0139] As described above, the single-output LLM utilization unit 926 outputs one dataset related to a specific concept for each output. The selection unit 922 sequentially selects datasets held in the holding units 924-1 to 924-n4 and supplies them to the single-output LLM utilization unit 926, so that the datasets sequentially output by the single-output LLM utilization unit 926 can be held in each of the multiple holding units 927. The number of holding units 927 is represented by n5 instead of n3 or n4 because there are various ways to use them, and the number varies depending on how they are used.

[0140] As described above, the single-output LLM utilization unit 929 outputs one dataset constituting a business concept plan for each output. The selection unit 928 sequentially selects datasets held in the holding units 927-1 to 927-n5 and supplies them to the single-output LLM utilization unit 929, so that the datasets sequentially output by the single-output LLM utilization unit 929 can be held in multiple holding units 930. The number of holding units 930 is represented by n6 instead of n3, n4, and n5 because there are various uses and the number varies depending on the use. The output unit 931 outputs the datasets held in the holding units 930, for example, a file generation device, an image display device, or a printer.

[0141] The selection unit 908 may select a dataset held in multiple storage units 907-i1 to 907-ix, rather than just one storage unit 907-i, and supply it to the multiple output utilization unit 908. In this case, the multiple output LLM utilization unit 909 may output the dataset by, for example, dividing it into cases for each dataset. For example, the selection unit 908 may select a dataset held in three storage units 907-i1 to 907-i3 and supply it to the multiple output LLM utilization unit 908. In this case, the multiple output LLM utilization unit 909 may, for example, divide it into three cases and output three datasets for each case (i.e., a total of nine datasets).

[0142] Furthermore, the multiple output LLM utilization unit 909 outputs, for example, three datasets for each dataset selected by the selection unit 909. Therefore, if the selection unit 909 sequentially selects three datasets, the multiple output LLM utilization unit 909 will output a total of 3 × 3 = 9 datasets. Similarly, the multiple output LLM utilization unit 916 outputs, for example, three datasets for each dataset selected by the selection unit 915. Therefore, if the selection unit 9159 sequentially selects nine datasets, the multiple output LLM utilization unit 916 will output a total of 9 × 3 = 27 datasets. Consequently, for example, each of the holding units 921-1 to 921-27 will hold its respective dataset. However, by using datasets in time-sharing mode, there is no need to increase the number of holding units.

[0143] Next, with reference to Figures 10 to 28, which show the display screen, a detailed explanation of the operation of the support device 900 according to the second embodiment will be given.

[0144] The support device 900 assists in generating a business plan through the following steps 0 to 5.

[0145] Step 0: Assets / Strategy Step 1: Market Selection Step 2: Problem Identification Step 3: Idea Generation Step 4: Concept Evaluation Step 5: Report

[0146] Figure 10 shows the screen for Step 0. Figures 11 to 14 show the screen for Step 1. Figures 15 to 17 show the screen for Step 2. Figures 18 to 22 show the screen for Step 3. Figures 23 to 26 show the screen for Step 4. Figures 27 and 28 show the screen for Step 5.

[0147] Referring to Figure 10, the Step 0 screen has a text box 1001 for entering the industry of the company to be developed. Figure 10 shows an example where the user enters "Zabuton Maker" as the industry of the company to be developed in the text box 1001 and then presses the "Submit" button 1002, resulting in the string "Zabuton Maker" 1003 being displayed. The string "Zabuton Maker" 1003 is held in the holding unit 901.

[0148] To the left of string 1003 are icons 1004 to activate the copy function, 1005 to activate the edit function, 1006 to hide the associated display object (also called a "toggle component"), and 1007 to activate the delete function. When the user selects icon 1004, string 1003 is copied to the clipboard. When the user selects icon 1005, a text box (not shown) for editing string 1003 is displayed, and the user can edit string 1003 using this text box. When the user selects icon 1006, string 1003 and icons 1004 through 1007 are hidden. When the user selects icon 1007, string 1003 is deleted. The functions of similar icons displayed elsewhere are the same.

[0149] The Step 0 screen has a text box 1008 for entering strengths or assets that could be used in business development. Figure 10 shows an example where the user enters "We can design, manufacture, and sell cushions with new shapes and new materials." as a strength or asset that could be used in business development into the text box 1008, and then presses the "Submit" button 1009, resulting in the string "We can design, manufacture, and sell cushions with new shapes and new materials." 1010 being displayed. The string "We can design, manufacture, and sell cushions with new shapes and new materials." 1010 is held in the holding unit 901.

[0150] The Step 0 screen has a text box 1011 for entering business development and innovation policies and strategies. Figure 10 shows an example where the user enters "Design, manufacture, and sell new donut-shaped cushions." as the business development and innovation policy or strategy in the text box 1011 and then presses the "Submit" button 1012, resulting in the string "Design, manufacture, and sell new donut-shaped cushions." 1013 being displayed. The string "Design, manufacture, and sell new donut-shaped cushions." 1013 is held in the holding unit 901.

[0151] The Step 0 screen has a text box 1014 for the user to confirm and modify the session name. Figure 10 shows an example where the user has already entered "donut-shaped cushion" in a screen not shown, and this is displayed in the text box 1014. When the user selects icon 1013, the user can edit the session name in the text box 1014.

[0152] Referring to Figure 11, the Step 1 screen has a text box 1101 for entering the target market for business development. Figure 11 shows an example where the user enters "Office," "Home," "Nursing Home," and "Hospital" as the target market for business development in the text box 1011, and presses the "Submit" button 1102 each time an entry is made, resulting in the display of the strings "Office" 1103, "Home" 1104, "Nursing Home" 1105, and "Hospital" 1106. Each of the strings "Office" 1103, "Home" 1104, "Nursing Home" 1105, and "Hospital" 1106 is held in the holding unit 901.

[0153] The Step 1 screen has an "Execute" button (also called a "starter component") 1107 to generate candidate market segments in the LLM.

[0154] When the user presses the "Execute" button 1107, the multiple output LLM utilization unit 902 is activated. The multiple output LLM utilization unit 902 causes the LLM to generate favorable segments in each market based on the following conditions (a) to (d).

[0155] (a) The industry of the company that conducts business development is "cushion manufacturer" as held in the holding section 901. (b) A strength or asset that could be utilized in business development is the "ability to design, manufacture, and sell cushions of new shapes and materials" held in the holding unit 901. (c) As a business development and innovation policy and strategy, the statement held in the holding unit 901 is "Design, manufacture, and sell new donut-shaped cushions." (d) The markets to be targeted for business development are "offices," "homes," "nursing homes," and "hospitals" held in the holding unit 901.

[0156] The multiple output LLM utilization unit 902 displays a results list 1108 based on the information about favorable segments in each market generated by the LLM.

[0157] As shown in Figures 12 and 13, the results list 1108 includes a table 1109 of candidate market segments, a first table 1110 of generated segments, a second table 1111 of generated segments, and a trace log table 1112.

[0158] As shown in Figure 12, a detailed explanation of the contents of Tables 1109 to 1111 will be omitted, but Table 1109 lists the industry and the market being entered. Table 1110 lists five segments. Table 1111 lists three segments. There are a total of eight records, and accordingly, the dataset for each record is stored in eight storage units 903. Trace logs will be discussed later.

[0159] Referring to Figure 13, the Step 1 screen has a text box 1113 for setting the market segment below the results list 1108.

[0160] The user copies the text of a market segment from either Table 1110 or Table 1111, pastes it into text box 1113, and then clicks the submit button 1114. The text will then be displayed below text box 1113. Note that the individual specific market segments in the market segment column of Table 1110 and the individual specific segments in the market segment column of Table 1111 may be displayed in a manner that distinguishes them from other texts, so that they can be selected as market segments to paste into text box 1113. A manner that distinguishes them from other texts may include, for example, using bold or colored fonts, or adding a marker.

[0161] For example, if you target "Office Cushion Market" in Table 1110 with the above operation, the string "Office Cushion Market" 1115 will be displayed as shown in Figure 13. Similarly, if you target "Health-Oriented Cushion Market" in Table 1111 with the above operation, the string "Health-Oriented Cushion Market" 1116 will be displayed as shown in Figure 13. If you target "Eco-Friendly Cushion Market" in Table 1111 with the above operation, the string "Eco-Friendly Cushion Market" 1117 will be displayed as shown in Figure 13.

[0162] The user can toggle the display of string 1115 by clicking its corresponding icon 1006B or by clicking the open / close screen control (also called a "toggle component") 1118. The same applies to strings 1116 and 1117.

[0163] Figure 14(a) shows an example where only the string 1115 is displayed by this operation, Figure 14(b) shows an example where only the string 1116 is displayed by this operation, and Figure 14(c) shows an example where only the string 1117 is displayed by this operation.

[0164] Text box 1113 corresponds to the editing unit 906 shown in Figure 8. The holding unit 907, also shown in Figure 8, holds the strings 1115 to 1117. The selection unit 908, also shown in Figure 8, selects the string that has become visible through the above operation.

[0165] Referring to Figure 15, the Step 2 screen has an "Execute" button 1501 to generate candidate tasks in the LLM.

[0166] When the user presses the "Execute" button 1501, the multiple output LLM utilization unit 909 is activated. The multiple output LLM utilization unit 909 causes the LLM to generate problems in the "office cushion market" based on the content of the necessary items from the following items (a) to (e).

[0167] (a) The industry of the company that conducts business development is "cushion manufacturer" as held in the holding section 901. (b) A strength or asset that could be utilized in business development is the "ability to design, manufacture, and sell cushions of new shapes and materials" held in the holding unit 901. (c) As a business development and innovation policy and strategy, the statement held in the holding unit 901 is "Design, manufacture, and sell new donut-shaped cushions." (d) The markets to be targeted for business development are "offices," "homes," "nursing homes," and "hospitals" held in the holding unit 901. (e) The "office cushion market" held in the holding unit 907 as a favorable segment.

[0168] The required items are, for example, items (a), (d), and (e). However, the configuration of this support device 900 is such that it is possible to switch the selection / deselection of each of items (a) to (e).

[0169] As shown in Figure 14(a), only the string "Office cushion market" 1115 is displayed as a favorable segment, and the selection unit 908 is assumed to have selected only the favorable segment indicated by the string "Office cushion market" 1115.

[0170] The multiple output LLM utilization unit 909 displays a results list 1502 based on the information generated by the LLM regarding challenges in the office cushion market.

[0171] As shown in Figure 15, the results list 1502 includes a prerequisites table 1503, a table of generated issues 1504, and a trace log 1505.

[0172] As it is shown in Figure 15, a detailed explanation of the contents of Tables 1503 and 1504 will be omitted, but Table 1503 lists the industry, target market, and segments. Table 1504 lists three challenges.

[0173] Here, as shown in Figure 14(b), if only the string "Health-Oriented Cushion Market" 1116 is displayed on the Step 1 screen and the selection unit 908 selects only the string "Health-Oriented Cushion Market" 1116, the contents of tables 1503 and 1504 will be different, but a detailed explanation will be omitted. Similarly, as shown in Figure 14(c), if only the string "Eco-Friendly Cushion Market" 1117 is displayed and the selection unit 908 selects only the string "Eco-Friendly Cushion Market" 1117, the contents of tables 1503 and 1504 will be different, but a detailed explanation will be omitted.

[0174] Referring to Figure 16, the Step 2 screen has a text box 1506 for setting the task below the results list 1502.

[0175] The user copies a string from one of the multiple sets of status, issue, target, competition, and obstacles listed in Table 1504, pastes it into text box 1506, and then clicks the submit button 1507. The string then appears below text box 1506. Note that the individual specific status, issue, target, competition, and obstacles in the status, issue, target, competition, and obstacle columns of Table 1504 may be displayed in a manner that distinguishes them from other strings, allowing them to be selected as the status, issue, target, competition, and obstacle to be pasted into text box 1506. Distinguishing them from other strings might include, for example, using bold or colored fonts, or adding a marker.

[0176] For example, if the set in the first record of table 1504 (the set consisting of "Remote work is becoming widespread, and work from home is increasing," "I want to maintain a comfortable posture while working," "Business people working from home," "Office chairs or regular chairs," and "Sitting for long periods of time is strenuous.") is targeted for the above operation, the string corresponding to that set, "Remote work is becoming widespread, and work from home is increasing. I want to maintain a comfortable posture while working. Business people working from home. Office chairs or regular chairs. Sitting for long periods of time is strenuous." 1508 will be displayed, as shown in Figure 16. Similarly, if the set in the second record of table 1504 is targeted for the above operation, the string corresponding to that set, 11509, will be displayed, as shown in Figure 16. If the set in the third record of table 1504 is targeted for the above operation, the string corresponding to that set, 1510, will be displayed, as shown in Figure 16.

[0177] The user can toggle the display of string 1508 by clicking its corresponding icon 1006C or by clicking the open / close screen control 1511. The same applies to strings 11509 and 11510.

[0178] Figure 17(a) shows an example where only the string 1508 is displayed by this operation, Figure 17(b) shows an example where only the string 1509 is displayed by this operation, and Figure 17(c) shows an example where only the string 11510 is displayed by this operation.

[0179] Text box 1506 corresponds to the editing unit 913 shown in Figure 8. The holding unit 914 shown in Figure 8 holds the strings 1508 to 1510. The selection unit 915 shown in Figure 8 selects the string that has become visible through the above operation.

[0180] Referring to Figure 18, the Step 3 screen has an "Execute" button 1801 to cause the LLM to generate candidate solutions.

[0181] When the user presses the "Execute" button 1801, the multiple output LLM utilization unit 916 is activated. The multiple output LLM utilization unit 916 causes the LLM to generate a solution based on the content of the necessary items from the following items (a) to (f).

[0182] (a) The industry of the company that conducts business development is "cushion manufacturer" as held in the holding section 901. (b) A strength or asset that could be utilized in business development is the "ability to design, manufacture, and sell cushions of new shapes and materials" held in the holding unit 901. (c) As a business development and innovation policy and strategy, the statement held in the holding unit 901 is "Design, manufacture, and sell new donut-shaped cushions." (d) The markets to be targeted for business development are "offices," "homes," "nursing homes," and "hospitals" held in the holding unit 901. (e) The "office cushion market" held in the holding unit 907 as a favorable segment, and the challenges in the "office cushion market" (f) The data set related to the issue is stored in the holding unit 914 as follows: "With the spread of remote work, the amount of work done at home is increasing. People want to maintain a comfortable posture while working. Business people working from home. Office chairs and regular chairs. Sitting for long periods of time puts a strain on the body."

[0183] The required items are, for example, items (a), (b), (c), (d), (e), and (f). However, the configuration of this support device 900 is such that it is possible to switch the selection / deselection of each item (a) through (f).

[0184] As shown in Figure 17(a), on the Step 2 screen, only the issue indicated by the string 1508, "With the spread of remote work, the amount of work done at home is increasing. I want to maintain a comfortable posture while working. Business people working at home. Office chairs and regular chairs become a burden when sitting for long periods of time," is displayed, and the selection unit 915 is assumed to have selected only the issue indicated by the string 1508, "With the spread of remote work, the amount of work done at home is increasing. I want to maintain a comfortable posture while working. Business people working at home. Office chairs and regular chairs become a burden when sitting for long periods of time."

[0185] The multiple output LLM utilization unit 916 displays solution 1802 based on the information generated by the LLM regarding challenges in the office cushion market.

[0186] As shown in Figures 18 and 19, solution 1802 includes a prerequisites table 1803, tables 1804, 1805 and 1806 concerning the generated solution, and a trace log 1807.

[0187] As explained in Figures 18 and 19, a detailed explanation of the contents of Tables 1803-1806 will be omitted. However, Table 1803 lists the industry, target market, segment, and challenges. Tables 1804-1806 each list one solution.

[0188] Here, as shown in Figure 17(b), if only string 1509 is displayed and the selection unit 915 selects only string 1509, the contents of tables 1803 to 1806 will be different, but a detailed explanation will be omitted. Similarly, as shown in Figure 17(c), if only string 11510 is displayed and the selection unit 915 selects only string 11510, the contents of tables 1803 to 1806 will be different, but a detailed explanation will be omitted.

[0189] The user can toggle the display of the contents of Solution 1802 by clicking the corresponding icon 1006D (see Figure 19) or the open / close screen control 1807 (see Figure 18). When the contents of Solution 1802 are hidden, only the title of Solution 1802 is displayed on the screen. In the screen shown in Figure 20, only the title of Solution 1802B is displayed.

[0190] The screen shown in Figure 20 also displays solutions 2001 and 2002, which only show the titles. Solution 2001 is the solution generated when the selection unit 915 selects only the string 1509. Solution 2002 is the solution generated when the selection unit 915 selects only the string 1510. The user can switch the display / hide of the contents of each of the three solutions by operating the open / close screen controls 1807, 2003, and 2004.

[0191] Referring to Figure 20, the Step 3 screen has a text box 2006 for setting the solution, under the results list 2005 which includes solutions 1807, 2003, and 2004.

[0192] For example, a user can copy the string representing the set of means, purpose, form of provision, and supplementary information listed in Table 1804, paste it into text box 2006, and then press the submit button 2007. This will display the same string, 2008, below text box 1506. Similarly, the same string, 2009, representing the set of means, purpose, form of provision, and supplementary information listed in Table 1805, can be displayed below it. Furthermore, the same string, 2010, representing the set of means, purpose, form of provision, and supplementary information listed in Table 1806, can be displayed below it. Note that the individual specific means, purpose, form of provision, and supplementary information in the "Means, Purpose, Form of Provision, and Supplementary Information" column of Table 1804 may be displayed in a manner that distinguishes them from other strings, allowing them to be selected as means, purpose, form of provision, and supplementary information to be pasted into text box 2006. The same applies to Tables 1805 and 1806. Examples of distinguishable text include bold or colored fonts, or text with markers.

[0193] The screen shown in Figure 20 displays the three solutions included in solution 1807. Similarly, the screen shown in Figure 21 displays the nine solutions, which are the three solutions 2101-2103 included in solution 2003 and the three solutions 2104-2106 included in solution 2004. Users can toggle the display / hide of the contents of these nine solutions 2008-2010 and 2101-2106 by selecting the corresponding hide icon or operating the open / close screen controls. For example, Figure 22 shows the display of only one of these solutions, 2008, titled "Ergonomic design cushion: Maintaining a comfortable posture for extended periods; Home cushion product focusing on ergonomics as a standard design."

[0194] Referring to Figure 23, the Step 4 screen has an "Execute" button 2301 that allows the LLM to generate a concept and then evaluate the generated concept.

[0195] When the user presses the "Execute" button 2301, the single-output LLM utilization unit 923 is activated. The single-output LLM utilization unit 923 causes the LLM to generate concepts and evaluations based on the content of the necessary items from (a) to (g) below.

[0196] (a) The industry of the company that conducts business development is "cushion manufacturer" as held in the holding section 901. (b) A strength or asset that could be utilized in business development is the "ability to design, manufacture, and sell cushions of new shapes and materials" held in the holding unit 901. (c) As a business development and innovation policy and strategy, the statement held in the holding unit 901 is "Design, manufacture, and sell new donut-shaped cushions." (d) The markets to be targeted for business development are "offices," "homes," "nursing homes," and "hospitals" held in the holding unit 901. (e) The "office cushion market" held in the holding unit 907 as a favorable segment, and the challenges in the "office cushion market" (f) The data set related to the issue is stored in the holding unit 914 as follows: "With the spread of remote work, the amount of work done at home is increasing. People want to maintain a comfortable posture while working. Business people working from home. Office chairs and regular chairs. Sitting for long periods of time puts a strain on the body." (g) As a solution, the "Ergonomic design cushion that maintains a comfortable posture for a long time, household cushion product that focuses on ergonomics as a standard proposal" is held in the retaining part 921.

[0197] The required items are, for example, all items (a) through (g). However, the configuration of this support device 900 is such that it is possible to switch the selection / deselection of each item (a) through (g).

[0198] As shown in Figure 22(a), it is assumed that on the Step 3 screen, only the solution indicated by the string 2008, "Ergonomic design cushion: Maintains a comfortable posture for extended periods; Home cushion product: Focusing on ergonomics as a standard proposal," is displayed, and that the selection unit 922 has selected only the solution indicated by the string 2008, "Ergonomic design cushion: Maintains a comfortable posture for extended periods; Home cushion product: Focusing on ergonomics as a standard proposal."

[0199] The single-output LLM utilization unit 923 displays the concept and evaluation 2302 based on the concept and evaluation information generated by the LLM.

[0200] As shown in Figure 23, the concept and evaluation 2302 includes a title 2303, summary 2304, table 2305, overall assessment 2306, and trace log 2307. The contents are as illustrated in Figure 23. Briefly, the summary includes the above-mentioned problem (f) and solution (g). Table 2305 contains five evaluation items. For each item in Table 2305, the item name, concept, evaluation criteria, score, and question are described.

[0201] Users can toggle the display of the Concept and Evaluation 2302 content by clicking the corresponding icon 1006E (see Figure 23) or the open / close screen control 2007 (see Figure 23). When the Concept and Evaluation 2302 content is hidden, only the title of Concept and Evaluation 2302 is displayed on the screen. In the screen shown in Figure 24, only the title of Concept and Evaluation 2302B is displayed.

[0202] The screen shown in Figure 24 also displays concepts and evaluations 2401-2408, which consist only of titles. The i-th concept and evaluation among the nine concepts and evaluations shown in Figure 24 are the concepts and evaluations generated when the i-th set of strings from the nine sets of strings representing means of delivery, purpose, form of delivery, and supplementary information shown in Figure 21 is selected.

[0203] Users can toggle the display / hide of the content for each of the nine concepts and evaluations by operating the screen controls 2008 and 2409-2416 for opening and closing.

[0204] Referring to Figures 24 and 25, the Step 4 screen has an "Execute" button 2418 to prompt the LLM to answer questions, below a results list 2417 that includes Concepts and Evaluations 2302 (2302B), 2401-2408. The questions here are the five questions in the Questions column of Table 2305 (see Figure 23) included in any of the Concepts and Evaluations.

[0205] When the user presses the "Execute" button 2418, the single-output LLM utilization unit 926 is activated. The single-output LLM utilization unit 926 generates a response based on the content of the required items from the following items (a) to (h).

[0206] (a) The industry of the company that conducts business development is "cushion manufacturer" as held in the holding section 901. (b) A strength or asset that could be utilized in business development is the "ability to design, manufacture, and sell cushions of new shapes and materials" held in the holding unit 901. (c) As a business development and innovation policy and strategy, the statement held in the holding unit 901 is "Design, manufacture, and sell new donut-shaped cushions." (d) The markets to be targeted for business development are "offices," "homes," "nursing homes," and "hospitals" held in the holding unit 901. (e) The "office cushion market" held in the holding unit 907 as a favorable segment, and the challenges in the "office cushion market" (f) The data set related to the issue is stored in the holding unit 914 as follows: "With the spread of remote work, the amount of work done at home is increasing. People want to maintain a comfortable posture while working. Business people working from home. Office chairs and regular chairs. Sitting for long periods of time puts a strain on the body." (g) As a solution, the "Ergonomic design cushion that maintains a comfortable posture for a long time, household cushion product that focuses on ergonomics as a standard proposal" is held in the retaining part 921. (h) Information held in the holding unit 924 as a concept and question

[0207] Note that this assumes that the first concept and evaluation content are displayed on the Step 4 screen, but the second through ninth concepts and evaluation content are not displayed.

[0208] The required items are, for example, items (a) through (e) and (h). However, the configuration of this support device 900 is such that it is possible to switch the selection / deselection of each item (a) through (h).

[0209] The single-output LLM utilization unit 926 displays answer 2419B based on the information regarding the answer generated by the LLM.

[0210] As shown in Figure 25, response 2419B includes title 2501, Q&A 2502, and trace log 2503. The contents are as illustrated in Figure 25. Briefly, the Q&A contains two questions for each of the five evaluation items included in concept table 2305, and the answers to each question.

[0211] The user can toggle the display of the content of answer 2419B to the question by clicking the corresponding icon 1006F (see Figure 25) or the open / close screen control 2504 (see Figure 25). When the content of answer 2419B is hidden, only the title of answer 2419B is displayed on the screen. In the screen shown in Figure 24, only the title of the concept and evaluation 2419 are displayed. In the screen shown in Figure 26, only the title of the concept and evaluation 2419 are displayed.

[0212] The screen shown in Figure 26 also displays responses 2601-2608, which consist only of titles. The i-th response among the nine responses shown in Figure 26 corresponds to the i-th concept and evaluation among the nine concepts and evaluations shown in Figure 24.

[0213] The user can toggle the display / hide of the content of each of the nine answers by operating the screen controls 2504 and 2609-2616 for opening and closing.

[0214] Referring to Figure 27, the Step 5 screen has an "Execute" button 2701 to cause the LLM to generate a business plan.

[0215] When the user presses the "Execute" button 2701, the single-output LLM utilization unit 929 is activated. The single-output LLM utilization unit 929 causes the LLM to generate a business plan based on the content of the necessary items from the following items (a) to (i).

[0216] (a) The industry of the company that conducts business development is "cushion manufacturer" as held in the holding section 901. (b) A strength or asset that could be utilized in business development is the "ability to design, manufacture, and sell cushions of new shapes and materials" held in the holding unit 901. (c) As a business development and innovation policy and strategy, the statement held in the holding unit 901 is "Design, manufacture, and sell new donut-shaped cushions." (d) The markets to be targeted for business development are "offices," "homes," "nursing homes," and "hospitals" held in the holding unit 901. (e) The "office cushion market" held in the holding unit 907 as a favorable segment, and the challenges in the "office cushion market" (f) The data set related to the issue is stored in the holding unit 914 as follows: "With the spread of remote work, the amount of work done at home is increasing. People want to maintain a comfortable posture while working. Business people working from home. Office chairs and regular chairs. Sitting for long periods of time puts a strain on the body." (g) As a solution, the "Ergonomic design cushion that maintains a comfortable posture for a long time, household cushion product that focuses on ergonomics as a standard proposal" is held in the retaining part 921. (h) Information held in the holding unit 924 as a concept and question (i) Information held in the holding unit 927 as the answer

[0217] The required items are, for example, items (a) through (c), (e), and (g) through (i). However, the configuration of this support device 900 is such that it is possible to switch the selection / deselection of each item (a) through (i).

[0218] Note that this assumes that the content of the first answer is displayed on the Step 4 screen, but the content of the second through ninth answers is not displayed.

[0219] The single-output LLM utilization unit 929 displays the business plan 2702 based on the information about the business plan generated by the LLM.

[0220] As shown in Figure 27, the business concept plan 2702 includes a title 2703, an outline 2704, content 2705, and a trace log 2710. A detailed explanation of content 2705 is omitted, but in short, content 2705 is divided into content for each of the five evaluation items included in the concept table 2305 (target customers and issues 2705, business concept 2706, solution 2707, market competitiveness 2708, and worldview 2709).

[0221] The user can toggle the display of the content of the answer 2419B to the question by clicking the corresponding icon 1006G (see Figure 27) or by clicking the open / close screen control 2712 (see Figure 27). When the content of the business plan 2702 is hidden, only the title of the business plan 2702 is displayed on the screen. In the screen shown in Figure 28, only the title of the business plan 2702B is displayed.

[0222] The screen shown in Figure 28 also displays business plan proposals 2801-2808, which only contain titles. The i-th answer among the nine business plan proposals shown in Figure 28 corresponds to the answer to the i-th question among the answers to the nine questions shown in Figure 26.

[0223] Users can toggle the display / hide of the contents of each of the nine business plan documents by operating the open / close screen controls 2809-2817.

[0224] <About prompts>

[0225] The prompts that the multiple output LLM user unit 902 gives to the LLM include the following description: • Description regarding the fact that the input items are items (a) to (d) held in the holding unit 901. • Description regarding LLM being an industry professional This instruction instructs the company to identify potential segments for the market it wishes to enter, and to identify potential market segments within that market, as well as potential market segments that the company should enter. This instruction instructs the company to identify the top five market segments it should enter, based on its strengths, assets, and strategy. This instruction instructs you to identify the top three potential market segments with a large number of potential customers and likely to have a need, based on the external environment of the industry and recent trends. • Instructions to output trace logs. • Description of output format

[0226] The prompt that the multiple output LLM user unit 909 gives to the LLM includes the following description: • Description regarding the fact that the input items are items (a) and (d) held in the holding unit 901 and item (e) selected by the selection unit 908. • Description regarding LLM being an industry professional • A statement regarding the LLM's deep experience and knowledge of the market and segments. This description instructs the system to identify the specific situations, backgrounds, and contexts in which the target customers of a particular market or segment find themselves, and to present them in a specified output format, along with the goals they wish to achieve and the challenges they wish to solve at that time. • Instructions to output trace logs. • Description of output format

[0227] The prompts that the multiple output LLM utilization unit 916 gives to the LLM include the following description: • Description regarding the fact that the input items are items (a) to (d) held in the holding unit 901, item (e) selected by the selection unit 908, and item (f) selected by the selection unit 915. • Instructions for creating three solutions • Description specifying the pattern that generates each solution. This description instructs the system to identify the specific situations, backgrounds, and contexts in which the target customers of a particular market or segment find themselves, and to present them in a specified output format, along with the goals they wish to achieve and the challenges they wish to solve at that time. • Instructions to output trace logs. • Description of output format

[0228] The prompts that the multiple output LLM user unit 923 gives to the LLM include the following description: ·A statement that the input items are items (a) to (d) held in holding unit 901, item (e) selected by selecting unit 908, item (f) selected by selecting unit 915, and item (g) selected by selecting unit 922 ·A statement that the LLM is an industry professional ·A statement instructing creation of a concept proposal ·A statement instructing evaluation of a concept proposal and creation of questions ·A statement instructing output of a trace log ·A statement regarding an output format

[0229] The prompt provided to the LLM by the multiple output LLM utilization unit 926 includes the following descriptions. ·A statement that the input items are items (a) to (d) held in holding unit 901, item (e) selected by selecting unit 908, and item (h) selected by selecting unit 925 ·A statement that the LLM is an industry professional ·A statement instructing creation of answers to questions generated by evaluating a concept proposal ·A statement instructing output of a trace log ·A statement regarding an output format

[0230] The prompt provided to the LLM by the multiple output LLM utilization unit 929 includes the following descriptions. ·A statement that the input items are items (a) to (c) held in holding unit 901, item (e) selected by selecting unit 908, item (g) selected by selecting unit 922, item (h) selected by selecting unit 925, and item (i) selected by selecting unit 928 ·A statement that the LLM is an industry professional ·A statement instructing creation of a new business project proposal ·A statement instructing output of a trace log ·A statement regarding an output format

[0231] <Regarding Trace Logs>

[0232] The trace log enables a user to understand the positioning of the current step based on its relative relationship with other steps.

[0233] The trace log includes a description of which step's data set the LLM utilization unit of the current step supplies to the LLM.

[0234] The trace log includes a description of whether the current step is a divergence step, a convergence & evaluation step, or a convergence step.

[0235] When the current step is a divergence step, the trace log includes a description of in which step the data set generated in the current step is selected. The trace log also includes a description of in which subsequent step the step selected in the step of selecting the data set generated in the current step is processed.

[0236] When the current step is a convergence & evaluation step or a convergence step, the trace log includes a description of in which step the data set generated in the current step is examined. The trace log also includes a description of in which subsequent step the examination result generated in the step of examining the data set generated in the current step is processed.

[0237] For example, the input to the trace log 1505 of step 2-1 (Figure 15) is "0-1 / 0-2 / 0-3", which indicates that the multiple output LLM utilization unit 909 corresponding to step 2-1 supplies the items output from these three steps to the LLM. Also, the trace log 1505 of step 2-1 (Figure 15) indicates that step 2-1 is a "divergence" step. Furthermore, the trace log 1505 of step 2-1 indicates that one of the multiple outputs from step 2-1 is selected in step 2-2. Furthermore, the trace log 1505 of step 2-1 indicates that the item selected in step 2-2 is used in steps 3-1, 4-1, and 5-1.

[0238] For example, the input to the trace log 2307 of step 4-1 (Figure 23) is "0-1 / 0-2 / 0-3 / 1-1 / 1-3 / 2-2 / 3-2", which indicates that the single-output LLM utilization unit corresponding to step 4-1 supplies the items output from these seven steps to the LLM. The trace log 2307 of step 4-1 also indicates that step 4-1 is a "convergence & evaluation" step. Furthermore, the trace log 2307 of step 4-1 indicates that the output from step 4-1 is selected in step 4-3 (not shown). Finally, the trace log of step 4-1 indicates that the items selected in step 4-3 are used in step 5-1.

[0239] <About Loops>

[0240] The user can freely select the steps to be executed by the support device by selecting a tab located at the top of the screen (for example, the tab with the text "1 Market Selection"). Basically, the user has the support device 900 execute steps 0 through 5 sequentially. However, the user can also have the support device 900 repeatedly execute a specific range of steps. For example, the user can repeatedly operate a predetermined range from the multiple output LLM utilization unit 909 onwards while switching the dataset selected by the selection unit 908. Any part can be selected as the end of the predetermined range. The operation can be repeated up to the holding unit 914, with multiple holding units 914 holding intermediate results, or up to the holding unit 921, with multiple holding units 921 holding intermediate results, or the range up to the holding unit 930 can be repeated. Furthermore, any part can be selected as the starting point for the repeated operation.

[0241] <Regarding input by third parties> As shown in Figure 10, a pair of an icon and text (1015) for inviting participants is displayed in the upper right corner of the screen. When a user selects this pair, a screen (not shown) for the user to invite a third party is displayed. This screen displays invitation information, such as a URL to access this screen. When a user invites a third party using the invitation information, that third party will be able to view, operate, and input text on the screen. In other words, the third party will be able to do the same things as the user. Using this function, the third party can input text into text boxes 1001, 1008, 1011, 1014, 1101, 1113, 1506, and 20006. Therefore, in addition to the answers from the LLM and the user, answers from third parties can also be used.

[0242] [Other embodiments] The configuration described in the second embodiment is merely an example. In other words, for example, the use of three multi-output LLM utilization units and three single-output LLM utilization units as LLM utilization units is merely an example. In general, the number of these units may be changed or their order may be changed depending on the application. The number of parallel holding units is also arbitrary. In particular, if a list is used, it does not need to be predetermined. Furthermore, the assignment of steps to each screen is flexible. For example, the steps in the multiple screens of the above embodiment may be consolidated into a single screen. Also, the prompts may be changed while maintaining the configuration, depending on the application. Depending on the application, the configuration may be changed and new prompts may be applied.

[0243] Within the scope of the concept of this invention, those skilled in the art can conceive of various modifications and alterations. Therefore, such modifications and alterations are understood to fall within the scope of this invention. For example, any addition, deletion, or design change of components, or addition, omission, or modification of processes, made by a person skilled in the art to the above-described embodiments, is also included within the scope of this invention, as long as it retains the essence of this invention. [Explanation of symbols]

[0244] S System 1 Support equipment 11 Control Unit 111 Basic Data Acquisition Unit 112 Input screen display section 113 Department for obtaining examination results 114 Candidate Generation Unit for Review Results 115 Information Provision Department 116 Department for Obtaining Results of Collaborator Review 117 Material generation department 118 Evaluation Generation Unit 119 Comparison display command section 13 Storage section 14 Communications Department N Network T terminal E Input screen A. Evaluation screen

Claims

1. A support device for assisting users in creating documents using multiple steps, The system includes a display control means that displays a screen corresponding to any one of the steps of interest included in the aforementioned plurality of steps, The aforementioned screen is A startup component for activating a large-scale language model utilization means to generate a string containing one or more answers to the questions asked to the user in the aforementioned focus step, based on user operations, A display area for displaying a string generated by the aforementioned large-scale language model, which includes one or more of the aforementioned answers, A text box for inputting individual responses from the user or the large-scale language model to the questions asked to the user in the aforementioned focus step, A list for storing the individual answers entered in the text boxes, A switching component for enabling / disabling the individual answers held in the list based on user operations, Equipped with, Based on user actions, it is possible to copy the individual answers displayed in the display area into the text box. Support equipment.

2. The large-scale language model utilization means supplies the large-scale language model with the answers in the preceding steps, which are activated by the switching component provided on the screen corresponding to at least one preceding step for the step of interest, in order to generate the answers in the step of interest. The support device according to claim 1.

3. The screen includes an area that lists the preceding steps to which the answer was supplied in order to generate the answer in the focus step. The support device according to claim 2.

4. The screen corresponding to at least one preceding step to the step of interest is the same as or different from the screen for the step of interest. The support device according to claim 2.

5. The response in the focus step, activated by the switching component, is supplied to the large language model by the large language model utilization means, which is activated by an activation component provided on a screen corresponding to at least one subsequent step to the focus step. The support device according to claim 1.

6. The aforementioned screen includes an area that lists the subsequent steps in which the answer in the focus step is supplied. The support device according to claim 5.

7. The screen corresponding to at least one subsequent step to the aforementioned focus step is the same as or different from the screen for the aforementioned focus step. The support device according to claim 5.

8. The system includes a second display control means that displays a second screen corresponding to any one of the other second steps included in the aforementioned plurality of steps, The second screen mentioned above is, A second activation component for activating a second large-scale language model utilization means for generating a single answer to a question posed to the user in the second step based on user operations, A second list for collectively storing one or more second responses generated by the large-scale language model when the second large-scale language model utilization means is activated one or more times based on user operations, A second switching component for enabling / disabling the one or more second answers held in the second list based on user operation, Equipped with, The support device according to claim 1.

9. The second large-scale language model utilization means supplies the large-scale language model with the answer in the second preceding step, which is activated by the switching component provided on the screen corresponding to at least one second preceding step for the second target step, in order to generate the answer in the second target step. The support device according to claim 8.

10. The screen corresponding to at least one second preceding step for the second step is the same as or different from the screen for the current second step. The support device according to claim 9.

11. The response in the second focus step, activated by the switching component, is supplied to the large language model by the large language model utilization means, which is activated by an activation component provided on a screen corresponding to at least one second subsequent step to the second focus step. The support device according to claim 8.

12. The screen corresponding to at least one second subsequent step for the second step is identical to or different from the screen for the second step. The support device according to claim 11.

13. Based on user actions, the user can show or hide individual answers or all answers within the user display area. The support device according to claim 1.

14. Each of the aforementioned answers held in the aforementioned list will be displayed if enabled, and not displayed if disabled. The support device according to claim 1.

15. Each of the answers held in the aforementioned list is displayed on a foldable card, with the card unfolded when activated and folded when deactivated. The support device according to claim 1.

16. The aforementioned screen displays content indicating that the current screen is the aforementioned screen, The second screen displays content that indicates that the current screen is the second screen. The support device according to claim 8.

17. Each screen has tabs, and when a user selects a tab, the screen containing the selected tab is displayed. The support device according to claim 1.

18. In the display area, the string is displayed in such a way that it can be distinguished between the one or more answers. The support device according to claim 1.

19. The text box can also be used to input responses from a third party who has been invited by the user. The support device according to claim 1.

20. A program for causing a computer to function as an auxiliary device according to any one of claims 1 to 19.

21. A computer-readable recording medium storing a program for causing a computer to function as an auxiliary device according to any one of claims 1 to 19.

22. A support method for assisting users in creating documents using multiple steps, The system includes a display control step that displays a screen corresponding to any one of the steps of interest included in the aforementioned plurality of steps, In the display control step, the screen displays: A startup component for activating a large-scale language model utilization means to generate a string containing one or more answers to the questions asked to the user in the aforementioned focus step, based on user operations, A display area for displaying a string generated by the aforementioned large-scale language model, which includes one or more of the aforementioned answers, A text box for inputting individual responses from the user or the large-scale language model to the questions asked to the user in the aforementioned focus step, A list for storing the individual answers entered in the text boxes, A switching component for enabling / disabling the individual answers held in the list based on user operations, Display and Based on user actions, it is possible to copy the individual answers displayed in the display area into the text box. How to help.

Images