Information providing device

The information providing device uses GPT to enhance application forms with business information and analysis, addressing the need for specialized knowledge and improving subsidy acceptance rates.

JP7584760B1Active Publication Date: 2024-11-18五味田 匡功 +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
JP2023090330
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2023-05-31
Publication Date
2024-11-18
Estimated Expiration
2043-05-31

AI Technical Summary

Technical Problem

Existing technologies, such as Patent Document 1, are inadequate in supporting the creation of application forms for subsidies as they do not adequately address the need for specialized knowledge in selecting and writing items, which affects the adoption rate.

Method used

An information providing device utilizing GPT to generate answers that prompt the provision of business information, analyze business data, and provide analysis results, thereby enhancing the content of application forms to increase their acceptance rate.

Benefits of technology

The device supports the creation of application forms by providing relevant business information and analysis results, significantly increasing the chances of subsidy acceptance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007584760000001_ABST
    Figure 0007584760000001_ABST
Patent Text Reader

Abstract

To provide a means of assisting in the preparation of applications for subsidies, etc. by providing information useful for document preparation, etc. [Solution] The information providing device 1 of the present invention comprises a question receiving unit 111 configured to receive questions regarding the preparation of subsidy application forms used to apply for subsidies, grants and other subsidies for businesses, an answer generating unit 112 configured to cause GPT to generate answers to the questions that encourage the provision of data related to the business, a business data receiving unit 114 configured to receive data for the business that is business data provided based on the answers, and an analysis unit 115 configured to generate analysis results for the business data.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an information providing device. [Background technology]

[0002] When conducting business, it is common to utilize subsidies, grants, and other payments (subsidies, etc.). When utilizing subsidies, etc., business operators normally submit an application to an entity that reviews the subsidies, etc. The decision as to whether or not the subsidy, etc. will be adopted is based on the review of this application. However, preparing application forms, which require specialized knowledge and labor, can be difficult for business operators. Thus, there is a demand for technology that supports the preparation of applications for subsidies, etc.

[0003] Regarding technology for supporting the creation of applications for subsidies, etc., Patent Document 1 discloses a method for supporting budget requests and subsidy applications in public sewerage works, characterized by creating a composite diagram from the results of outline design and detailed design. The technology of Patent Document 1 eliminates the need for a lot of effort in creating a control diagram for budget requests and subsidy applications, and can have the effect of significantly reducing the number of work hours required for budget requests and subsidy applications. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2006-004022 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology of Patent Document 1 can only reduce the labor required to create a control chart for applications for subsidies in public sewerage projects. In general, when creating an application, the likelihood of the application being adopted can change depending on the selection of items to be described and how they are written. Therefore, there is room for further improvement in the technology of Patent Document 1 in terms of supporting the creation of general application forms that require specialized knowledge in the selection of items to be described and how they are written.

[0006] The present invention has been made in consideration of the above circumstances, and its object is to provide a means for supporting the preparation of applications for subsidies and the like by providing information useful for document preparation, etc. [Means for solving the problem]

[0007] As a result of intensive research into solving the above problems, the inventors have found that the above object can be achieved by having GPT generate a response that prompts the provision of information about the business, and generating an analysis result for the provided business data based on the response. The inventors have then completed the present invention. Specifically, the present invention provides the following.

[0008] The present invention provides an information providing device comprising: a question receiving unit configured to receive questions regarding the preparation of subsidy application forms used to apply for subsidies, grants, and other benefits (subsidies, etc.) for a business; an answer generating unit configured to cause a GPT to generate answers to the questions that encourage the provision of information about the business (business information); a business data receiving unit configured to receive the business information, which is business data provided based on the answers; and an analysis unit configured to generate analysis results for the business data.

[0009] If the business information, which is information about the business, is properly described in the application form for subsidies, etc., the success rate of the subsidy, etc. can be increased. However, since the business information to be described in the application form for subsidies, etc. varies depending on the type of business, etc., it may not be easy to list the business information without omission or excess.

[0010] The information providing device of the present invention generates answers to questions about the preparation of subsidy applications that encourage the provision of business information using a GPT trained to generate human-like text using a large-scale corpus of text data. The corpus of text data contains knowledge about business information that should be included in applications for subsidies, etc., for various businesses. Thus, the information providing device of the present invention can provide information that further increases the acceptance rate of subsidies, etc., such as answers indicating business information that should be further included in applications for subsidies, etc. As a result, the information providing device of the present invention can support the preparation of applications for subsidies, etc.

[0011] Incidentally, the business information to be included in the application form may include business data exemplified by projected financial statements, etc. By attaching the results of an analysis of such business data to the subsidy application form, the adoption rate of the subsidy may be further increased.

[0012] The information providing device of the present invention can receive business data provided based on a response to a request for providing business information generated by GPT, and generate the analysis results. In other words, the information providing device of the present invention can provide information that further increases the acceptance rate of subsidies, etc., such as the analysis results of business data to be included in applications for subsidies, etc. In this way, the information providing device of the present invention can support the creation of applications for subsidies, etc. Effect of the Invention

[0013] The present invention can provide a means for supporting the preparation of applications for subsidies and the like by providing information that is useful for document preparation. [Brief description of the drawings]

[0014] [Figure 1] FIG. 1 is a block diagram showing an example of the hardware configuration and software configuration of a system S in this embodiment. [Diagram 2] FIG. 2 is a main flowchart showing an example of a preferable flow of the information provision process executed by the information provision device 1 of this embodiment. [Diagram 3] FIG. 3 is a continuation of FIG. [Figure 4] FIG. 4 is a continuation of FIG. [Diagram 5] FIG. 5 shows an example of a response for encouraging the provision of information, which is generated in the information provision process of this embodiment. [Figure 6] FIG. 6 shows an example of an analysis result generated in the information provision process of this embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] Hereinafter, an example of an embodiment of the present invention will be described in detail with reference to the drawings.

[0016] 1 is a block diagram showing an example of a hardware configuration and a software configuration of a system S in this embodiment. The information providing system S (system S) is configured to include at least an information providing device 1. The system S is preferably configured to include a terminal T capable of communicating with the information providing device 1 via a network N.

[0017] [Subsidy application form] In this embodiment, the "subsidy application form" refers to various documents used to apply for subsidies, grants, and other payments (subsidies, etc.) for a business. The content of the business and the type of subsidy, etc. are not particularly limited. Subsidies, etc. include, for example, subsidies, grants, support funds, cooperation funds, and payments for businesses. The information providing device 1 of this embodiment is configured to cause the GPT to generate a response that prompts the provision of information about the business (business information), and therefore can provide information for creating a subsidy application form without limiting the content of the business and the type of subsidy, etc.

[0018] [Information provision device 1] The information providing device 1 includes at least a control unit 11, a storage unit 12, and a communication unit 13. There is no particular limitation on the type of the information providing device 1. Examples of the type include a server, a cloud server, and the like.

[0019] [Control unit 11] The control unit 11 includes a central processing unit (CPU), a random access memory (RAM), and a read only memory (ROM).

[0020] The control unit 11 cooperates with the storage unit 12 and / or the communication unit 13 as necessary. The control unit 11 then realizes software components of the program of this embodiment executed by the information provision device 1, such as a question receiving unit 111, an answer generating unit 112, a reading unit 113, a business data receiving unit 114, an analyzing unit 115, an application draft receiving unit 116, an amendment draft generating unit 117, and a providing unit 118. The functions provided by each of the software components of the program of this embodiment will be shown in the description of a preferred flow of the information provision process described later.

[0021] [Storage section 12] The storage unit 12 is a device in which data and / or files are stored, and includes a data storage unit such as a hard disk, a semiconductor memory, a recording medium, and a memory card.

[0022] The storage unit 12 may have a mechanism that enables connection via the network N to a storage device or storage system such as a NAS, a SAN, cloud storage, a file server, and / or a distributed file system.

[0023] The memory unit 12 stores programs executed by the microcomputer, received questions, answers to the questions, dialogue history, a language model (GPT), an application form database, a voice recognition model, a voice synthesis model, etc.

[0024] (Language Model) The language model is not particularly limited as long as it is a Generative Pre-trained Transformer or a language model based thereon. Hereinafter, such a language model is also simply referred to as "GPT". The language model may be stored in the storage unit 12, or may be stored in another server available by communication via the communication unit 13 in order to simplify the configuration of the storage unit 12 of the information providing device 1.

[0025] Providing the correct information about a project (project information) on an application for a subsidy, etc. can increase the rate of grant acceptance. However, because the project information to be included on an application for a subsidy, etc. varies depending on the type of project, etc., it may not be easy to list the project information without omission or excess.

[0026] The language model, which is a GPT, may include knowledge about business information that should be included in an application for a subsidy, etc., for various businesses. This allows the information providing device 1 of this embodiment to provide information that further increases the acceptance rate of a subsidy, etc., in response to a question about the creation of a subsidy application, such as an answer indicating business information that should be further included in the application for the subsidy, etc.

[0027] Specific language models include, for example, ChatGPT, GPT-3, GPT-4 by OpenAI, and GPT-J by EleutherAI. By storing a language model trained on a corpus of large-scale text data in the storage unit 12, etc., the information providing device 1 of the present embodiment can generate a response indicating business information that should be further included in an application for a subsidy, etc., regardless of the content of the business. In addition, storing such a language model in the storage unit 12, etc., enables the information providing device 1 of the present embodiment to generate a response that sounds like it was written by a human.

[0028] The language model is preferably a GPT that has been pre-trained using training data that associates the business content included in the question with business information to be written in an application for a subsidy or the like. This allows the information providing device 1 of the present embodiment to generate an answer indicating business information to be further included in the application for a subsidy or the like so that the business information is written in the application without excess or deficiency according to the business content. The pre-training referred to here is not particularly limited. Examples of the pre-training include transfer learning and / or fine tuning using the above-mentioned training data.

[0029] The language model may be a lightweight language model that is produced by "distillation," which uses an already-trained language model. This allows the information providing device 1 to provide information quickly with lighter processing. In particular, by using a language model that has been "distilled" after pre-learning using the above-mentioned learning data, the information providing device 1 can prevent the time from a question to a response from being lengthened by pre-learning.

[0030] (Application database) The application database (database) stores information on adopted subsidy applications whose applications have been adopted. The information is not particularly limited as long as it includes information that enables acquisition of part or all of the data of the adopted subsidy application. The information includes, for example, part or all of the data of the adopted subsidy application and one or more of the URL of the adopted subsidy application. In order to enable the information providing device 1 to provide information corresponding to the content of the project included in the received question, it is preferable that the information is stored in association with the content, classification, etc. of the project related to the adopted subsidy application.

[0031] (Speech recognition model) The voice recognition model is not particularly limited as long as it is a model that can convert the form of a question (question data) from voice to text. The voice recognition model includes, for example, a model related to acoustic analysis, an acoustic model, a language model, and the like.

[0032] By including a model related to acoustic analysis in the voice recognition model, the information providing device 1 can examine the features (frequency and volume) of the voice constituting the question, and extract and convert it into data that is easy to handle in the acoustic model.

[0033] By including an acoustic model in the voice recognition model, the information providing device 1 can compare the feature amount digitized in the acoustic analysis with the data of the pre-trained acoustic model, and extract phonemes, which are the smallest units of sound waves.

[0034] By including a language model in the speech recognition model, the information providing device 1 can convert the extracted phonemes into accurate text data. In order to improve the recognition accuracy, the speech recognition model is preferably a model using a neural network, such as the speech recognition model Whisper by OpenAI.

[0035] (Speech synthesis model) The voice synthesis model is not particularly limited as long as it is a model that can convert text data such as an answer generated by the answer generating unit 112 into a voice that reads out the text data. The voice synthesis model may be, for example, a rule-based voice synthesis model, a waveform concatenation type voice synthesis model, a statistical parametric voice synthesis model, or the like.

[0036] Among them, the voice synthesis model is preferably a statistical parametric voice synthesis model exemplified by hidden Markov model voice synthesis and neural network voice synthesis, which allows the information providing device 1 to perform voice synthesis with higher expressiveness and more naturalness than the hidden Markov model.

[0037] By storing the voice recognition model and the voice synthesis model in the storage unit 12, the information providing device 1 can provide information for creating documents related to subsidy applications through voice dialogue. In particular, by storing the voice recognition model and the voice synthesis model of statistical parametric voice synthesis in the storage unit 12, the information providing device 1 can provide information for creating documents related to subsidy applications through natural voice dialogue like when interacting with a human.

[0038] [Communications Section 13] The communication unit 13 is not particularly limited as long as it connects the information providing device 1 to the network N and enables communication with the terminal T, etc. Examples of the communication unit 13 include a wireless device compatible with a mobile phone network, a device connectable to a wireless LAN, and a network card compatible with the Ethernet standard.

[0039] [Network N] The type of the network N is not particularly limited as long as it enables communication between the information providing device 1 and the terminal T. The type of the network N is, for example, the Internet, a mobile phone network, a wireless LAN, or the like.

[0040] [Terminal T] The terminal T is, for example, a personal computer, a laptop computer, a smartphone, a tablet terminal, etc. The terminal T can execute a process of displaying information provided by the information providing device 1, a process of providing a user's input via a user interface to the information providing device 1, and the like.

[0041] [Main Flowchart of Information Provision Processing] Fig. 2 is a main flow chart showing an example of a preferred flow of the information provision processing executed by the information provision device 1 of this embodiment. Fig. 3 is a diagram continuing from Fig. 2. Fig. 4 is a diagram continuing from Fig. 3. The following is an example of a preferred flow of the information provision processing executed by the information provision device 1 of this embodiment using Figs. 2 to 4.

[0042] First, the control unit 11 executes the question receiving unit 111 in cooperation with the storage unit 12 and the communication unit 13. Then, the control unit 11 executes the process of step S1 related to the question receiving unit 111.

[0043] [Step S1: Receiving a question] The control unit 11, in cooperation with the storage unit 12 and the communication unit 13, executes a process of receiving a question from the terminal T etc. (step S1, question receiving step). The control unit 11 stores the received question in the storage unit 12 and moves the process to step S2.

[0044] (Regarding questions) The form of the question that can be received in the question receiving step includes at least a text form. The questions received in the question receiving step include questions related to the preparation of a subsidy application form used to apply for subsidies for a project. Hereinafter, when there is simply a "received question", it refers to a question related to the preparation of a subsidy application form received in the question receiving step. Furthermore, when there is simply a "project related to the received question", it refers to a project corresponding to the subsidy application form related to the question. In addition, the questions that can be received in the question receiving step include questions that include information that the user is prompted to provide in the answer generated by the information providing device 1. Hereinafter, such questions are also referred to as "information-attached questions".

[0045] In order to enable the information providing device 1 to provide information for document creation related to a subsidy application form by voice dialogue, it is preferable that the form of the question that can be received in the question receiving step includes a voice form. In this case, it is preferable that the management process includes the processes of steps S2 to S3 related to the question receiving unit 111.

[0046] [Step S2: Determine whether the question is in the form of voice] The control unit 11, in cooperation with the storage unit 12 and the communication unit 13, executes a process of determining whether the question received in step S1 is in the form of voice (step S2, question form determination step). If it is determined that the question is in the form of voice, the control unit 11 moves the process to step S3. If it is not determined that the question is in the form of voice, the control unit 11 moves the process to step S4.

[0047] [Step S3: Convert the question into text format] The control unit 11, in cooperation with the storage unit 12, executes a process of converting the question form determined to be a voice form in step S2 into a text form using a voice recognition model stored in the storage unit 12 (step S3, voice recognition step). The control unit 11 moves the process to step S4.

[0048] Next, the control unit 11 executes the answer generation unit 112 in cooperation with the storage unit 12 and the communication unit 13. Then, the control unit 11 executes the process related to the answer generation unit 112, that is, step S4.

[0049] [Step S4: Generate answers to questions] The control unit 11, in cooperation with the storage unit 12 and the communication unit 13, executes a process of causing the GPT stored in the storage unit 12 or the like to generate an answer to the question received in step S1 (step S4, answer generation step). The control unit 11 moves the process to step S5.

[0050] The answer generation step causes the GPT to generate an answer to the question received in step S1, which prompts the provision of information (business information) related to the received question. That is, the control unit 11 processes the neural network of the GPT stored in the storage unit 12 or the like with an engine corresponding to the GPT, and executes a process of generating an answer using the question and dialogue history as input.

[0051] Such generation is realized, for example, by a prompt such as "The business content is (※Business content included in the received question). When completing a grant application for this business, please list the information and data necessary to make the application have a high chance of being accepted from the list of elements required for the application." However, in the above prompt, "(※Business content included in the received question)" is replaced by the question received in step S1 or the business content extracted from the question.

[0052] The answer generation step preferably includes causing the GPT to generate an answer that includes the example business information. Examples of prompts that cause the GPT to generate such an answer include various prompts, such as those illustrated below.

[0053] Example prompts: "How do you highlight the relevance of your project to the grant program objectives?", "How do you demonstrate the innovation or uniqueness of your project?", "How do you explain the feasibility, planning, and measurement methods?", "How do you explain the project's impact on the local economy and industry?", "How do you explain the project's commitment to the environment and sustainability?", "How do you properly demonstrate risks and how they are managed?", "How do you clearly demonstrate the project implementation structure and team member roles?", "How do you explain the budget plan and cost-effectiveness evaluation method?", etc.

[0054] If the question received in step S1 is an information-attached question, the answer generation step preferably determines whether there is business information that should be further included in the application for a subsidy or the like, and generates an answer that encourages the provision of further information if such business information is present. Such generation is realized, for example, by a prompt such as "(※The received information-attached question) Please let us know if there is a lack of information and data necessary to make the application for a high acceptance rate." However, in the above prompt, "(※The received information-attached question)" is replaced by the information-attached question received in step S1. This allows the information providing device 1 to provide information that further increases the acceptance rate of the subsidy or the like, that is, an answer indicating business information that should be further included in the application for a subsidy or the like.

[0055] To prevent processing from becoming too large, the answer generation step is preferably a procedure in which an upper limit is set on the number of tokens in answer generation (e.g., 4096 tokens). A "token" is a processing unit equivalent to a character, word, etc. in GPT. The number of tokens in answer generation is the sum of the number of tokens in the history, the number of tokens in the question, and the number of tokens in the answer. This allows the information providing device 1 of this embodiment to prevent processing from becoming too large in answer generation and provide information more quickly.

[0056] The answer generating step preferably includes a procedure of generating an answer structure for a question by GPT and further generating answer elements included in the structure by GPT. This allows the information providing device 1 to generate an answer with a length that exceeds the upper limit of the number of tokens even if an upper limit is set for the number of tokens in answer generation.

[0057] It is preferable that the control unit 11 cooperates with the storage unit 12 and the communication unit 13 to execute the reading unit 113. At this time, the control unit 11 executes the process of step S5 which is a process related to the reading unit 113. This enables the information providing device 1 to provide accurate information for various questions in the field of the target service through voice dialogue.

[0058] [Step S5: Read the answer aloud] The control unit 11 cooperates with the storage unit 12 to execute a process of reading out the answer generated in step S4 by voice (step S5, reading step). The control unit 11 moves the process to step S6.

[0059] The procedure for reading out in the reading step is not particularly limited, and may be a procedure using a conventional voice synthesis technique. The procedure preferably includes a procedure using a voice synthesis model of parametric voice synthesis. This allows the information providing device 1 to perform voice synthesis that is more expressive and natural than a hidden Markov model. The procedure preferably includes a procedure using neural network voice synthesis, which is more expressive and natural than hidden Markov model voice synthesis.

[0060] The control unit 11 cooperates with the storage unit 12 and the communication unit 13 to execute the business data receiving unit 114. Then, the control unit 11 executes the process related to the business data receiving unit 114, that is, the process of step S6.

[0061] [Step S6: Determine whether business data has been provided] The control unit 11, in cooperation with the storage unit 12 and the communication unit 13, executes a process of determining whether business data has been provided from the terminal T or the like (step S6, business data provision determination step). If it is determined that the business data has been provided, the control unit 11 receives the provided business data, stores it in the storage unit 12, and moves the process to step S7. If it is not determined that the business data has been provided, the control unit 11 moves the process to step S11.

[0062] The "business data" in the business data provision determination step may be business data provided based on the response generated in step S4, and is not particularly limited. The business data includes, for example, data on business plans exemplified by a schedule, forecast financial statements, etc., data on effectiveness measurement exemplified by the results of market research, etc.

[0063] When business data is received, the control unit 11 cooperates with the storage unit 12 to execute the analysis unit 115. The control unit 11 then executes processing related to the analysis unit 115. Hereinafter, as an example of processing related to the analysis unit 115, a series of processing for generating a chart related to the analysis result from step S7 to step S8 and a series of processing for generating a quantitative evaluation from step S9 to step S10 will be described. The processing related to the analysis unit 115 is not limited to these processing, and includes various processing for analyzing business data. The processing includes, for example, a processing for making GPT generate a comment that compares general data of a similar business with the business data to be analyzed.

[0064] By executing the process related to the analysis unit 115, the information providing device 1 can receive business data provided based on a response to a request for providing business information generated by GPT, and generate an analysis result. That is, the information providing device 1 of this embodiment can provide information that further increases the adoption rate of subsidies, etc., which is the analysis result of business data to be included in an application for a subsidy, etc. In this way, the information providing device 1 can support the creation of an application for a subsidy, etc.

[0065] [Step S7: Determine whether to generate a chart] The control unit 11, in cooperation with the storage unit 12 and the communication unit 13, executes a process of determining whether or not to generate a chart related to the analysis results of the business data received in step S6 (step S7, chart generation determination step). If it is determined that a chart should be generated, the control unit 11 shifts the process to step S8. If it is determined that a chart should not be generated, the control unit 11 shifts the process to step S9.

[0066] The diagram generation determination step is not particularly limited, and may be, for example, a procedure of determining that a diagram should be generated when a user instructs the system to generate a diagram, or a procedure of determining that a diagram should be generated when the GPT generates a response based on business data indicating that a diagram can be generated.

[0067] [Step S8: Generate charts and graphs regarding the analysis results] The control unit 11, in cooperation with the storage unit 12, executes a process of generating a chart relating to the analysis results determined to be generated in step S7 (step S8, chart generation step). The control unit 11 moves the process to step S9.

[0068] The diagram generating step is not particularly limited and may include, for example, a procedure for generating a business data flow diagram, a business data chart, a business data graph, a business data table, or a business data infographic by various conventional techniques such as RPA.

[0069] By generating a chart according to the above-mentioned procedure, the information providing device 1 can generate a chart showing the analysis results of business data. That is, the information providing device 1 configured to execute an information providing process including a chart generation step can provide a function equivalent to a data analysis plug-in. That is, the information providing device 1 configured in this way can provide information that further increases the acceptance rate of subsidies, etc., such as a chart of the analysis results of business data to be included in an application for a subsidy, etc. In this way, the information providing device 1 can support the creation of an application for a subsidy, etc.

[0070] [Step S9: Determine whether to generate a quantitative evaluation] The control unit 11, in cooperation with the storage unit 12 and the communication unit 13, executes a process of determining whether to generate a quantitative evaluation to be included in the analysis result of the business data received in step S6 (step S9, quantitative evaluation generation determination step). If it is determined that a quantitative evaluation is to be generated, the control unit 11 shifts the process to step S10. If it is determined that a quantitative evaluation is not to be generated, the control unit 11 shifts the process to step S11.

[0071] The quantitative evaluation generation determination step is not particularly limited, and may be, for example, a procedure of determining that a quantitative evaluation should be generated when a user instructs that a quantitative evaluation should be generated, or a procedure of determining that a quantitative evaluation should be generated when the GPT generates a response based on business data indicating that a quantitative evaluation can be generated.

[0072] [Step S10: Generate quantitative evaluation] The control unit 11 cooperates with the storage unit 12 to execute a process of generating the quantitative evaluation determined to be generated in step S9 (step S10, quantitative evaluation generating step). The control unit 11 shifts the process to step S11.

[0073] The quantitative evaluation generation step is not particularly limited. For example, the step includes a procedure for generating a quantitative evaluation of the impact of a business on society or the environment by a conventional technology such as RPA. The step also includes a procedure for similarly generating various cash flows, real period profits and losses, etc. based on projected financial statements.

[0074] By generating a quantitative evaluation according to the above-mentioned procedure, the information providing device 1 can generate a quantitative evaluation to be included in the analysis result of the business data. Thus, the information providing device 1 configured to execute the quantitative evaluation generation step can provide a function equivalent to an impact evaluation plug-in. That is, the information providing device 1 configured in this way can provide information that further increases the adoption rate of subsidies, etc., that is, a quantitative evaluation of business data to be included in an application for a subsidy, etc. Thus, the information providing device 1 can support the creation of an application for a subsidy, etc.

[0075] The information provision process preferably includes a process of providing various information related to the subsidy application draft. In this process, the control unit 11 first executes the application draft receiving unit 116 in cooperation with the storage unit 12. Then, the control unit 11 executes the process of step S11 related to the application draft receiving unit 116.

[0076] [Step S11: Determine whether the subsidy application draft can be received] The control unit 11, in cooperation with the memory unit 12 and the communication unit 13, executes a process to determine whether a draft subsidy application form can be received from the terminal T or the like (step S11, draft application form reception determination step). If it is determined that the draft subsidy application form can be received, the control unit 11 receives the draft subsidy application form, stores it in the memory unit 12, and moves the process to step S12. If it is not determined that the draft subsidy application form can be received, the control unit 11 moves the process to the provision step. The draft subsidy application form is, for example, a draft subsidy application form etc. that has been drafted based on various information provided from the information providing device 1 of this embodiment.

[0077] Below, as an example of the process of providing various information executed when a subsidy application draft is received, the process of step S12 of generating a revision proposal for the subsidy application draft, the process of steps S13 to S14 of adding information on similar applications, and the process of steps S15 to S16 of generating a summary of the application draft will be described. The process of providing various information is not limited to these processes, but includes various processes of providing information on the subsidy application draft received in step S11. The process includes, for example, a process of having GPT generate a translation of the subsidy application draft.

[0078] In the process of step S12 for generating an amendment to the subsidy application proposal, the control unit 11 first executes the amendment generation unit 117 in cooperation with the storage unit 12. Then, the control unit 11 executes the process of step S12 related to the amendment generation unit 117.

[0079] [Step S12: Generate a revision proposal for the subsidy application] The control unit 11, in cooperation with the storage unit 12 and the communication unit 13, executes a process of generating an amendment to the subsidy application proposal received in step S11 in the GPT stored in the storage unit 12 or the like (step S12, amendment generation step). The control unit 11 shifts the process to step S13.

[0080] The revision proposal generation step is not particularly limited. For example, the revision proposal generation step is realized by a prompt such as "(※Subsidy application draft) Please list typos, grammar, and expression errors in this subsidy application draft and suggestions for their correction." However, in the above prompt, "(※Subsidy application draft)" is replaced by the subsidy application draft received in step S11.

[0081] In the creation of an application form that requires specialized knowledge on how to write the details, it may not be easy to create an application form that is written in an appropriate manner. The information providing device 1 configured to execute an information providing process including a revision proposal generation step can provide a function equivalent to a text proofreading plug-in. In other words, the information providing device 1 configured in this manner can provide information for revising an application form to an applicant who does not have specialized knowledge on how to write the details to make it written in an appropriate manner. In this way, the information providing device 1 can support the creation of applications for subsidies, etc.

[0082] In the process of steps S13 to S14 for adding information on similar applications, the control unit 11 first executes the analysis unit 115 in cooperation with the storage unit 12. Then, the control unit 11 executes the process of steps S13 to S14 related to the analysis unit 115.

[0083] [Step S13: Determine whether information on similar applications is stored] The control unit 11, in cooperation with the storage unit 12, executes a process of determining whether information on an adopted subsidy application (similar application) similar to the subsidy application proposal received in step S11 is stored in the application database (step S13, similar application determination step). If it is determined that the information is stored, the control unit 11 shifts the process to step S14. If it is not determined that the information is stored, the control unit 11 shifts the process to step S15.

[0084] The similar application determination step is not particularly limited. For example, the similar application determination step includes a procedure of executing a query to the application database to search for similar applications based on whether the business content and classification match or are similar to the subsidy application proposal received in step S11, and determining that one or more similar applications are stored when the query execution result includes the similar applications.

[0085] [Step S14: Adding information on similar applications] The control unit 11 cooperates with the storage unit 12 to execute a process of adding information on the similar application form determined to be stored in step S13 to the output (step S14, similar application form information adding step). The control unit 11 moves the process to step S15.

[0086] The similar application information providing step is not particularly limited, and may include, for example, a procedure of providing, to the output, one or more of the data of a part or all of the selected subsidy application and the URL of the selected subsidy application.

[0087] The information providing device 1 configured to execute the similar application information providing step can provide information that further increases the adoption rate of subsidies, etc., such as subsidy applications when subsidy applications for similar projects / subsidies have been adopted. In other words, the information providing device 1 configured in this manner allows the user to compare similar past applications and learn from successful applications.

[0088] In the process of generating a summary of the draft application from step S15 to step S16, the control unit 11 first executes the analysis unit 115 in cooperation with the storage unit 12. Then, the control unit 11 executes the process of steps S15 to S16 related to the analysis unit 115.

[0089] [Step S15: Determine whether to generate a summary of the application proposal] The control unit 11, in cooperation with the storage unit 12 and the communication unit 13, executes a process of determining whether to generate a summary of the application draft received in step S11 (step S15, summary generation determination step). If it is determined that a summary is to be generated, the control unit 11 shifts the process to step S16. If it is not determined that a summary is to be generated, the control unit 11 shifts the process to a provision step.

[0090] [Step S16: Generate a summary of the proposed application] The control unit 11, in cooperation with the storage unit 12 and the communication unit 13, executes a process of causing the GPT stored in the storage unit 12 or the like to generate a summary of the subsidy application proposal received in step S11 (step S16, summary generation step). The control unit 11 shifts the process to the provision step.

[0091] By using GPT trained on a large-scale corpus of text data, the information providing device 1 can generate a summary of a subsidy application draft regardless of the content of the business. The information providing device 1 can provide information useful for document creation, such as information in which a subsidy application draft written in a professional style is summarized in the form of an easy-to-read summary. In other words, the information providing device 1 configured to execute the summary generation step can provide an automatic summary function.

[0092] [Provision steps] After the processing of steps S1 to S16, the control unit 11 executes the providing unit 118 in cooperation with the storage unit 12 and the communication unit 13. Then, the control unit 11 executes a process (a providing step) related to the providing unit 118, which is a process of providing the information generated, assigned, etc. in these steps to the terminal T, etc. This allows the information providing device 1 of this embodiment to provide various information for document creation. Therefore, the information providing device 1 of this embodiment can support the creation of applications for subsidies, etc., by providing various information.

[0093] [Interactive feedback] The information provision process is preferably configured to include a process of providing feedback in real time and suggesting improvements when a user inputs a suggestion. This allows the information provision device 1 to provide an interactive feedback function. This process is realized, for example, by a procedure in which, when a question is input in a text box, the question is sent to the information provision device 1, and an instruction is given to the terminal T to display a Web page including a script configured to receive and display an answer to the question from the information provision device 1.

[0094] <Usage example> An example of how the information providing device 1 is used will be described below.

[0095] [Submit a question] First, a user who uses the information providing device 1 via the terminal T transmits a first question Q1 regarding the preparation of a subsidy application for the project "new software development", "Please let me know if there are any good subsidies for new software development", to the information providing device 1. The information providing device 1 receives the transmitted question.

[0096] [Answer Generation] Fig. 5 shows an example of a response that prompts the user to provide information, which is generated in the information provision process of this embodiment. The screen of the terminal T shown in Fig. 5 displays the transmitted question (first question Q1) and the responses (first answer A1 and second answer A2) generated in the information provision device 1.

[0097] In response to the first question Q1 "Please let me know if there are any good subsidies for new software development," the information providing device 1 generates a first answer A1 that suggests using the "Manufacturing Subsidy" administered by the Ministry of Economy, Trade and Industry. The information providing device 1 further generates a second answer A2 that prompts the provision of various information such as "Detailed explanation of the development content," "Objectives and effects of the development," "Development process plan," "Technical details of the development," "Financial information," and "Business operator information."

[0098] The second answer A2 encourages users to provide specific information required when applying for a subsidy for a software development project, by answering "Detailed description of development content: A detailed description of the software to be developed and the technical issues to be solved are required," "Objectives and effects of development: It is necessary to clearly explain what objectives and effects will be achieved through software development and prove the validity of the business plan," and "Technical details of development: Details of the technology to be used, the development environment and development tools, related patents and licenses, etc. are required." This allows users to include specific information in their subsidy application that may increase the chances of receiving the subsidy, even if they do not have specialized knowledge about subsidy applications.

[0099] In addition, the second answer A2 encourages users to provide specific business data required when applying for a subsidy for a software development business, by answering, "Development process plan: It is necessary to clearly plan the software development process, period, personnel, budget, etc., and to prove the progress management of the project," "Financial information: It is necessary to submit financial information based on the business plan (budget, investment amount, income and expenditure plan, etc.)," ​​"Business information: Information such as the business's corporate registration certificate, list of officers, number of employees, and capital is required," and "Other necessary documents: There are documents required for the application (application form, business plan, development plan, request for materials, etc.)." This allows users to include specific business data in their subsidy application that may increase the chance of receiving the subsidy, even if they do not have specialized knowledge about subsidy applications.

[0100] [Provision of information and data] The user inputs information regarding "detailed description of the development content," "purpose and effect of the development," "development process plan," "technical details of the development," etc., via input area I in Figure 5. Terminal T transmits the input information to information providing device 1. Information providing device 1 receives the transmitted information. The user also provides business data regarding "financial information," such as budget, investment amount, and income / expense plan, via input area I in Figure 5. Terminal T transmits the business data provided to information providing device 1. Information providing device 1 receives the transmitted business data.

[0101] [Generation of analysis results] The information providing device 1 generates an analysis result of the received business data. In addition, the information providing device 1 generates an analysis result including an analysis result in the form of a chart or the like and a quantitative evaluation of the impact of the business on society or the environment in response to an instruction from a user.

[0102] Fig. 6 is an example of an analysis result generated in the information provision process of this embodiment. The screen of the terminal T shown in Fig. 6 displays a second question Q2 commanding the generation of charts and quantitative evaluation, charts and the like generated by the information provision device 1 (first chart F1, second chart F2, third chart F3), and a third answer A3 including a quantitative evaluation generated in the information provision device 1.

[0103] The first chart F1 is a flowchart showing the flow of development in a new software development project. The information providing device 1 generates a flowchart from project data related to the flow of development, thereby supporting the creation of a subsidy application form in which the flow of development is visualized to increase the acceptance rate.

[0104] The second chart F2 is a Gantt chart showing a development schedule in a new software development project. The information providing device 1 generates a Gantt chart from project data related to the development schedule, thereby supporting the creation of a subsidy application form in which the development schedule is visualized to increase the adoption rate.

[0105] The third chart F3 is a graph showing the trend of forecast sales in a new software development business. The information providing device 1 generates a graph from business data related to the trend of forecast sales, thereby supporting the creation of a subsidy application form in which the trend of forecast sales is visualized to increase the acceptance rate.

[0106] The third answer A3 shown in FIG. 6 includes quantitative evaluations of environmental assessment, social responsibility assessment, life cycle assessment, and economic impact assessment. Since the third answer A3 includes a quantitative evaluation of environmental assessment, the information providing device 1 can support the creation of a subsidy application showing a quantitative evaluation of the impact on the environment, measures, environmental conservation costs, etc., evaluated in advance. Since the third answer A3 includes a quantitative evaluation of social responsibility assessment, the information providing device 1 can support the creation of a subsidy application showing a quantitative evaluation of the social and environmental efforts, policies, reports, etc., of a company, evaluated in advance. Since the third answer A3 includes a quantitative evaluation of life cycle assessment, the information providing device 1 can support the creation of a subsidy application showing a quantitative evaluation of the environmental load in the production, use, disposal, etc. phases of a product or service, evaluated in advance. Since the third answer A3 includes a quantitative evaluation of economic impact assessment, the information providing device 1 can support the creation of a subsidy application showing a quantitative evaluation of the economic effect, employment effect, etc., brought about by a project, evaluated in advance.

[0107] [Submitting a draft application] The user transmits a draft subsidy application form created based on the answer provided by information providing device 1 to information providing device 1 via terminal T. Information providing device 1 receives the transmitted draft subsidy application form. Information providing device 1 generates a revised draft of the received subsidy application form and transmits it to terminal T. The user checks the revised draft via terminal T and revises the draft subsidy application form.

[0108] The user instructs the information providing device 1 to provide a summary of the transmitted draft subsidy application form via terminal T. The information providing device 1 generates a summary of the transmitted draft subsidy application form and transmits it to terminal T. The user checks the summary via terminal T. This allows the user to ascertain whether the draft application form is as intended by checking the summary. If the draft application form is different from what the user intended, the user can use the summary to improve the draft application form.

[0109] It should be noted that within the scope of the concept of the present invention, a person skilled in the art may come up with various modifications and alterations. It is understood that such modifications and alterations fall within the scope of the present invention. For example, even if a person skilled in the art appropriately adds, deletes, or modifies the design of the above-described embodiment, or adds, omits, or modifies the conditions of a process, these modifications and alterations are included in the scope of the present invention as long as they include the gist of the present invention. [Explanation of symbols]

[0110] S System 1 Information provision device 11 Control section 111 Question Reception Department 112 Answer generation part 113 Reading section 114 Business Data Reception Department 115 Analysis Department 116 Application Draft Reception Section 117 Amendment generator 118 Provision Department 12 Storage section 13. Communications Department N Network T-Terminal

Claims

1. A question receiving unit configured to receive questions regarding the preparation of a grant application form used to apply for a grant, subsidy, or other benefit (subsidy) for a business; an answer generation unit configured to generate an answer to the question, which encourages the provision of information about the business (the question or business details extracted from the question), in a language model including knowledge about business information to be written in an application form for a subsidy for the business; A communication unit that provides the generated answer to a terminal; a business data receiving unit configured to receive the business information, the business data being provided by the terminal based on the response; an analysis unit configured to generate an analysis result for the business data; an application proposal receiving unit configured to receive a subsidy application proposal drafted based on the analysis result; an amendment generator configured to generate an amendment to the grant application; Further equipped with Further comprising an application database in which information on adopted subsidy applications is stored; The information providing device, wherein the analysis unit is configured to generate analysis results including information on accepted subsidy applications that are similar in business content to the proposed subsidy application.

2. The information providing apparatus according to claim 1 , wherein the analysis unit is configured to generate an analysis result including a chart.

3. The information providing device according to claim 1 , wherein the analysis unit is configured to generate an analysis result including a quantitative evaluation of an impact of the business on society or the environment.

4. The information providing apparatus of claim 1 , wherein the analyzer is configured to generate a summary of the proposed grant application.

Citation Information

Patent Citations

  • Method and system for supporting budget request and / or grant application concerning public sewer business

    JP2006004022A

  • Management support device, management support program and management support method

    JP2018067162A