JP7591175B1 - Information processing system, information processing method, and program - Google Patents

Abstract

[Problem] To provide a mechanism for using a generation AI to assist in the creation of a document for evaluating the business viability of a business. [Solution] An information processing system for assisting in the creation of a document for evaluating the business viability of a business, the information processing system comprising: a first reception means for receiving the designation of a business, a first acquisition means for instructing the generation AI to output input information for input items of a document for evaluating the business viability of the designated business and acquiring the outputted input information, a second acquisition means for instructing the generation AI to output a question for acquiring missing information from the designated business about the acquired input information and acquiring the output question, and a first output means for outputting the acquired question. [Selected Figure] Figure 5

Description

The present invention relates to an information processing system, an information processing method, and a program.

Non-Patent Document 1 is a document disclosing background technology in this technical field. Non-Patent Document 1 states, "The system supports activities in accordance with the Regional Business Support Package (Cabinet Office) and the Basic Policy for Financial Monitoring (Financial Services Agency). It can be used to strengthen the ability to appropriately evaluate the business content and growth potential of companies at various life stages."
Furthermore, this non-patent document 1 states, "Business feasibility evaluation sheets can be efficiently created. By importing financial data, attribute information, etc., the administrative burden of assisting in the creation of business overviews can be reduced."

Information Planning Co., Ltd., "Business Evaluation Support System", [online], [Retrieved July 19, 2024], Internet <URL: https://www.jyohokikaku.co.jp/loan-solution/129.html>

The above-mentioned non-patent document 1 describes that, with regard to the creation of a business feasibility evaluation sheet, the administrative burden of supporting the creation of a business overview map, etc. is reduced by importing financial data, attribute information, etc. However, non-patent document 1 does not mention the use of generation AI.
The present invention has been made in consideration of these circumstances, and provides a mechanism that uses generation AI to assist in the creation of documents for evaluating the business viability of businesses.

In order to solve the above problems, for example, the configurations described in the claims are adopted .

According to the present invention, a mechanism can be provided that utilizes generation AI to assist in the creation of documents for evaluating the business viability of a company.
Problems, configurations and effects other than those described above will become apparent from the following description of the embodiments.

FIG. 1 shows an example of the configuration of an information processing system 100 . FIG. 2 shows an example of the configuration of the business support server 101 . FIG. 3 shows an example of the configuration of the LLM server 102 . FIG. 4 shows an example of the configuration of the user terminal 103 . FIG. 5 shows an example of a business support process 500 . FIG. 6 shows an example of a business designation screen 600 . FIG. 7 shows an example of input information 700 corresponding to the input item "company overview." FIG. 8 shows an example of input information 800 corresponding to the input item "commercial flow." FIG. 9 shows an example of input information 900 corresponding to the input item "market." FIG. 10 shows an example of input information 1000 corresponding to the input item "3C analysis." FIG. 11 shows an example of input information 1100 corresponding to the input item "Five Forces Analysis." FIG. 12 shows an example of input information 1200 corresponding to the input item "SWOT analysis." FIG. 13 shows an example of input information 1300 corresponding to the input item "management system." FIG. 14 shows an example of input information 1400 corresponding to the input item "financial summary." FIG. 15 shows an example of input information 1500 corresponding to the input item "management issue." FIG. 16 shows an example of input information 1600 corresponding to the input item "management policy." FIG. 17 shows an example of a question list 1700 related to the input item "Five Forces Analysis." FIG. 18 shows an example of a question list 1800 related to the input item "SWOT analysis." FIG. 19 shows an example of the configuration of the business support server 1900 . FIG. 20 shows an example of a business support process 2000 . FIG. 21 shows an example of a business support process 2000 . FIG. 22 shows an example of a question list 2200 related to the input item "Five Forces Analysis." FIG. 23 shows an example of a business support process 2300 .

1. Examples Hereinafter, examples of the present invention will be described with reference to the drawings.

1-1. Configuration Fig. 1 shows an example of the configuration of an information processing system 100 according to this embodiment.
The information processing system 100 includes a business support server 101, an LLM server 102, and a plurality of user terminals 103. The information processing system 100 is a system for supporting the creation of documents for evaluating the business feasibility of an enterprise. Each device constituting the system is connected via a wired or wireless network and is capable of transmitting and receiving information to and from each other.

Of the devices constituting this system, the business support server 101 is a server that supports the user in creating a business feasibility evaluation sheet.
The LLM server 102 is a server that receives a prompt from the business support server 101 and outputs a response corresponding to the received prompt.
The plurality of user terminals 103 are terminal devices used by respective users. Each user creates a business feasibility evaluation sheet by using the business support server 101. In this embodiment, a bank employee is assumed as an example of a user.

Each of these devices includes a processor that executes an operating system, applications, programs, etc.; a main storage device such as a RAM (Random Access Memory); an auxiliary storage device such as an IC card, a hard disk drive, an SSD (Solid State Drive), or a flash memory; a communication control unit such as a network card, a wireless communication module, or a mobile communication module; input devices such as a touch panel, keyboard, mouse, voice input, and input based on motion detection captured by a camera unit; and an output device such as a monitor or display. The output device may be a device or terminal that transmits information to be output to an external monitor, display, printer, or device.

The main memory stores various programs and applications (modules), and the processor executes these programs and applications to realize each functional element of the overall system. Each module may be implemented in hardware, for example by integration. Each module may be an independent program or application, or may be implemented as a subprogram or function within a single integrated program or application.

In this specification, each module is described as the entity (subject) that performs the processing, but in reality, the processing is carried out by a processor that processes various programs, applications, etc. (modules).

Various databases (DB) are stored in the auxiliary storage device. A "database" is a functional element (storage unit) that stores a set of data so that it can handle any data manipulation (e.g., extraction, addition, deletion, overwriting, etc.) from a processor or an external computer. There are no limitations on how the database is implemented, and it may be, for example, a database management system, spreadsheet software, or a text file such as XML or JSON.

1-1-1. Business support server 101
FIG. 2 shows an example of the configuration of the business support server 101 .
The business support server 101 is composed of, for example, one or more servers arranged on a cloud.
The server's main memory device 201 stores programs and applications such as a first receiving module 210, a first acquisition module 211, a second acquisition module 212, and a first output module 213. The processor 203 executes these programs and applications to realize each functional element of the business support server 101. Each module will be described below.

The first reception module 210 receives the designation of a business from the user.

The first acquisition module 211 instructs the generation AI to output input information for input items of a document for evaluating the business feasibility of a business entity specified via the first reception module 210. Then, the module acquires the input information output from the generation AI.
In this case, the module can instruct the generation AI to output input information for the input fields for the specified business operator based on at least one of the accounting information and financial statement data of the specified business operator.

The document for evaluating business feasibility referred to here is, specifically, for example, a business feasibility evaluation sheet. This business feasibility evaluation sheet is mainly prepared by financial institutions and is used to determine whether or not to grant a loan. Note that although the name "business feasibility evaluation sheet" is used here, documents with different names can be processed in the same way as in this book, so long as they are documents for evaluating business feasibility.
The input items referred to here are specifically the input items of the business feasibility evaluation sheet, which may include at least one of the following: company overview, trade flow, market, 3C analysis, 5 forces analysis, SWOT analysis, management structure, financial summary, management issues, and management measures.
Furthermore, the generating AI referred to here specifically refers to the generating AI module 310 described later.

Next, the second acquisition module 212 instructs the generation AI to output a question for acquiring missing information from a specified business entity for the input information acquired by the first acquisition module 211. Then, the module acquires the question output from the generation AI.
At that time, the module instructs the generation AI to identify the missing information based on the input information for the above input items previously output by the generation AI.

The input information for the above input items previously output by the generation AI is, specifically, model input information. This input information is, for example, input information generated by a veteran bank employee using the generation AI, or input information created by the veteran bank employee himself.
Furthermore, the generating AI referred to here specifically refers to the generating AI module 310 described later.

Next, the first output module 213 outputs the question acquired by the second acquisition module 212. Specifically, this module displays the acquired question on the display of the user terminal 103.

Next, the auxiliary storage device 202 will be described.
The auxiliary storage device 202 stores an accounting information DB 220, a company information DB 221, a financial statement DB 222, a public relations memo DB 223, a corporate statistics DB 224, an industry-specific examination dictionary DB 225, an examination manual DB 226, an industry trend information DB 227, and an input item table 228. Each DB will be described below.

The accounting information DB220 is a database that stores the accounting information of each business. The accounting information stored in this DB includes information on the balance and deposit/withdrawal details of one or more accounts of the business, and information on invoices issued or received by the business. The accounting information of each business is stored in association with a business ID.

The company information DB221 is a database that stores the company information of each business operator. The company information stored in this DB includes the company's name, location, telephone number, representative's name, capital, number of employees, industry ID, business type, business content, products/services, history, bases, major business partners, shareholders, corporate philosophy, and future vision. The company information of each business operator is stored in association with the business operator ID.

The financial statement DB222 is a database that stores financial statement data for each business. The financial statement data stored in this DB includes balance sheets, profit and loss statements, and cash flow statements. The financial statement data for each business is stored in association with a business ID.

The public relations memo DB223 is a database that stores the public relations memo of each business. The public relations memo stored in this DB includes the history of negotiations between the business and the bank. This negotiation history includes documents, emails, and conversations exchanged between the two parties. The public relations memo of each business is stored in association with the business ID.

The corporate statistics DB224 is a database that stores corporate statistics for each industry. The corporate statistics stored in this DB include the results of the corporate statistics survey conducted by the Ministry of Finance. These survey results include sales, operating profits, ordinary profits, and capital investments. Corporate statistics for each industry are stored in association with an industry ID.

The Industry Review Dictionary DB225 is a database that stores industry information for each industry. The industry information stored in this DB includes information from the Industry Review Dictionary published by the Institute for Financial Affairs, a general incorporated association. This information includes industry characteristics, market size, regional characteristics, demand trends, issues and prospects, product knowledge, production methods, trading methods, capital needs, how to read financial statements, key points for management support, relevant laws and regulations, institutional loans, and industry associations. The industry information for each industry is stored in association with an industry ID.

The screening manual DB226 is a database that stores loan screening manuals for each industry. The loan screening manuals stored in this DB include the loan screening manuals of each financial institution and the viewpoints of industry-specific support issued by the Financial Services Agency. The loan screening manuals for each industry are stored in association with the industry ID.

The industry trend information DB227 is a database that stores trend information for each industry. The trend information stored in this DB includes industry trend information issued by government agencies or private research companies. The trend information for each industry is stored in association with an industry ID.

The input item table 228 is a table for managing the correspondence between the input items of the business feasibility evaluation sheet and the DBs. This table is used to identify the DBs corresponding to each input item.
In addition, since the format (including input items) of the business feasibility assessment sheet usually differs for each financial institution, it is conceivable that the correspondence between the input items and the DB may differ for each financial institution. Therefore, before starting the business support process 500 described later, the business support server 101 sets the format of the business feasibility assessment sheet for the financial institution and the correspondence between the input items and the DB (i.e., the input item table 228).
In this table, for example, the input item "company overview" is associated with an accounting information DB 220 and a business information DB 221. As another example, the input item "SWOT analysis" is associated with the business information DB 221, financial statements DB 222, public relations memo DB 223, corporate statistics DB 224, and industry-specific review dictionary DB 225.

1-1-2. LLM server 102
FIG. 3 shows an example of the configuration of the LLM server 102 .
The LLM server 102 is composed of, for example, one or more servers arranged on a cloud.
The server's main memory 301 stores programs and applications such as a generation AI module 310. The processor 303 executes these programs and applications to realize each functional element of the LLM server 102.

Among the functional elements realized, the generative AI module 310 is a deep learning model trained in advance on a large-scale data set. In other words, this module is an LLM (i.e., a large-scale language model). This module outputs a response to an input prompt (in other words, a command sentence).
This module is a deep learning model that has already been trained using a large dataset, so it can be used without training data or additional training.

1-1-3. User terminal 103
FIG. 4 shows an example of the configuration of the user terminal 103 .
The user terminal 103 is, for example, a terminal device such as a smartphone, a tablet terminal, a notebook PC, a desktop PC, etc. In this embodiment, a bank employee is assumed as an example of a user.

The main memory device 401 of this terminal stores programs and applications such as a browser module 410. The processor 403 executes these programs and applications to realize each functional element of the user terminal 103.

The browser module 410 exchanges information with the business support server 101.

1-2. Operation Next, the business support process 500 executed by the business support server 101 will be described with reference to Fig. 5. Fig. 5 is a flow diagram showing an example of the business support process 500. The process shown in the figure supports the creation of a business feasibility evaluation sheet.
As described above, the format of the business feasibility evaluation sheet for the financial institution and the input item table 228 are set in the business support server 101 before the business support process 500 described below is started.

First, the first reception module 210 receives a request for the business operator selection screen 600 from the user terminal 103 (step 501). In response to this request, the first reception module 210 causes the user terminal 103 to display the business operator selection screen 600 (step 502).

6 shows an example of a business designation screen 600. The screen shown in the figure includes a business name input field 601, a website information input field 602, and a creation instruction button 603.
Of these display elements, the company name input field 601 is used to input the name of the business for which a business feasibility evaluation sheet is to be created.
The website information input field 602 is used to input the website address (URL) of the business entity for which the business feasibility evaluation sheet is to be created.
The creation instruction button 603 is a button for instructing the creation of a business feasibility evaluation sheet.
6 includes only the company name input field 601, website information input field 602, and creation instruction button 603, but is not limited to this. For example, it is also possible to input an ID in a database of company information held by the bank (ID managed by CRM: Customer Relationship Management) as information for identifying the company. In that case, if information such as company name and website address is managed in the database of company information, it is not necessarily necessary to provide the company name input field 601 and website information input field 602.

The user inputs the business name and website address on the screen, and selects the creation instruction button 603. As a result, the input information is sent from the user terminal 103 to the business support server 101, and the first receiving module 210 acquires the input information (step 503).
After acquiring the input information, the first receiving module 210 searches the company information DB 221 to identify the company ID and industry ID corresponding to the acquired company name (step 504).

Next, the first acquisition module 211 selects input items to be processed from among the input items of the business feasibility evaluation sheet to be created (step 505). The business feasibility evaluation sheet that is finally created includes a total of 10 input items, such as company overview, distribution channel, market, 3C analysis, 5 forces analysis, SWOT analysis, management structure, financial summary, management issues, and management measures.

Next, the first acquisition module 211 refers to the input item table 228 to identify the DB corresponding to the selected input item (step 506). The module then refers to the identified DB to acquire information on the target business and information on the industry to which the target business belongs (step 507). In so doing, the module acquires information corresponding to the business ID or industry ID identified in step 504 from each DB. Hereinafter, the information acquired in this step will be referred to as "reference information."

Next, the first acquisition module 211 generates a prompt to be input to the generation AI module 310 (step 508). The generated prompt (in other words, a command sentence) includes the following information:
(1) The website address of the business acquired in step 503; (2) Detailed items constituting the input items to be processed; (3) Reference information acquired in step 507; (4) A statement commanding the output of input information corresponding to the detailed items in (2) for the business designated by the website address in (1) with reference to the reference information in (3).

In addition, the detailed items in (2) are, for example, in the case of the input item "Company Overview," the company name, representative, industry, business content, base, number of employees, business history, and history (see Figure 7).

Next, the first acquisition module 211 inputs the generated prompt to the generation AI module 310 (step 509). The generation AI module 310 outputs an answer corresponding to the input prompt. The output answer (in other words, the input information) is acquired by the first acquisition module 211 (step 510). The first acquisition module 211 inputs the acquired input information into the corresponding input item field of the business feasibility evaluation sheet (step 511).

Next, the second acquisition module 212 receives the output of the input information and generates a prompt to be input to the generation AI module 310 (step 512). The prompt generated here includes the following information:
(1) Input information acquired in step 510; (2) Model input information; (3) A command to output a question for acquiring information that is missing in the input information in (1) from the target business entity in light of the input information in (2).

The input information in (2) refers to one or more pieces of input information for the input items to be processed that have been previously output by the generation AI module 310. This input information is, for example, input information generated by a veteran bank employee using the generation AI module 310.

Next, the second acquisition module 212 inputs the generated prompt to the generation AI module 310 (step 513). The generation AI module 310 outputs an answer corresponding to the input prompt. The output answer is acquired by the second acquisition module 212 (step 514). If the acquired answer includes a question, the second acquisition module 212 inputs the question into the corresponding input item field of the question sheet (step 515).

Next, the first acquisition module 211 determines whether all input items have been processed (step 516). If the result of this determination is that all input items have not been processed (NO in step 516), the module returns to step 505 and selects the next input item. On the other hand, if the result of this determination is that all input items have been processed (YES in step 516), the first output module 213 displays the generated business feasibility evaluation sheet and questionnaire sheet on the user terminal 103 (step 517).

The user completes the displayed business feasibility evaluation sheet by modifying it as necessary. In doing so, the user can use the displayed questionnaire to find out any missing information from the target business. The user can then complete the business feasibility evaluation sheet by completing the information found out.

In the above example, steps 505 to 515 are performed for all input items, but this is not limited to the above. For example, input information may be acquired in step 510, and the input information may be entered into a business feasibility evaluation sheet in step 511. After examining whether there is any information missing, the processing related to the second acquisition module 212 from step 512 onwards may be performed only if there is information missing. In this case, it may be possible to determine whether there is a lack of information by studying a business feasibility evaluation sheet with a sufficient amount of information, created by a veteran bank employee, and a business feasibility evaluation sheet with insufficient information. This also applies to the flows in Figures 20 to 21 and 23 below.

Alternatively, after determining whether there is a lack of information, when outputting the questionnaire in step 517, it is also possible to distinguish between questions about input items that are determined to have an adequate amount of information (i.e., questions for improvement) and questions about input items that are determined to have an insufficient amount of information (questions that need to be checked to make up for the lack of information). This also applies to the flows in Figures 20 to 21 and 23 below.

Below, examples of the business feasibility assessment sheet and question sheet displayed on the user terminal 103 will be explained with reference to Figures 7 to 18. First, the business feasibility assessment sheet will be explained for each input item.

Figure 7 shows an example of input information 700 corresponding to the input item "Company Overview." The input information 700 shown in the figure is composed of multiple pieces of input information for multiple detailed items. The multiple detailed items include the company name, representative, industry, business content, bases, number of employees, business history, and history.

8 shows an example of input information 800 corresponding to the input item "commercial flow." The input information 800 shown in the figure is composed of a plurality of pieces of input information for a plurality of detailed items. The detailed items include the business model, main business, sub-business, suppliers, subcontractors, sales destinations, and consumer segments.
The underlined text contains links to related websites. Users can check the links as necessary to see if the content is correct, and prevent hallucination from occurring. The same applies to the links to websites in the examples in and after Figure 10.

Figure 9 shows an example of input information 900 corresponding to the input item "Market." The input information 900 shown in the figure is composed of multiple pieces of input information for multiple detailed items. The multiple detailed items include market size, TAM (Total Addressable Market), SAM (Serviceable Available Market), SOM (Serviceable Obtainable Market), and industry trends.

10 shows an example of input information 1000 corresponding to the input item "3C analysis." The input information 1000 shown in the figure is composed of a plurality of pieces of input information for a plurality of detailed items. The detailed items include customers, competitors, and the company itself.
The underlined text contains links to relevant websites.

11 shows an example of input information 1100 corresponding to the input item "Five Forces Analysis." The input information 1100 shown in the figure is composed of a plurality of pieces of input information for a plurality of detailed items. The detailed items include competitors, sellers, buyers, new entrants, and substitutes.
The underlined text contains links to relevant websites.

12 shows an example of input information 1200 corresponding to the input item "SWOT analysis." The input information 1200 shown in the figure is composed of multiple pieces of input information for multiple detailed items. The multiple detailed items include strengths, weaknesses, opportunities, threats, strengths*opportunities, weaknesses*opportunities, strengths*threats, and weaknesses*threats.
The underlined text contains links to relevant websites.

13 shows an example of input information 1300 corresponding to the input item "management system." The input information 1300 shown in the figure is composed of multiple pieces of input information for multiple detailed items. The multiple detailed items include the management vision, organizational structure, and key people.
The underlined text contains links to relevant websites.

14 shows an example of input information 1400 corresponding to the input item “financial summary.” The input information 1400 shown in the drawing is composed of an income statement 1401 and a balance sheet 1402.
Income statement 1401 shows sales, cost, gross profit, selling expenses, operating profit, ordinary profit, and EBITDA (Earnings Before Interest Taxes Depreciation and Amortization) for the months 2020 through 2024.
On the other hand, balance sheet 1402 shows cash and deposits, fixed assets, total assets, long-term borrowings, equity capital, and CF (Cash Flow) multiple for December 2020 to 2024.

Incidentally, the input information corresponding to the input item "Financial summary" is merely copied data from the income statement and balance sheet, so it may be generated by the business support server 101 rather than by the generation AI module 310.

Next, FIG. 15 shows an example of input information 1500 corresponding to the input item "management issues." The input information 1500 shown in the figure is composed of multiple pieces of input information for multiple detailed items. The multiple detailed items include short-term issues and long-term issues.

Figure 16 shows an example of input information 1600 corresponding to the input item "management policy." The input information 1600 shown in the figure is composed of multiple pieces of input information for multiple detailed items. The multiple detailed items include personnel policies, business policies, financial policies, and DX (Digital Transformation) policies.

Next, we will explain the questionnaire. As an example, we will only explain the input items "5 Forces Analysis" and "SWOT Analysis".

Figure 17 shows an example of a question list 1700 for the input item "Five Forces Analysis." The question list 1700 shown in the figure is composed of multiple questions about multiple question items. The multiple question items include company overview, market environment, supply chain, customer information, barriers to new entry, and substitutes.

18 shows an example of a question list 1800 for the input item "SWOT analysis." The question list 1800 shown in the figure is composed of a plurality of questions for a plurality of question items. The plurality of question items include strengths, weaknesses, opportunities, threats, strengths*opportunities, weaknesses*opportunities, strengths*threats, and weaknesses*threats.
The above is the explanation of the business support process 500.

According to the business support process 500 described above, a business feasibility evaluation sheet can be created using generation AI. This reduces the effort and time required for information gathering and document creation that was previously required to create a business feasibility evaluation sheet.

In addition, according to this business support process 500, reference information such as accounting information and financial statement data can be included as prompts depending on the input items to be generated. This improves the accuracy of the input information output from the generation AI.

Furthermore, according to this business support process 500, if there is any information missing from the input information once it has been output, a question is output to elicit the missing information from the business operator. The bank employee can use this question to elicit the missing information from the business operator and complete the contents of the business feasibility evaluation sheet. This allows even a bank employee with poor document creation skills or listening skills to create a high-quality business feasibility evaluation sheet. As a result, the variation in the quality of business feasibility evaluation sheets between bank employees is reduced.

2. Modifications The above embodiment may be modified as follows. The following modifications may be combined with each other.

2-1. Business feasibility evaluation sheet The business feasibility evaluation sheet created in the above embodiment is an example of a document for evaluating the business feasibility of a business. A document with a different name (for example, a business plan) may be created as long as it is a document for evaluating the business feasibility of a business.

2-2. First Receiving Module 210
In the above embodiment, the first reception module 210 receives the input of a business name as information for identifying the target business (see step 503 in FIG. 5). However, the business name is merely one example of information for identifying the target business. Instead of the business name, the first reception module 210 may receive other identification information such as a corporate number, a CIF (Customer Information File) number, a location, and a representative's name. The first reception module 210 may then search the business information DB 221 to identify the business ID and industry ID corresponding to the received identification information.

2-3. Method of acquiring input information, etc. In the above-mentioned business support process 500, input information and a question sentence are acquired for each input item. However, such an acquisition method is not necessarily required. As another acquisition method, input information, etc. may be acquired for each of a plurality of input items. For example, input information, etc. may be acquired at once for two input items, "company profile" and "sales flow". As another example, input information, etc. may be acquired at once for all input items. In that case, the first acquisition module 211 may include format information of the business feasibility evaluation sheet in the prompt, and instruct the generation AI module 310 to output the business feasibility evaluation sheet according to the format information.

2-4. Generative AI module 310
In the above-mentioned business support process 500, the second prompt includes model input information (see step 512 in FIG. 5). This enables the generation AI module 310 to identify information that is lacking in the input information once generated. However, the method of making it possible to identify the lacking information is not limited to this. Another method is to fine-tune the generation AI module 310. Specifically, the generation AI module 310 is made to learn multiple pairs of input items and input information for the input items. This enables the generation AI module 310 to identify the lacking information in light of the learned input information.

2-5. Correction of business feasibility evaluation sheet In the above business support process 500, if there is missing information, the user corrects the business feasibility evaluation sheet. However, such a correction method is not necessarily required. As another correction method, the generation AI module 310 may correct the business feasibility evaluation sheet.

In this case, the business support server 101 further includes a second reception module 1901, a third acquisition module 1902, and a second output module 1903. Hereinafter, the business support server 101 including these modules will be referred to as the business support server 1900. FIG. 19 shows an example of the configuration of this business support server 1900.

A second reception module 1901 shown in the figure receives an answer to the output question from the user.
The third acquisition module 1902 inputs the answer received by the second reception module 1901 to the generation AI to acquire input information in which the missing information is supplemented. Note that the generation AI referred to here is specifically the generation AI module 310 described later.
The second output module 1903 outputs the input information acquired by the third acquisition module 1902. Specifically, this module causes the acquired input information to be displayed on the display of the user terminal 103.

Next, the operation of the business support server 1900 will be described. The operation to be described is business support process 2000 that supports the creation of a business feasibility evaluation sheet. Figures 20 and 21 are flow diagrams showing an example of the same process. Process 2000 shown in these figures is common to process 500 shown in Figure 5 in that it includes steps 501 to 516. Below, a description of these steps will be omitted.

In the business support process 2000, step 2001 is executed following step 516. In step 2001, the first output module 213 displays the generated business feasibility evaluation sheet and question sheet on the user terminal 103. At this time, the displayed question sheet has an answer input field for each question.

Figure 22 shows an example of a question list 2200 related to the input item "Five Forces Analysis" from among the displayed questionnaire sheets. The question list 2200 shown in the figure has an answer input field 2201 for each question.

The user uses the questions in this question list 2100 to find out the missing information from the business. The user then enters the information found out in the answer input field 2201. Hereinafter, the information entered in the answer input field 2201 and the corresponding questions will be referred to as "additional information."

The user fills in the answer input fields 2201 for the other input items in a similar manner. The user then selects a correction instruction button (not shown). This correction instruction button is a button for instructing corrections to the business feasibility evaluation sheet. By selecting this button, the entered additional information is sent from the user terminal 103 to the business support server 1900, and the second reception module 1901 acquires the additional information (step 2002).

Next, the third acquisition module 1902 selects the input item to be processed from among the input items of the business feasibility evaluation sheet to be corrected (step 2003). Then, the module generates a prompt for the selected input item to be input to the generation AI module 310 (step 2004). The generated prompt includes the following information:
(1) Input information for the input item to be processed obtained in step 510; (2) Additional information for the input item to be processed; (3) Model input information; (4) A statement instructing the user to correct the input information in (1) based on the additional information in (2) with reference to the input information in (3).

The input information in (2) refers to one or more pieces of input information for the input items to be processed that have been previously output by the generation AI module 310. This input information is, for example, input information generated by a veteran bank employee using the generation AI module 310.

Next, the third acquisition module 1902 inputs the generated prompt to the generation AI module 310 (step 2005). The generation AI module 310 outputs an answer corresponding to the input prompt. The output answer (in other words, the input information) is acquired by the third acquisition module 1902 (step 2006). The third acquisition module 1902 inputs the acquired input information into the corresponding input item field of the business feasibility evaluation sheet (step 2007). At that time, the module replaces the old input information already entered in that field with the new input information.

Next, the third acquisition module 1902 judges whether all input items have been processed (step 2008). If the result of this judgment is that all input items have not been processed (NO in step 2008), the module returns to step 2003 and selects the next input item. On the other hand, if the result of this judgment is that all input items have been processed (YES in step 2008), the second output module 1903 displays the corrected business feasibility evaluation sheet on the user terminal 103 (step 2009).
The above is the explanation of the business support process 2000.

According to the business support process 2000 described above, the generation AI module 310 can correct the business feasibility evaluation sheet. This can save the effort and time required for correction.

Note that in the above process, the user manually inputs the additional information, but the method of inputting the additional information is not limited to this. As another input method, for example, an image of the additional information captured by a camera may be uploaded to the business support server 1900. As another method, audio data containing the additional information may be uploaded to the business support server 1900.

In addition, in the above process, the user (i.e., the bank employee) inputs the additional information, but the business operator may also input this additional information. In that case, the business support server 1900 may display only the questionnaire on the user terminal 103 used by the business operator, and obtain the additional information from the business operator's terminal. This eliminates the need for the bank employee to input the additional information.

2-6. Conditions for outputting a question In the above-described business support process 500, a command is issued to output a question at the second prompt (see step 512 in FIG. 5). At that time, the second acquisition module 212 may specifically specify conditions for outputting a question.

For example, the second acquisition module 212 may instruct the generation AI to output a specified question when the input information acquired by the first acquisition module 211 is information about the economic situation of the business operator and the change in the economic situation satisfies a specified condition.

In this case, the information on the business's economic situation is, for example, information input to the input item "financial summary."
An example of a change in economic conditions that meets a predetermined condition is when a financial indicator (e.g., sales) of an enterprise falls by a predetermined value or a predetermined percentage or more compared to the previous year.
The predetermined question is a question for obtaining information regarding the cause of the change in economic situation from the target business as missing information.
Specifically, the generating AI is the generating AI module 310 described later.

In this way, by outputting questions when a business operator's financial indicators deteriorate, the user is prompted to investigate the cause of the deterioration. The user can then use the cause of the deterioration that he or she finds out from the business operator to revise the business feasibility evaluation sheet.

2-7. Number of times prompts are input In the above business support process 500, two prompts are input for each input item to the generation AI module 310. The first prompt generates input information, and the second prompt generates a question to obtain missing information. However, it is not necessarily required to input two prompts, and both the input information and the question may be generated with only one prompt.

In this case, the first acquisition module 211 instructs the generation AI to output a question for acquiring information that is lacking when generating input information for an input item of a document for evaluating the business feasibility of the business designated via the first reception module 210 from the designated business. Then, the module acquires the question output from the generation AI.
The specific operation of the first acquisition module 211 will be described below.

Figure 23 is a flow diagram showing an example of a business support process 2300 according to this modified example. Process 2300 shown in the figure is common to process 500 shown in Figure 5 in that it includes steps 501 to 507. In the following, a description of these steps will be omitted.

In the business support process 2300, step 2301 is executed following step 507. In step 2301, the first acquisition module 211 generates a prompt to be input to the generation AI module 310. The prompt generated here includes the following information:
(1) the website address of the business acquired in step 503; (2) detailed items constituting the input items to be processed; (3) reference information acquired in step 507; (4) a statement commanding the output of input information corresponding to the detailed items in (2) for the business designated by the website address in (1) with reference to the reference information in (3); (5) model input information; (6) a statement commanding the output of a question to acquire information from the target business that is missing in light of the input information in (5) in the input information to be output in accordance with the command in (4);

In addition, the detailed items of (2) are, for example, in the case of the input item "company overview", the company name, representative, industry, business content, base, number of employees, business history, and history (see FIG. 7).
The input information in (5) is one or more pieces of input information for the input item to be processed that have been previously output by the generating AI module 310. This input information is, for example, input information generated by a veteran bank employee using the generating AI module 310.

Next, the first acquisition module 211 inputs the generated prompt to the generation AI module 310 (step 2302). The generation AI module 310 outputs an answer corresponding to the input prompt. The output answer (in other words, the input information and question) is acquired by the first acquisition module 211 (step 2303). The first acquisition module 211 inputs the acquired input information into the corresponding input item field of the business feasibility evaluation sheet (step 2304). Furthermore, if the acquired answer contains a question, the module inputs the question into the corresponding input item field of the question sheet (step 2305).

Next, the first acquisition module 211 judges whether all input items have been processed (step 2306). If the result of this judgment is that all input items have not been processed (NO in step 2306), the module returns to step 505 and selects the next input item. On the other hand, if the result of this judgment is that all input items have been processed (YES in step 2306), the first output module 213 displays the generated business feasibility evaluation sheet and questionnaire sheet on the user terminal 103 (step 2307).
The above is the explanation of the business support process 2300.

The process described above can also be used to create a business feasibility assessment sheet and a questionnaire sheet.

2-8. Creator of business feasibility evaluation sheet In the above embodiment, a bank employee is assumed to be the creator of the business feasibility evaluation sheet. However, a bank employee is merely one example of a creator. A business feasibility evaluation sheet may be created by another creator, for example, a business operator.

In the above-described business support process 500, information on the target business is acquired as reference information (see step 507 in FIG. 5). The acquired reference information may include information on the business of the business (in other words, business data).

In this case, the business support server 101 further includes a business information DB. This business information DB is a database that stores information about the business of each business. The information stored in this DB includes information about the attendance of employees employed by the business. The business information of each business is stored in association with the business ID.

When the first acquisition module 211 identifies a business information DB in step 506 of FIG. 5, the first acquisition module 211 acquires business data of the target business entity from the business information DB in the following step 507. The first acquisition module 211 then instructs the generation AI module 310 to output input information for the input items based on the acquired business data.

2-10. Method of selecting reference information In the above business support process 500, the input item table 228 is referenced to identify the DB from which reference information is obtained (see step 506 in FIG. 5). That is, the DB from which reference information is obtained is changed for each input item. However, this change in DB is not necessarily required. As another aspect, reference information may be obtained from the same DB for all input items. Even in such an aspect, a business feasibility evaluation sheet can be created.

2-11. Model input information In the above business support process 500, model input information is included in the second prompt (see step 512 in FIG. 5). This model input information may be changed according to the attributes of the target business. For example, it may be changed according to the industry of the target business. In that case, the selected input information is input information previously output for a business of the same industry as the target business. As another example, it may be changed according to the scale of the target business. In that case, the selected input information is input information previously output for a business of the same scale as the target business. In this way, by making the model input information variable, the accuracy of the answer output from the generation AI module 310 is improved.

2-12. Management of business feasibility evaluation sheet In the above embodiment, a business feasibility evaluation sheet is created, but when the information on the created business feasibility evaluation sheet becomes outdated, an alert may be output to the user. This allows the user to notice that the information on the business feasibility evaluation sheet has become outdated.

As a specific method, the business support server 101 records the creation date of the financial statement data used as a reference when creating the business feasibility evaluation sheet along with the created business feasibility evaluation sheet. The server then outputs an alert to the user when a specified period of time has passed since the creation date. This allows the user to realize that the information on the business feasibility evaluation sheet they created has become outdated.

2-13. Functional layout In the above embodiment, the business support process 500 is executed by two servers, the business support server 101 and the LLM server 102. However, this functional layout is merely an example. The functions of each server may be appropriately arranged according to the execution environment of the business support process 500. For example, some or all of the functions of the LLM server 102 may be provided in the business support server 101.

2-14. System Components Each device constituting the information processing system 100 may be, for example, a mobile terminal such as a smartphone, tablet, mobile phone, or personal digital assistant (PDA), or may be a wearable terminal such as glasses, wristwatch, or clothing. Each device may be a stationary or mobile computer, or a server located on the cloud or network. Each device may function as a VR (Virtual Reality) terminal, an AR (Augmented Reality) terminal, or an MR (Mixed Reality) terminal. Alternatively, each device may be a combination of a plurality of these terminals. For example, a combination of one smartphone and one wearable terminal may function as one terminal logically. Other information processing terminals may also be used.

2-15. Others The present invention is not limited to the above-mentioned embodiments, and various modifications are included. For example, the above-mentioned embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the configurations described. It is also possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.

Furthermore, the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in hardware, in part or in whole, for example by designing them as integrated circuits. Furthermore, the above-mentioned configurations, functions, etc. may be realized in software, by a processor interpreting and executing a program that realizes each function. Information on the programs, tables, files, etc. that realize each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are connected to each other.
The above-mentioned embodiment discloses at least the following configurations.
(1)
An information processing system for supporting the creation of documents for evaluating the business viability of a business operator,
A first reception means for receiving a designation of a business operator;
A first acquisition means for instructing a generation AI to output input information for an input item of a document for evaluating the business feasibility of the specified business and acquiring the output input information;
A second acquisition means for instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
a first output means for outputting the acquired question sentence;
An information processing system comprising:
(2)
The second acquisition means instructs the generation AI to identify the missing information based on input information for the input items previously output by the generation AI.
An information processing system according to (1).
(3)
The generation AI is a language model that has been trained on a plurality of pairs of the input items and input information for the input items in order to identify the missing information.
An information processing system according to (1).
(4)
A second receiving means for further receiving an answer to the question;
A third acquisition means for inputting the answer to the generating AI and acquiring another input information in which the missing information is supplemented;
A second output means for outputting the acquired other input information;
The information processing system according to (1), further comprising:
(5)
The first acquisition means instructs the generation AI to output input information for the input items for the specified business operator based on at least one of accounting information and financial statement data of the specified business operator.
An information processing system according to (1).
(6)
The input items are any of company profile, trade flow, market, 3C analysis, 5 forces analysis, SWOT analysis, management system, financial summary, management issues, and management measures.
An information processing system according to (1).
(7)
The second acquisition means instructs the generation AI to output a question for acquiring information on the cause of the change from the designated business operator as the missing information when the acquired input information is information on the economic situation of the designated business operator and a change in the economic situation satisfies a predetermined condition.
An information processing system according to (1).
(8)
An information processing method for supporting the creation of a document for evaluating the business viability of a business operator, the method comprising:
accepting the designation of the business operator;
A step of instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business operator, and acquiring the output input information;
Instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
outputting the acquired question sentence;
An information processing method comprising:
(9)
A program for supporting the creation of documents for evaluating the business feasibility of a business operator,
On the computer,
accepting the designation of the business operator;
A step of instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business operator, and acquiring the output input information;
Instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
outputting the acquired question sentence;
A program for executing.
(10)
An information processing system for supporting the creation of documents for evaluating the business viability of a business operator,
A reception means for receiving the designation of a business operator;
An acquisition means for instructing a generation AI to output a question sentence for acquiring missing information from the specified business when generating input information for an input item of a document for evaluating the business feasibility of the specified business, and acquiring the output question sentence;
an output means for outputting the acquired question sentence;
An information processing system comprising:

100: Information processing system, 101: Business support server, 102: LLM server, 103: User terminal

Claims (15)

An information processing system for supporting the creation of documents for evaluating the business viability of a business operator,
A first reception means for receiving a designation of a business operator;
A first acquisition means for instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business and acquiring the output input information;
A second acquisition means for instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
a first output means for outputting the acquired question sentence;
Equipped with
The second acquisition means instructs the generation AI to identify the missing information based on input information for the input items previously output by the generation AI.
Information processing system. An information processing system for supporting the creation of documents for evaluating the business viability of a business operator,
A first reception means for receiving a designation of a business operator;
A first acquisition means for instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business and acquiring the output input information;
A second acquisition means for instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
a first output means for outputting the acquired question sentence;
Equipped with
The generation AI is a language model that has been trained on a plurality of pairs of the input items and input information for the input items in order to identify the missing information .
Information processing system. An information processing system for supporting the creation of documents for evaluating the business viability of a business operator,
A first reception means for receiving a designation of a business operator;
A first acquisition means for instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business and acquiring the output input information;
A second acquisition means for instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
a first output means for outputting the acquired question sentence;
Equipped with
The second acquisition means instructs the generation AI to output a question for acquiring information on the cause of the change from the designated business operator as the missing information when the acquired input information is information on the economic situation of the designated business operator and a change in the economic situation satisfies a predetermined condition.
Information processing system. A second receiving means for further receiving an answer to the question;
A third acquisition means for inputting the answer to the generating AI and acquiring another input information in which the missing information is supplemented;
A second output means for outputting the acquired other input information;
The information processing system according to claim 1 , further comprising: The first acquisition means instructs the generation AI to output input information for the input items for the specified business operator based on at least one of accounting information and financial statement data of the specified business operator.
The information processing system according to any one of claims 1 to 3 . The input items are any of company profile, trade flow, market, 3C analysis, 5 forces analysis, SWOT analysis, management system, financial summary, management issues, and management measures.
The information processing system according to any one of claims 1 to 3 . An information processing method for supporting the creation of a document for evaluating the business viability of a business operator, the method comprising:
accepting the designation of the business operator;
A step of instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business operator, and acquiring the output input information;
Instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
outputting the acquired question sentence;
Including,
In the step of acquiring the question sentence, the generation AI is instructed to identify the missing information based on input information for the input item previously output by the generation AI.
Information processing methods. An information processing method for supporting the creation of a document for evaluating the business viability of a business operator, the method comprising:
accepting the designation of the business operator;
A step of instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business operator, and acquiring the output input information;
Instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
outputting the acquired question sentence;
Including,
The generation AI is a language model that has been trained on a plurality of pairs of the input items and input information for the input items in order to identify the missing information.
Information processing methods. An information processing method for supporting the creation of a document for evaluating the business viability of a business operator, the method comprising:
accepting the designation of the business operator;
A step of instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business operator, and acquiring the output input information;
Instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
outputting the acquired question sentence;
Including,
In the step of acquiring the question, when the acquired input information is information about the economic situation of the designated business operator and a change in the economic situation satisfies a predetermined condition, the generation AI is instructed to output a question for acquiring information about the cause of the change from the designated business operator as the missing information.
Information processing methods. A program for supporting the creation of documents for evaluating the business feasibility of a business operator,
On the computer,
accepting the designation of the business operator;
A step of instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business operator, and acquiring the output input information;
Instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
outputting the acquired question sentence;
Run the command ,
In the step of acquiring the question sentence, the generation AI is instructed to identify the missing information based on input information for the input item previously output by the generation AI.
program. A program for supporting the creation of documents for evaluating the business feasibility of a business operator,
On the computer,
accepting the designation of the business operator;
A step of instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business operator, and acquiring the output input information;
Instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
outputting the acquired question sentence;
Run the command,
The generation AI is a language model that has been trained on a plurality of pairs of the input items and input information for the input items in order to identify the missing information.
program. A program for supporting the creation of documents for evaluating the business feasibility of a business operator,
On the computer,
accepting the designation of the business operator;
A step of instructing a generation AI to output input information for an input item of a document for evaluating the business viability of the specified business operator, and acquiring the output input information;
Instructing the generation AI to output a question for acquiring missing information from the specified business entity with respect to the acquired input information, and acquiring the output question;
outputting the acquired question sentence;
Run the command,
In the step of acquiring the question, when the acquired input information is information about the economic situation of the designated business operator and a change in the economic situation satisfies a predetermined condition, the generation AI is instructed to output a question for acquiring information about the cause of the change from the designated business operator as the missing information.
program. An information processing system for supporting the creation of documents for evaluating the business viability of a business operator,
A reception means for receiving the designation of a business operator;
An acquisition means for instructing a generation AI to output a question sentence for acquiring missing information from the specified business when generating input information for an input item of a document for evaluating the business feasibility of the specified business, and acquiring the output question sentence;
an output means for outputting the acquired question sentence;
Equipped with
The acquisition means instructs the generation AI to identify the missing information based on input information for the input items previously output by the generation AI.
Information processing system. An information processing system for supporting the creation of documents for evaluating the business viability of a business operator,
A reception means for receiving the designation of a business operator;
An acquisition means for instructing a generation AI to output a question sentence for acquiring missing information from the specified business when generating input information for an input item of a document for evaluating the business feasibility of the specified business, and acquiring the output question sentence;
an output means for outputting the acquired question sentence;
Equipped with
The generation AI is a language model that has been trained on a plurality of pairs of the input items and input information for the input items in order to identify the missing information.
Information processing system. An information processing system for supporting the creation of documents for evaluating the business viability of a business operator,
A reception means for receiving the designation of a business operator;
An acquisition means for instructing a generation AI to output a question sentence for acquiring missing information from the specified business when generating input information for an input item of a document for evaluating the business feasibility of the specified business, and acquiring the output question sentence;
an output means for outputting the acquired question sentence;
Equipped with
The acquisition means instructs the generation AI to output a question for acquiring information on the cause of the change from the designated business operator as the missing information when the acquired input information is information on the economic situation of the designated business operator and a change in the economic situation satisfies a predetermined condition.
Information processing system.