Information processing system, information processing method, and information processing program

Abstract

To stably display high-quality proposals even if information input by a user is varied. [Solution] The information processing system 1 accepts input of analysis target information to be analyzed, accepts a proposal instruction from the user which is either a guarantee proposal instruction to analyze the analysis target information and make a proposal regarding guarantee to the organization, or an advice proposal instruction to analyze the analysis target information and make a proposal regarding advice to the organization, and generates display information that displays the proposal information generated by analyzing the analysis target information in accordance with the proposal instruction.

Description

[Technical Field]

[0001] The present invention relates to an information processing system, an information processing method, and an information processing program for making proposals for assurance services (audits, inspections, ISO certification, and other assurance services) or advisory services (advisory and consulting services) for organizations. In particular, the present invention relates to a mechanism for realizing integrated optimization of risk assessment and proposal generation by utilizing a trained model (described later). [Background technology]

[0002] 2. Description of the Related Art Conventionally, there are audit support devices that accept input of information such as company information from a user and support audit work based on the accepted information.

[0003] For example, Patent Publication No. 2022-115543 discloses an audit support device that accepts input of corporate information indicating the characteristics of a client's company, determines questionnaire items related to the audit and items to be submitted based on the corporate information, accepts responses to the questionnaire items and submission of items from the client, determines interview items related to the audit based on the corporate information, responses, and items, accepts input of interview results related to the interview items from a consultant, and generates a draft audit report based on the interview results. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-115543 Summary of the Invention [Problem to be solved by the invention]

[0005] However, conventional audit support systems have been limited to collecting information from audited companies and streamlining audit procedures using checklists and questionnaires. In particular, in actual audit work, clients possess a wide variety of information and have diverse needs. Therefore, conventional audit support systems may provide low-quality proposals that do not meet the client's needs, depending on the information input by the user. Furthermore, in actual audit work, clients may be required to provide advice in addition to audits. In particular, assurance work, including audits, may require analysis and proposals based on a variety of information. Therefore, conventional audit support systems may have difficulty providing proposals that meet the client's needs or are of high quality.

[0006] Therefore, an embodiment of the present invention aims to provide an information processing system, an information processing method, and an information processing program that can display high-quality suggestions even if the information input by the user is diverse. [Means for solving the problem]

[0007] The information processing system according to the embodiment includes: Purpose and scope reception department, risk information reception department, risk assessment department, The analysis information receiving unit, the proposal instruction unit, the proposal generating unit, and the display generating unit are included. The purpose and scope receiving unit receives purpose and scope information, including input of key stakeholders to be analyzed, input of the results and / or goals expected by the key stakeholders, and input of physical scope, and inputs the purpose and scope information into a large-scale language model so that it can be identified in natural language. The risk information receiving unit receives risk information, including setting of risk categories including at least one of external environment risk, internal environment risk, and operational risk, input of text information of risks corresponding to the risk categories, and setting of risk levels corresponding to the risk categories, and inputs the risk information into the large-scale language model so that it can be identified in natural language. The risk assessment unit instructs the large-scale language model to analyze the purpose and scope information and the risk information to generate a risk assessment, including key risks, a basic assessment, and an interdependency assessment. The analysis information receiving unit receives input of analysis target information to be analyzed. and inputting the analysis target information into the large-scale language model so that the information can be identified by natural language. The proposal instruction unit receives, from a user, a proposal instruction that is either a guarantee proposal instruction for analyzing the analysis target information and making a proposal regarding guarantee to the organization, or an advice proposal instruction for analyzing the analysis target information and making a proposal regarding advice to the organization. and inputting the suggested instruction into the large-scale language model so that the suggested instruction can be specified in natural language. . The proposal generation unit The above-mentioned proposal instructions the objectives and scope information, the significant risks, and the risk assessment Analyze the analysis target information according to Let us know your suggestions Generate The display generation unit issues an instruction to the large-scale language model to Proposal information , obtained from the large-scale language model Generate display information to be displayed.

[0008] The information processing method according to the embodiment includes: A computer executes a plurality of processes, the computer including a process of receiving purpose and scope information including an input of a key stakeholder to be analyzed, an input of the results and / or objectives expected by the key stakeholder, and an input of a physical scope, and inputting the purpose and scope information into a large-scale language model so that the purpose and scope information can be specified in natural language, a process of receiving risk information including a setting of risk categories including at least one of external environment risk, internal environment risk, and operational risk, an input of text information of risks according to the risk categories, and a setting of risk levels according to the risk categories, and inputting the risk information into the large-scale language model so that the risk information can be specified in natural language, and a process of issuing an instruction to the large-scale language model to analyze the purpose and scope information and the risk information and generate a risk assessment including significant risks, a basic assessment, and an interdependency assessment. Accepts input of information to be analyzed and inputting the analysis target information into the large-scale language model so that the information can be identified by natural language.and receiving a proposal instruction from the user, which is either an assurance proposal instruction for analyzing the analysis target information and making a proposal regarding assurance to the organization, or an advice proposal instruction for analyzing the analysis target information and making a proposal regarding advice to the organization. inputting the suggested instruction into the large-scale language model so that the suggested instruction can be identified in natural language; The above-mentioned proposal instructions the objectives and scope information, the significant risks, and the risk assessment The analysis target information is analyzed according to Proposal information Generate a process of issuing an instruction to the large-scale language model to The proposed information , obtained from the large-scale language model Generate display information to display Processing and executing .

[0009] The information processing program according to the embodiment includes: Code for the purpose and scope reception section, code for the risk information reception section, code for the risk assessment section, The code of the analysis information receiving unit, the code of the proposal instruction unit, the code of the proposal generating unit, and the code of the display generating unit are executed by a computer. The code for the purpose and scope receiving unit receives purpose and scope information, including input of key stakeholders to be analyzed, input of the outcomes and / or objectives expected by the key stakeholders, and input of physical scope, and inputs the purpose and scope information into a large-scale language model so that it can be identified in natural language. The code for the risk information receiving unit receives risk information, including setting of risk categories including at least one of external environmental risk, internal environmental risk, and operational risk, input of text information of risks corresponding to the risk categories, and setting of risk levels corresponding to the risk categories, and inputs the risk information into the large-scale language model so that it can be identified in natural language. The code for the risk assessment unit instructs the large-scale language model to analyze the purpose and scope information and the risk information and generate a risk assessment, including significant risks, a basic assessment, and an interdependency assessment. The code for the analysis information receiving unit receives input of analysis target information to be analyzed. and inputting the analysis target information into the large-scale language model so that the information can be identified by natural language. The code of the proposal instruction unit receives from a user a proposal instruction that is either a guarantee proposal instruction for analyzing the analysis target information and making a proposal regarding guarantee to the organization, or an advice proposal instruction for analyzing the analysis target information and making a proposal regarding advice to the organization. and inputting the suggested instruction into the large-scale language model so that the suggested instruction can be specified in natural language. . The code for the suggestion generator may include: The above-mentioned proposal instructions the objectives and scope information, the significant risks, and the risk assessment Analyze the analysis target information according to Let us know your suggestions Generate The display generator code instructs the large-scale language model to: Proposal information , obtained from the large-scale language model Generate display information to be displayed. [Effects of the Invention]

[0010] The embodiments of the present invention can provide an information processing system, an information processing method, and an information processing program that can display high-quality suggestions even if the information input by the user is varied. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram illustrating an example of the overall configuration of an information processing system according to an embodiment. [Figure 2]1 is a block diagram illustrating an example of an internal configuration of an information processing system according to an embodiment. [Figure 3] FIG. 2 is a block diagram illustrating an example of a storage unit of a server in the information processing system according to the embodiment. [Figure 4] FIG. 10 is a diagram showing an example of an image for inputting a suggestion instruction in the information processing system according to the embodiment. [Figure 5A] FIG. 10 is a diagram showing an example of an image for accepting input of main stakeholders from among purpose and scope information in the information processing system according to the embodiment. [Figure 5B] FIG. 10 is a diagram showing an example of an image for receiving input of results and / or objectives expected by major stakeholders from among the objective and scope information in the information processing system according to the embodiment. [Figure 5C] FIG. 10 is a diagram showing an example of an image for accepting input of a physical range of purpose / scope information in the information processing system according to the embodiment. [Figure 6A] FIG. 10 is a diagram showing an example of an image for accepting risk information including setting a risk category designated as an external environment risk, inputting a risk, and setting a risk level in an information processing system according to an embodiment. [Figure 6B] A figure showing an example of an image for accepting risk information including setting a risk category designated as an internal environment risk, inputting a risk, and setting a risk level in an information processing system according to an embodiment. [Figure 6C] A figure showing an example of an image that accepts risk information including setting a risk category designated as an operational risk, inputting the risk, and setting the risk level in an information processing system according to an embodiment. [Figure 7A] 10 is a diagram showing an example of an image for accepting settings for a risk assessment including a basic assessment in the information processing system according to the embodiment. FIG. [Figure 7B] FIG. 10 is a diagram showing an example of an image for accepting settings for risk assessment including interdependency assessment in the information processing system according to the embodiment. [Figure 8]FIG. 10 is a diagram showing an example of an image for instructing whether to make settings related to warranty proposals or settings related to advice proposals in the information processing system according to the embodiment. [Figure 9] FIG. 10 is a diagram showing an example of an image for accepting settings for an audit purpose among settings related to warranty proposals in the information processing system according to the embodiment. [Figure 10A] 10 is a diagram showing an example of an image for accepting selection of a data collection method for an audit method among settings related to an assurance proposal in the information processing system according to the embodiment; FIG. [Figure 10B] 10 is a diagram showing an example of an image for accepting selection of an analysis method of an audit method among settings related to an assurance proposal in the information processing system according to the embodiment; FIG. [Figure 10C] 10 is a diagram showing an example of an image for accepting selection of a verification method for an audit method among settings related to an assurance proposal in the information processing system according to the embodiment; FIG. [Figure 11] 10 is a diagram showing an example of an image for accepting settings for audit procedures among settings related to assurance proposals in the information processing system according to the embodiment; FIG. [Figure 12] 10 is a diagram showing an example of an image for accepting a setting of an advice goal among settings related to advice proposals in the information processing system according to the embodiment; FIG. [Figure 13] 10 is a diagram showing an example of an image for accepting selection of a solution area from among settings related to advice proposals in the information processing system according to the embodiment; FIG. [Figure 14] 10 is a diagram showing an example of an image for accepting settings for a tentative advice plan among settings related to advice proposals in the information processing system according to the embodiment; FIG. [Figure 15] This figure shows an example of an image in which display information is output in an information processing system related to an embodiment, displaying proposed information including information to be analyzed, important risks, risk assessment, audit objectives, expected controls, audit key points, audit methods, and recommended audit procedures. [Figure 16]FIG. 10 is a diagram showing an example of an image outputted from an information processing system relating to an embodiment, in which display information is displayed, displaying proposed information including information to be analyzed, important risks, risk assessment, advice goals, solution areas, points of focus (one of objective achievement level (KPI), risk management (key risk indicators), success factors (CSF), internal control perspective (control key points), sustainability perspective (EHS / ESG indicators), etc., or a combination of two or more of these), and a tentative advice plan. [Figure 17] This is a flowchart showing an example of pre-processing based on the code of the purpose / scope receiving unit, risk information receiving unit, risk assessment unit, setting instruction unit, audit objective receiving unit, perspective information receiving unit, audit method receiving unit, audit procedure receiving unit, advice goal receiving unit, solution area receiving unit, interim advice plan receiving unit, and organizational information receiving unit in an information processing system related to an embodiment. [Figure 18] 10 is a flowchart illustrating an example of a series of operations based on code of an analysis information receiving unit, a proposal instruction switching unit, a proposal instruction unit, a proposal generation unit, and a display generation unit in the information processing system according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] (Configuration of the embodiment) Hereinafter, an embodiment will be described with reference to the drawings.

[0013] FIG. 1 is a block diagram showing an example of the overall configuration of an information processing system 1. As shown in FIG.

[0014] 1, the information processing system 1 includes a terminal device 10, a server 20, and a trained model 30. The terminal device 10 is connected to the server 20 via a network such as a LAN, the Internet, or mobile communications. The server 20 is connected to the trained model 30 via a network such as a LAN, the Internet, or mobile communications.

[0015] The terminal device 10 is used by a user such as an auditor in an organization such as a company, etc. The information processing system 1 is used to provide assurance and advice to the organization.

[0016] Assurance here refers to the act of systematically and objectively evaluating and verifying the consistency of a specific object (activity, process, deliverable, etc.) with predetermined standards and requirements, and providing reasonable assurance based on the results. Specifically, assurance includes the following acts: (1) Third-party evaluations such as audits, inspections, and reviews (2) Evaluation by the parties involved, such as self-inspection and self-assessment (3) Technical verification such as system testing and quality inspection (4) Institutional verification of compliance assessment, internal control assessment, etc. (5) Operational verification such as business process review and performance evaluation

[0017] Advice refers to the act of formulating and proposing solutions in a systematic and practical manner to achieve a target state or value for a specific subject (activity, process, deliverable, etc.), and supporting their implementation. Specifically, advice includes the following acts: (1) Providing direction for strategic planning, planning, etc. (2) Providing specific solutions such as process improvement and system implementation (3) Support for organizational reform, human resource development, and other reforms (4) Support for risk response, problem solving, etc. (5) Support for improvements and innovations such as performance improvement and value creation

[0018] FIG. 2 is a block diagram showing an example of the internal configuration of the information processing system 1. As shown in FIG.

[0019] 2, the terminal device 10 is an information terminal such as a personal computer or a smartphone. Based on an instruction input by a user, the terminal device 10 transmits analysis target information to be analyzed to the server 20. The terminal device 10 has a control unit 11, a storage unit 12, a display unit 13, and an instruction input unit 14.

[0020] The control unit 11 has a processing unit such as a CPU, and executes a program read from the storage unit 12 to realize its functions.

[0021] The storage unit 12 has storage elements such as a ROM, a RAM, an HDD, and an SSD, and stores various information for controlling the terminal device 10 and various programs that constitute the processing unit.

[0022] The display unit 13 is configured with an LCD, an LED, an OLED, or the like, and is capable of displaying various types of information input from the control unit 11.

[0023] The instruction input unit 14 has input devices such as a keyboard and a mouse, and outputs to the control unit 11 various instruction inputs input by the user.

[0024] The server 20 includes a control unit 21 and a storage unit 22 .

[0025] The control unit 21 has a processing device such as a CPU, and realizes its functions by executing a program read from the storage unit 22. The control unit 21 also inputs various information to the trained model 30, and receives output generated by the trained model 30 after analyzing the input information.

[0026] The storage unit 22 has storage elements such as a ROM, a RAM, an HDD, and an SSD, and stores various programs that constitute a processing unit for controlling the server 20, and various information used by the programs.

[0027] 3, for example, the memory unit 22 stores programs for an analysis information receiving unit P1, a proposal instruction switching unit P2, a proposal instruction unit P3, a purpose / scope receiving unit P4, a risk information receiving unit P5, a risk assessment unit P6, a setting instruction unit P7, an audit objective receiving unit P8, a viewpoint information receiving unit P9, an audit method receiving unit P10, an audit procedure receiving unit P11, an advice goal receiving unit P12, a solution area receiving unit P13, a tentative advice plan receiving unit P14, an organization information receiving unit P15, a proposal generation unit P16, and a display generation unit P17. The memory unit 22 also stores a knowledge base DB.

[0028] The trained model 30 is generated so that a machine learning model performs machine learning in accordance with a model learning program. The trained model 30 is, for example, a parameterized composite function that combines multiple functions and is defined by a combination of multiple adjustable functions and parameters. The trained model 30 performs natural language processing, for example, by analyzing input text information and extracting information that means a question. The trained model 30 can be realized by a model that uses Transformer technology that performs natural language processing. The trained model 30 is realized by, for example, GPT-4. The trained model 30 is trained by, for example, supervised learning. The trained model 30 is, for example, a general-purpose model.

[0029] FIG. 4 is a diagram showing an example of an image (suggestion instruction receiving image A) for receiving a suggestion instruction from a user.

[0030] 2 and 4, the proposal instruction acceptance image A is displayed on the display unit 13. The analysis information acceptance unit P1 accepts input of analysis target information on the proposal instruction acceptance image A.

[0031] As shown in FIG. 4, for example, when the user operates the instruction input unit 14 and inputs into the input area A1 of the suggestion instruction reception image A, the analysis information reception unit P1 receives text information that is the information to be analyzed.

[0032] The analysis information receiving unit P1 adds the received analysis target information to a predetermined formula, and inputs the result into the trained model 30. The formula is not particularly limited, but may be defined in natural language to identify the analysis target information, such as "The analysis target information is as follows." The trained model 30 temporarily stores the information input here as analysis target information and can use it for subsequent processing as appropriate. Although not explained further below, all or part of the information input into the trained model 30 can also be used for subsequent processing as appropriate.

[0033] As shown in FIGS. 2 and 4, the proposal instruction switching unit P2 can switch the proposal instruction on the proposal instruction reception image A to either a guarantee proposal instruction or an advice proposal instruction.

[0034] As shown in Figure 4, for example, when a user operates the instruction input unit 14 to check either the "guarantee mode" button A21 or the "advice mode" button A22 in the proposal instruction reception image A, the proposal instruction switching unit P2 switches the proposal instruction to a guarantee proposal instruction or an advice proposal instruction corresponding to the checked button.

[0035] The suggestion instruction switching unit P2 inputs the switching result of the suggestion instruction to the trained model 30 in natural language.

[0036] As shown in Figure 4, for example, when a user operates the instruction input unit 14 to check the "Start analysis" button A3 on the proposal instruction acceptance image A, the proposal instruction unit P3 accepts either a guarantee proposal instruction or an advice proposal instruction switched by the proposal instruction switching unit P2.

[0037] The guarantee proposal instruction is a proposal instruction to analyze the analysis target information input by the analysis information receiving unit P1 and make a proposal regarding guarantee to the organization. The advice proposal instruction is a proposal instruction to analyze the analysis target information input by the analysis information receiving unit P1 and make a proposal regarding advice to the organization.

[0038] The proposal instruction unit P3 inputs the guarantee proposal instruction or advice proposal instruction received from the user into the trained model 30.

[0039] Here, the preliminary settings for performing the analysis will be described.

[0040] Figure 5A is a diagram showing an example of an image (first image for receiving purpose / scope information B1) that accepts input from the main stakeholders to be analyzed among the purpose / scope information. Figure 5B is a diagram showing an example of an image (second image for receiving purpose / scope information B2) that accepts input from the main stakeholders regarding the results and / or objectives they expect among the purpose / scope information. Figure 5C is a diagram showing an example of an image (third image for receiving purpose / scope information B3) that accepts the physical scope among the purpose / scope information.

[0041] 2 and 5A to 5C, objective / scope information reception images B1 to B3 are displayed on the display unit 13. The objective / scope reception unit P4 receives objective / scope information on the objective / scope information reception images B1 to B3.

[0042] Purpose and scope information includes key stakeholders, their expected outcomes and / or objectives, and the physical scope.

[0043] 5A, for example, when a user operates the instruction input unit 14 to input text information into the input field B11 of the first purpose / scope information reception image B1 and then checks the "Send response" button B12 on the first purpose / scope information reception image B1, the purpose / scope reception unit P4 receives this text information and the major stakeholders included therein. Major stakeholders include, for example, requests from management, expectations from related departments, and the interests of external stakeholders.

[0044] As shown in FIG. 5B, for example, when the user operates the instruction input unit 14 to input text information into the input area B21 of the second purpose / scope information reception image B2, and then checks the "Send response" button B22 of the second purpose / scope information reception image B2, the purpose / scope reception unit P4 accepts this text information and the results and / or objectives expected by the major stakeholders contained therein.

[0045] 5C, for example, when the user operates the instruction input unit 14 to input text information into the input field B31 of the third purpose / scope information reception image B3 and then checks the "Send Answer" button B32 on the third purpose / scope information reception image B3, the purpose / scope reception unit P4 accepts the text information and the physical range included therein. The physical range includes, for example, a geographic range, an organizational range, a temporal range, etc.

[0046] The purpose and scope receiving unit P4 adds the received purpose and scope information to a predetermined boilerplate phrase and inputs it to the trained model 30. The boilerplate phrase is defined in natural language so that the purpose and scope information can be input to the trained model 30 according to the input area that received the text information, such as "The main stakeholders are as follows" in the case of the input area B11.

[0047] Fig. 6A is a diagram showing an example of an image (first risk information receiving image C1) for receiving risk information including the setting of a risk category determined to be an external environmental risk. Fig. 6B is a diagram showing an example of an image (second risk information receiving image C2) for receiving risk information including the setting of a risk category determined to be an internal environmental risk. Fig. 6C is a diagram showing an example of an image (third risk information receiving image C3) for receiving risk information including the setting of a risk category determined to be an operational risk.

[0048] 2 and 6A to 6C, first to third risk information reception images C1 to C3 are displayed on the display unit 13. The risk information reception unit P5 receives risk information on the risk information reception images C1 to C3. In addition to setting a risk category, the risk information includes input of a risk according to the set risk category and setting of a risk level.

[0049] As shown in Figures 6A to 6C, for example, when the user operates the instruction input unit 14 to check either the ``External Environment Risk'' button C11, the ``Internal Environment Risk'' button C21, or the ``Operational Risk'' button C31 in each of the first to third risk information reception images C1 to C3, the risk information reception unit P5 accepts the selection of the risk category corresponding to the checked button and then proceeds to setting that risk category.

[0050] For example, when the user operates the instruction input unit 14 to input text information into the input areas C12, C22, and C32 of the first to third risk information reception images C1 to C3 according to the selected risk category, the risk information reception unit P5 becomes ready to accept this text information and the risks contained therein.

[0051] For example, when the user operates the instruction input unit 14 to check any of the "High Risk" buttons C131, C231, C331, "Medium Risk" buttons C132, C232, C332, "Low Risk" buttons C133, C233, C333, or "Not Applicable" buttons C134, C234, C334 of the first to third risk information reception images C1 to C3, depending on the selected risk category, the risk information reception unit P5 becomes ready to accept the checked risk level.

[0052] Here, the risk levels are set in descending order of "high risk," "medium risk," and "low risk." If the risk level setting is not suitable for risk assessment, the risk level is set to "not applicable."

[0053] Furthermore, for example, when the user operates the instruction input unit 14 to check the "Perform analysis" buttons C14, C24, C34 of the first to third risk information reception images C1 to C3, the risk information reception unit P5 receives the risk information set as described above.

[0054] The risk information receiving unit P5 adds the content entered in the first to third risk information receiving images C1 to C3 to predetermined standard phrases and inputs the result into the trained model 30. For example, if the button C11 is checked, the text information entered into the input area C12 is added to a standard phrase such as "External environment risks are as follows," and the result is input into the trained model 30. If the "High Risk" button C131 is checked, a standard phrase such as "The risk level is high" is input into the trained model 30.

[0055] 7A is a diagram showing an example of an image (first risk assessment reception image D1) for receiving settings for risk assessment including basic assessment, and FIG. 7B is a diagram showing an example of an image (second risk assessment reception image D2) for receiving settings for risk assessment including interdependency assessment.

[0056] 2, 7A, and 7B, the first and second risk assessment reception images D1 and D2 are displayed on the display unit 13. The risk assessment unit P6 performs risk assessment on the first and second risk assessment reception images D1 and D2.

[0057] 7A and 7B, the risk assessment includes a basic assessment and an interdependency assessment. For example, when the user operates the instruction input unit 14 to check either the "basic assessment" button D11 or the "interdependency assessment" button D21 of the first and second risk assessment reception images D1 and D2, the risk assessment corresponding to the checked button is displayed.

[0058] When the "Basic Evaluation" button D11 is checked, the risk evaluation unit P6 instructs the trained model 30 to evaluate each of the external environment risks, internal environment risks, and operational risks received by the risk information receiving unit P5 using predetermined phrases. The evaluation results output from the trained model 30 are displayed and set in the output area D12.

[0059] Furthermore, when the "Interdependency Assessment" button D21 is checked, the risk assessment unit P6 instructs the trained model 30 to assess interdependency using a predetermined formula. The assessment results output from the trained model 30 are displayed and set in the output area D22.

[0060] As shown in Figure 7A, the basic assessment includes, for example, an assessment of the likelihood, impact, and velocity of a risk. As shown in Figure 7B, the interdependency assessment includes, for example, an assessment of risk cascading, amplification, and hidden factors. Boilerplate can be defined to include these assessments.

[0061] The risk assessment unit P6 can output the major risks and risk assessments generated by analyzing the purpose and scope information and risk information through the basic assessment and interdependency assessment. The proposal generation unit P16, which will be described later, analyzes the risk information through the basic assessment and interdependency assessment and generates the major risks and risk assessments.

[0062] FIG. 8 is a diagram showing an example of an image (setting instruction image E) for instructing whether to make settings related to a guarantee proposal or settings related to an advice proposal.

[0063] 2 and 8, the setting instruction image E is displayed on the display unit 13. The setting instruction unit P7 makes it possible to switch between settings related to a guarantee proposal instruction and settings related to an advice proposal instruction on the setting instruction image E.

[0064] As shown in FIG. 8, for example, when the user operates the instruction input unit 14 to check either the "guarantee mode" button E1 or the "advice mode" button E2 of the setting instruction image E, the setting instruction unit P7 starts setting the suggested instruction for the mode corresponding to the checked button.

[0065] FIG. 9 is a diagram showing an example of an image (inspection purpose reception image F) for receiving the setting of the inspection purpose among the settings related to the warranty proposal.

[0066] 2 and 9, the inspection purpose reception image F is displayed on the display unit 13. The inspection purpose reception unit P8 receives, on the inspection purpose setting image F, the setting of the inspection purpose among the settings related to the assurance proposal.

[0067] As shown in Figure 9, for example, when a user operates the instruction input unit 14 to input text information into the input area F1 of the audit purpose setting image F, and then checks the "Perform AI analysis" button F3 of the audit purpose setting image F, the audit purpose receiving unit P8 accepts the setting of this text information, which is the audit purpose.

[0068] The audit purpose receiving unit P8 adds text information to a predetermined fixed phrase and inputs it into the trained model 30. The fixed phrase is defined to instruct the setting of the audit purpose, for example, "The audit purpose is as follows."

[0069] Furthermore, the viewpoint information receiving unit P9 receives, on the inspection purpose setting image F, the viewpoint information setting among the settings related to the assurance proposal.

[0070] 9, the viewpoint information is set to include at least one of a compliance approach, an internal leadership assessment approach, and a risk-based approach. Furthermore, although not specifically shown, the viewpoint information can also be set to include at least one of a compliance approach, an internal leadership assessment approach, a risk-based approach, a strategic alignment approach, and a social evaluation approach.

[0071] As shown in Figure 9, along with setting the audit objective, for example, by operating the user's instruction input unit 14, at least one of the ``Compliance Approach'' button F21, the ``Internal Comprehensive Evaluation Approach'' button F22, and the ``Risk-Based Approach'' button F23 in the audit objective setting image F is checked, and further, when the ``Perform AI Analysis'' button F3 in the audit objective setting image F is checked, the perspective information receiving unit P9 receives the perspective information corresponding to the checked button.

[0072] The viewpoint information receiving unit P9 inputs a set phrase for indicating viewpoint information to the trained model 30 in response to the checked button.

[0073] The perspective information will be explained in detail below. In the compliance approach, the audit scope is the "individual business level," the key areas are "confirmation of compliance with laws and regulations," "prevention of fraud," etc., the evaluation target is "confirmation of compliance with laws and regulations," and the main stakeholder is the "person in charge of operations." In the compliance approach, the perspective information can be input to the trained model 30 using boilerplate phrases that indicate attention to the audit scope, compliance with laws and regulations, compliance risks, and fraud prevention perspectives.

[0074] In the internal control approach, the audit scope is the "process level," the focus area is "assessment and improvement of existing internal controls," the evaluation target is "effectiveness of internal controls," and the main stakeholder is "management." In the internal control approach, viewpoint information can be input into the trained model 30 using standard phrases that direct attention to these audit scopes, the state of development of internal controls, the effectiveness of the operational status, and the appropriateness of control activities.

[0075] In the risk-based approach, the audit scope is the "department / function level," the focus area is "identification and evaluation of significant risks," the evaluation target is "effectiveness of significant risk management," and the main stakeholders are "department heads and functional managers (CxOs)." In the risk-based approach, viewpoint information can be input into the trained model 30 using standard phrases that focus on these audit scopes, identification of significant risks, appropriateness of risk responses, and effectiveness of risk management processes.

[0076] In the strategic alignment approach, the audit scope is the "enterprise level," the focus area is "enterprise risk management aligned with mid- to long-term strategy," the evaluation target is "effectiveness of enterprise risk management aligned with the organization's mid- to long-term strategy," and the key stakeholder is "management." In the strategic alignment approach, perspective information can be input into the trained model 30 using boilerplate phrases that direct attention to these audit scopes, alignment with strategic goals, effectiveness of enterprise risk management, and mid- to long-term impacts.

[0077] In the social impact approach, the audit scope is "institutional efforts that take social systems into consideration," the focus area is "assessment of the organization's purpose (raison d'être) and social impact," the evaluation target is "the organization's purpose and the effectiveness of social impact management," and the main stakeholders are "the board of directors and society as a whole." In the social impact approach, perspective information can be input into the trained model30 using phrases that indicate attention to these audit scopes, alignment with the organization's purpose, the scope and extent of social impact, contribution to sustainability, impact on stakeholders, and long-term social value creation.

[0078] Even when the suggestion instruction is an advice suggestion instruction, viewpoint information can be input to the trained model 30 from the same viewpoint as above. Also, when the suggestion instruction is an advice suggestion instruction, viewpoint information different from the above can be input, which will be described later.

[0079] Fig. 10A is a diagram showing an example of an image (first inspection method reception image G1) that accepts the selection of a data collection method for the inspection method among the settings related to the warranty proposal. Fig. 10B is a diagram showing an example of an image (second inspection method reception image G2) that accepts the selection of an analysis method for the inspection method among the settings related to the warranty proposal. Fig. 10C is a diagram showing an example of an image (third inspection method reception image G3) that accepts the selection of a verification method for the inspection method among the settings related to the warranty proposal.

[0080] 2 and 10A to 10C, the first to third inspection method reception images G1 to G3 are displayed on the display unit 13. The inspection method reception unit P10 receives the selection of an inspection method from among the settings related to the warranty proposal on the first to third inspection method reception images G1 to G3.

[0081] Audit methods are selected from at least one of the following, or a combination of two or more of the following: inquiry / interview, inspection of documents, physical inspection (verification of tangible assets), survey, data analysis, process analysis, control testing, quality assessment, walkthrough, flowchart analysis, observation, inspection / verification, cross-checking, recalculation, re-execution, confirmation, tracing, and analytical procedures.

[0082] As shown in Figures 10A to 10C, for example, when the user operates the instruction input unit 14 and checks any of the "Data Collection Method" button G11, "Analysis Method" button G21, and "Verification Method" button G31 in the first to third inspection method reception images G1 to G3, settings are made for the "Data Collection Method," "Analysis Method," or "Verification Method" corresponding to the checked button.

[0083] As shown in FIG. 10A, for example, when the user operates the instruction input unit 14 to select a data collection method, at least one of the buttons displayed as the "Question / Interview" button G12, the "Document Viewing" button G13, the "Observation" button G14, and the "Survey" button G15 in the first inspection method reception image G1 is checked, and further, when the "Complete Selection" button G16 in the first inspection method reception image G1 is checked, the inspection method reception unit P10 will accept the setting of the data collection method corresponding to the checked button.

[0084] As shown in FIG. 10B, for example, when the user operates the instruction input unit 14 to select an analysis method, at least one of the "Data Analysis" button G22, the "Process Method" button G23, and the "Control Test" button G24 on the second inspection method reception image G2 is checked, and further, when the "Selection Complete" button G25 on the second inspection method reception image G2 is checked, the inspection method reception unit P10 will accept the setting of the analysis method corresponding to the checked button.

[0085] As shown in FIG. 10C, for example, when the user operates the instruction input unit 14 to select a verification method, at least one of the "Sampling Verification" button G32, the "Recalculation" button G33, and the "Verification" button G34 on the third inspection method reception image G3 is checked, and further, when the "Complete Selection" button G35 on the third inspection method reception image G3 is checked, the inspection method reception unit P10 will accept the setting of the verification method corresponding to the checked button.

[0086] The audit method receiving unit P10 inputs predetermined phrases indicating the audit method selection results by the user into the trained model 30. For example, when buttons G11 and G12 are checked, the phrases are defined so that the audit method selection results can be input into the trained model 30, such as "As a result of the audit method selection, questionnaires and interviews have been selected as the data collection method."

[0087] FIG. 11 is a diagram showing an example of an image (audit procedure reception image H) for receiving the setting of the audit procedure among the settings related to the assurance proposal.

[0088] 2 and 11, the audit procedure reception image H is displayed on the display unit 13. The audit procedure reception unit P11 receives detailed settings of the audit procedure, among the settings related to the assurance proposal, on the audit procedure reception image H.

[0089] As shown in Fig. 11, for example, when the "Add Procedure" button H1 in the audit procedure reception image H is checked by the user operating the instruction input unit 14, a button corresponding to the audit procedure is added to the "List of Audit Procedures" in the audit procedure reception image H. For example, a button H2 for "Evaluate the appropriateness of system access privileges" and a button H3 for "Confirm compliance with purchase approval process" are added to the "List of Audit Procedures" in the audit procedure reception image H.

[0090] For example, when a user operates the instruction input unit 14 and checks a button corresponding to an audit procedure, the user can set the details of that audit procedure.

[0091] For example, when the user operates the instruction input unit 14 to check the "Evaluate the appropriateness of system access privileges" button H2, selects the "Category" drop-down list H21 in the audit procedure reception image H, inputs "Procedure title" into the input field H22, inputs "Purpose" into the input field H23, inputs "Implementation method" into the input field H24, inputs "Evaluation criteria" into the input field H25, and inputs "Required evidence" into the input field H26, the audit procedure reception unit P11 will accept the detailed settings of this audit procedure.

[0092] The audit procedure receiving unit P11 adds detailed settings of the audit procedure made by the user to a predetermined standard phrase and inputs the result into the trained model 30. The standard phrase is defined so that detailed settings of the audit method can be input into the trained model 30, for example, when the button H2 is checked and "on-site procedure" is selected in the drop-down list H21, such as "The evaluation of the appropriateness of system access privileges will be determined in the following on-site investigation."

[0093] FIG. 12 is a diagram showing an example of an image (advice goal receiving image I) for receiving the setting of an advice goal among the settings related to advice suggestions.

[0094] 2 and 12, the advice goal acceptance image I is displayed on the display unit 13. The advice goal acceptance unit P12 accepts, on the advice goal acceptance image I, the setting of the advice goal among the settings related to the advice proposal.

[0095] As shown in Figure 12, for example, when a user operates the instruction input unit 14 to input text information into the input area I1 of the advice goal acceptance image I, and then checks the "Generate KPI proposal" I3 button on the advice goal acceptance image I, the advice goal acceptance unit P12 accepts the text information and the settings of the advice goals contained therein.

[0096] As shown in FIG. 12, the viewpoint information is set to include at least one of process reform, system introduction / improvement, and organizational reform.

[0097] Along with setting the advice goal, for example, the user operates the instruction input unit 14 to check at least one of the "Process Reform" button I21, the "System Introduction / Improvement" button I22, and the "Organizational Reform" button I23 in the advice goal reception image I, and further checks the "Generate KPI Proposal" button I3 in the advice goal reception image I, and the viewpoint information reception unit P9 then accepts the viewpoint information corresponding to the checked button.

[0098] The advice goal setting receiving unit P12 adds the input results of the viewpoint information and text information regarding the setting of the advice goal to a predetermined boilerplate phrase and inputs it into the trained model 30. For example, when the button I21 is checked, the boilerplate phrase is defined so that the viewpoint information and the advice goal can be input into the trained model 30, such as "The advice goals based on an improvement approach centered on process reform are as follows. Please perform KPI analysis in accordance with these goals."

[0099] FIG. 13 is a diagram showing an example of an image (solution area receiving image J) for receiving the selection of a solution area from among the settings related to advice proposals.

[0100] 2 and 13, the solution area receiving image J is displayed on the display unit 13. The solution area receiving unit P13 receives, on the solution area receiving image J, the selection of a solution area from among the settings related to the advice proposal.

[0101] As shown in Figure 13, for example, when the user operates the instruction input unit 14 to check any of the buttons J11, J12, J13 related to setting the "process improvement method," buttons J21, J22, J23 related to setting the "systemization method," and buttons J31, J32, J33 related to setting the "organization / management method" in the solution area reception image J, and further checks the button J4 for "analyze selected method" in the solution area reception image J, the solution area reception unit P13 will accept the selection of such a solution area.

[0102] For example, the buttons J11, J12, and J13 related to the setting of the "process improvement methodology" may include a "Lean methodology" button J11, a "Six Sigma" button J12, and a "BPR" button J13. The buttons J21, J22, and J23 related to the setting of the "systemization method" may include the "RPA implementation" button J21, the "workflow automation" button J22, and the "system integration" button J23. The buttons J31, J32, and J33 related to the setting of the "organization and management method" may include a "change management" button J31, a "training and education" button J32, and a "strengthening governance" button J33.

[0103] The solution domain receiving unit P13 inputs the result of the user's selection of the solution domain into a predetermined fixed phrase into the trained model 30. The fixed phrase is defined as, for example, "The process improvement methods, systemization methods, and organization / management methods are as follows."

[0104] FIG. 14 is a diagram showing an example of an image (provisional advice plan receiving image K) for receiving the setting of a provisional advice plan among the settings related to advice proposals.

[0105] 2 and 14, the tentative advice plan reception image K is displayed on the display unit 13. The tentative advice plan reception unit P14 receives, on the tentative advice plan reception image K, settings for a tentative advice plan among settings related to advice proposals.

[0106] As shown in Figure 14, for example, by operating the instruction input unit 14, the user can check either the "Current situation analysis" button K11, the "Solution design" button K12, or the "Preparation for execution" button K13 on the interim advice plan reception image K, allowing detailed settings of the interim advice plan.

[0107] Furthermore, for example, when a user operates the instruction input unit 14 to input text information into the "Implementation Structure" input area K21 and the "Milestone" input area K22 of the provisional advice plan reception image K, the provisional advice plan reception unit P14 accepts this text information and the implementation structure and milestones contained therein.

[0108] The interim advice plan receiving unit P14 adds the input result of the text information regarding the setting of the interim advice plan to a predetermined standard phrase and inputs it into the trained model 30. For example, when the button K11 is checked, the standard phrase is defined so that the interim advice plan can be input into the trained model 30, such as "The implementation structure and milestones in the current situation analysis are as follows."

[0109] As shown in FIG. 2, the organization information receiving unit P15 receives organization information input from the knowledge base DB.

[0110] The organizational information includes at least one of the target organization's internal regulations, audit regulations, past assurance proposals, and past advisory proposals. Furthermore, the organizational information may include the target organization's purpose, mission, vision, values, strategy, business plan, corporate governance, organizational structure, internal policies, regulations, internal rules, etc., as well as at least one of the target organization's internal audit regulations, past audit papers, audit reports, findings (including judgment criteria, factual explanations of events and situations including audit evidence, root causes, actual problems / potential risks and impacts), improvement proposals (measures to prevent recurrence), follow-up information on the implementation of measures to prevent recurrence, and past advisory proposals.

[0111] Figure 15 shows an example of an image (guaranteed output image L) that displays display information for displaying proposed information including the information to be analyzed, important risks, risk assessment, audit objectives, expected controls, audit key points, audit methods, and recommended audit procedures.

[0112] Figure 16 is a diagram showing an example of an image (advice output image M) that outputs display information for displaying proposed information including the information to be analyzed, important risks, risk assessment, advice goals, solution areas, points of focus (one of the following, or a combination of two or more of these: degree of objective achievement (KPI), risk management (key risk indicators), success factors (CSF), internal control perspective (control key points), sustainability perspective (EHS / ESG indicators), etc.), and a tentative advice plan.

[0113] When a user instructs the proposal generator P16 to make an assurance proposal, the proposal generator P16 transmits predetermined assurance proposal directive phrases and analysis target information to the trained model 30 to analyze the analysis information E. The predetermined assurance proposal directive phrases are written in natural language, for example, "Please analyze the following document based on {approach} and propose appropriate audit key points and procedures," and are stored in the storage unit 22. The viewpoint information received by the viewpoint information receiver P9 is set in {approach}.

[0114] Furthermore, when the user instructs the proposal generator P16 to make an advice proposal, the proposal generator P16 transmits predetermined advice proposal directive phrases and analysis target information to the trained model 30 to analyze the analysis information E. The predetermined advice proposal directive phrases are written in natural language and stored in the memory unit 22, such as, for example, "Analyze the following document based on {approach} and propose appropriate points of focus (outcomes (KPIs), risk management (key risk indicators), conditions for success (CSFs), internal control (control points), sustainability (EHS / ESG)) and an advice plan."

[0115] The trained model 30 is trained to, for example, interpret predetermined guarantee proposal directives or predetermined advice proposal directives in response to a proposal instruction, and to analyze the analysis target information according to at least one of organizational information and viewpoint information. The trained model 30 is trained to, for example, generate proposal information based on the analysis results of the analysis target information.

[0116] The display generation unit P17 instructs the trained model 30 to generate a predetermined template for display so that the proposed information is displayed in a desired order corresponding to the proposed guarantee or advice, and generates the display information.

[0117] 2, 15, and 16, the guarantee and advice output images L, M are displayed on the display unit 13. The proposal generation unit P16 analyzes the analysis target information in accordance with the proposal instructions to generate proposal information, and the display generation unit P17 generates display information for displaying the proposal information.

[0118] As shown in Figures 2 and 15, when the proposal instruction is an assurance proposal instruction, the proposal generation unit P16 generates proposal information (assurance proposal information) including at least one of a focus point and an audit procedure, and the display generation unit P17 generates display information (assurance display information) for displaying the assurance proposal information.

[0119] In this case, the points of focus are generated by analyzing the information to be analyzed in accordance with the assurance proposal instructions. The audit procedures are generated by analyzing the information to be analyzed and the points of focus in accordance with the assurance proposal instructions. The points of focus can be, for example, audit key points.

[0120] In addition, when the proposal generation unit P16 analyzes the analysis target information in accordance with the assurance proposal instructions and generates points of interest, particularly audit key points, it also generates expected controls. The audit key points and expected controls are generated while being associated with each other.

[0121] Furthermore, the proposal generation unit P16 can generate assurance proposal information by analyzing the information to be analyzed in accordance with at least one instruction from among objective / scope information, risk information, significant risks, risk assessment, setting of audit objectives, viewpoint information, selection of audit method, setting of audit procedures, and organizational information, in addition to the assurance proposal instructions.

[0122] For example, the proposal generation unit P16 can analyze the information to be analyzed in accordance with at least one instruction from among objective / scope information, risk information, significant risks, risk assessment, setting of audit objectives, viewpoint information, selection of audit method, and organizational information in addition to the assurance proposal instructions, to generate points of focus, particularly audit key points.

[0123] For example, the proposal generation unit P16 can generate audit procedures by analyzing the information to be analyzed and points of focus, particularly audit key points, in accordance with at least one instruction from among objective / scope information, risk information, significant risks, risk assessment, setting of audit objectives, viewpoint information, selection of audit method, setting of audit procedures, and organizational information, in addition to assurance proposal instructions.

[0124] The assurance proposal information includes at least one of the analysis target information, significant risks, risk assessment, audit objectives, assumed controls, audit key points, audit methods, and recommended audit procedures.

[0125] As shown in Figure 15, the display generation unit P17 generates assurance display information for displaying text information related to the above assurance proposal information, for example, in the output area L1 for "information to be analyzed," the output area L2 for "important risks," the output area L3 for "risk assessment," the output area L4 for "audit objectives," the output area L5 for "anticipated controls," the output area L6 for "audit key points," the output area L7 for "audit method," and the output area L8 for "audit procedures" of the assurance output image L.

[0126] Here, the audit key points are organized into the following system: First, the audit key points are classified into basic audit key points, audit key points specific to non-financial information, area-specific audit key points, and risk-based audit key points.

[0127] The basic audit points are organized from the perspectives of existence / occurrence, completeness / completeness, accuracy, allocation / attribution, and presentation / disclosure. The audit points specific to non-financial information are organized from the perspectives of appropriateness of measurement methods, reliability of data, comparability, and appropriateness of context.

[0128] The specific audit points for each area are organized from the perspectives of environmental information, social information, and governance information. The risk-based audit points are organized from the perspectives of materiality risk, measurement risk, and reporting risk.

[0129] On the other hand, if the proposal instruction is an advice proposal instruction, the proposal generation unit P16 generates advice proposal information including at least one of a point of view and a tentative advice plan, and the display generation unit P17 generates display information (advice display information) for displaying the advice proposal information.

[0130] In this case, the viewpoints are generated by analyzing the analysis target information in accordance with the advice proposal instructions, and the tentative advice plan is generated by analyzing the analysis target information and the viewpoints in accordance with the advice proposal instructions.

[0131] The focus can also be one of the following: objective achievement (KPI), risk management (key risk indicators), success factors (CSF), internal control perspective (control points), sustainability perspective (EHS / ESG indicators), etc., or a combination of two or more of these.

[0132] Furthermore, the proposal generation unit P16 can generate advice proposal information by analyzing the information to be analyzed according to at least one of the following in addition to the advice proposal instructions: purpose / scope information, risk information, significant risks, risk assessment, setting of advice goals, viewpoint information, selection of solution area, setting of interim advice plan, and organizational information.

[0133] For example, the proposal generation unit P16 can generate a solution area by analyzing the analysis target information according to at least one of the following in addition to the advice proposal instructions: purpose / scope information, risk information, significant risks, risk assessment, setting of advice goals, viewpoint information, selection of solution area, and organizational information.

[0134] For example, the proposal generation unit P16 can generate a provisional advice plan by analyzing the analysis target information and the solution area according to at least one of the purpose / scope information, risk information, significant risks, risk assessment, setting of advice goals, viewpoint information, selection of solution area, setting of provisional advice plan, and organizational information in addition to the advice proposal instructions.

[0135] The advice proposal information may include at least one of analysis target information, significant risks, risk assessment, advice goals, solution areas, points of focus, and a tentative advice plan.

[0136] As shown in Figure 16, the display generation unit P17 generates advice display information for displaying text information related to the above advice proposal information, for example, in the output area M1 for "information to be analyzed," the output area M2 for "important risks," the output area M3 for "risk assessment," the output area M4 for "advice goals," the output area M5 for "solution area," the output area M6 for "point of view," and the output area M7 for "tentative advice plan" of the advice output image M.

[0137] KPIs are target indicators that measure the results of proposed and implemented plans. Key risk indicators are warning indicators that supplement risk awareness. CSFs are prerequisites for ensuring factors essential to success. Control points are key points in internal control that mitigate risks and ensure the effectiveness and efficiency of processes. EHS / ESG indicators are non-financial indicators that respond to sustainability and societal demands.

[0138] The proposal generation unit P16 generates proposal information, for example, by natural language processing using the trained model 30. The proposal information is generated by analyzing the analysis target information in accordance with at least the proposal instruction.

[0139] (operation) Next, we will explain an example of a series of operations based on the codes of the analysis information receiving unit P1, proposal instruction switching unit P2, proposal instruction unit P3, objective / scope receiving unit P4, risk information receiving unit P5, risk assessment unit P6, setting instruction unit P7, audit objective receiving unit P8, perspective information receiving unit P9, audit method receiving unit P10, audit procedure receiving unit P11, advice goal receiving unit P12, solution area receiving unit P13, interim advice plan receiving unit P14, organizational information receiving unit P15, proposal generation unit P16, and display generation unit P17.

[0140] Figure 17 is a flowchart showing an example of pre-processing based on the code of the purpose / scope receiving section P4, risk information receiving section P5, risk assessment section P6, setting instruction section P7, audit objective receiving section P8, perspective information receiving section P9, audit method receiving section P10, audit procedure receiving section P11, advice goal receiving section P12, solution area receiving section P13, interim advice plan receiving section P14, and organizational information receiving section P15.

[0141] As shown in Figures 5A to 5C and 17, the purpose and scope reception unit P4 receives text information from the user on the first to third purpose and scope information reception images B1 to B3, including the main stakeholders to be analyzed, the results and / or purposes expected by the main stakeholders, and purpose and scope information having a physical scope (S1).

[0142] As shown in FIGS. 6A to 6C and 17, the risk information receiving unit P5 receives, from the user on the first to third risk information receiving images C1 to C3, a risk category setting, text information including a risk according to the risk category, and a risk level setting (S2).

[0143] As shown in Figures 7A, 7B and 17, the risk assessment unit P6 analyzes the purpose / scope information and risk information using basic assessment and interdependence assessment on the first and second risk assessment reception images D1 and D2 to perform risk assessment (S3).

[0144] 8 and 17, the setting instruction unit P7 receives a switching instruction from the user to select whether to make a setting related to a warranty proposal instruction or a setting related to an advice proposal instruction on the setting instruction image E. At this time, it is determined whether or not there is a setting related to a warranty proposal instruction (S4).

[0145] As shown in FIG. 17, when the setting instruction unit P7 receives an instruction to switch to the setting related to the guarantee proposal instruction (YES), the setting related to the guarantee proposal instruction is performed (S5).

[0146] As shown in FIGS. 9 and 17, the inspection purpose receiving unit P8 receives text information including the inspection purpose from the user on the inspection purpose receiving image F (S6).

[0147] Furthermore, the viewpoint information receiving unit P9 receives, from the user on the audit purpose receiving image F, the setting of viewpoint information including at least one of the compliance approach, the internal integration evaluation approach, and the risk-based approach (S7).

[0148] As shown in FIGS. 10A to 10C and 17, the inspection method receiving unit P10 receives a selection of an inspection method including a data collection method, an analysis method, and a verification method from the user on first to third inspection method receiving images G1 to G3 (S8).

[0149] As shown in FIGS. 11 and 17, the audit procedure reception unit P11 receives detailed settings of the audit procedure from the user on the audit procedure reception image H (S9).

[0150] As shown in Fig. 17, it is determined whether or not additional settings related to a proposal instruction are to be made (S10). If additional settings related to a proposal instruction are to be made (YES), the process returns to S4. If settings related to a proposal instruction are not to be made (NO), the process receives organization information from the knowledge base DB (S11). Thereafter, the pre-processing ends, and the process proceeds to the main processing for a warranty proposal or an advice proposal (S12).

[0151] In S4, when the setting instruction unit P7 receives an instruction to switch to the setting related to the advice suggestion instruction (NO), the setting instruction unit P7 performs the setting related to the advice suggestion instruction (S13).

[0152] As shown in FIGS. 12 and 17, the advice goal receiving unit P12 receives text information including the setting of an advice goal from the user on an advice goal receiving image I (S14).

[0153] Furthermore, the viewpoint information receiving unit P9 receives, from the user on the advice goal receiving image I, the setting of viewpoint information including at least one of process reform, system introduction / improvement, and organizational reform (S15).

[0154] As shown in FIGS. 13 and 17, the solution area receiving unit P13 receives the selection of a solution area including a process improvement method, a systemization method, and an organization / management method from the user in a solution area receiving image J (S16).

[0155] As shown in Figures 14 and 17, the interim advice plan reception unit P14 receives text information from the user on the interim advice plan reception image K, including current situation analysis, solution design, implementation structure for execution preparation, and settings for an interim advice plan with milestones (S17).

[0156] As shown in Fig. 17, it is determined whether or not additional settings related to a proposal instruction are to be made (S10). If additional settings related to a proposal instruction are to be made (YES), the process returns to S4. If settings related to a proposal instruction are not to be made (NO), the process receives organization information from the knowledge base DB (S11). Thereafter, the pre-processing ends, and the process proceeds to the main processing for a warranty proposal or an advice proposal (S12).

[0157] FIG. 18 is a flowchart showing an example of a series of main processes based on the codes of the analysis information receiving unit P1, the proposal instruction switching unit P2, the proposal instruction unit P3, the organization information receiving unit P15, the proposal generating unit P16, and the display generating unit P17.

[0158] As shown in FIGS. 4 and 18, the analysis information receiving unit P1 receives text information including analysis target information from the user on a proposal instruction receiving image A (S21).

[0159] The proposal instruction switching unit P2 receives an instruction from the user to switch the proposal instruction to a guarantee proposal instruction or an advice proposal instruction on the proposal instruction reception image A. At this time, it is determined whether the proposal instruction is a guarantee proposal instruction (S22).

[0160] If the proposal instruction switching unit P2 receives an instruction from the user on the proposal instruction reception image A to switch the proposal instruction to a guarantee proposal instruction (YES), the proposal instruction unit P3 accepts a guarantee proposal instruction to analyze the analysis target information input by the analysis information reception unit P1 and make a proposal regarding guarantee to the organization (S23).

[0161] As shown in Figure 18, the proposal generation unit P16 generates points of focus, particularly audit key points, by analyzing the information to be analyzed according to the objective / scope information, risk information, significant risks, risk assessment, setting of audit objectives, viewpoint information, selection of audit method, and organizational information in addition to the assurance proposal instructions (S24).

[0162] The proposal generation unit P16 generates audit procedures by analyzing the information to be analyzed and the points of focus, particularly the key audit points, in accordance with the assurance proposal instructions, objective and scope information, risk information, significant risks, risk assessment, setting of audit objectives, viewpoint information, selection of audit methods, setting of audit procedures, and organizational information (S25).

[0163] 15 and 18, the display generation unit P17 generates assurance display information (S26) that displays text information related to the analysis target information, important risks, risk assessment, audit objectives, expected controls, audit key points, audit methods, and audit procedures on the assurance output image L. Based on the assurance display information, the assurance output image L is displayed on the display unit 13 (S27).

[0164] As shown in Figures 4 and 18, when the proposal instruction switching unit P2 receives an instruction from the user on the proposal instruction reception image A to switch the proposal instruction to an advice proposal instruction (NO), the proposal instruction unit P3 receives an advice proposal instruction to analyze the analysis target information input by the analysis information reception unit P1 and make a proposal regarding advice to the organization (S28).

[0165] As shown in Figure 18, the proposal generation unit P16 generates a solution area by analyzing the analysis target information according to the advice proposal instructions, purpose / scope information, risk information, important risks, risk assessment, setting of advice goals, viewpoint information, selection of solution area, and organizational information (S29).

[0166] The proposal generation unit P16 generates a provisional advice plan by analyzing the analysis target information and the solution area according to the advice proposal instructions, purpose / scope information, risk information, important risks, risk assessment, setting of advice goals, viewpoint information, selection of solution area, setting of provisional advice plan, and organizational information (S30).

[0167] 16 and 18, the display generation unit P17 generates advice display information (S31) that displays text information related to the analysis target information, important risks, risk assessment, advice goal, solution area, viewpoint, and tentative advice plan on the advice output image M. Based on the advice display information, the advice output image M is displayed on the display unit 13 (S32).

[0168] According to the embodiment, the information processing system 1 can quickly provide appropriate guarantees and advice proposals by performing risk assessment and proposal generation using the trained model 30. Furthermore, by introducing a feedback loop, the accuracy of the proposals can be continuously improved. The proposal generation unit can use the trained model 30 to generate optimal proposals based on proposal instructions from the user, risk assessment data, and similar past cases.

[0169] According to the embodiment, accurate information regarding warranties can be extracted from the analysis target information input by the user in various states in accordance with warranty proposal instructions that are clearly set to make proposals regarding warranties, accurate warranty proposal information can be generated based on the extracted accurate information, and warranty display information that displays the accurate warranty proposal information can be generated.

[0170] In addition, in accordance with advice suggestion instructions that are clearly set to make suggestions regarding advice, accurate information regarding advice can be extracted from the information to be analyzed that is input by the user in various states, accurate advice suggestion information can be generated based on the extracted accurate information, and advice display information that displays the accurate advice suggestion information can be generated.

[0171] In particular, the proposal instruction switching unit P2 clearly switches between guarantee proposal instructions and advice proposal instructions, so that proposal instructions can be clearly set. This enables integrated processing of guarantees and advice, and makes it possible to stably display high-quality proposals even if the information input by the user is diverse.

[0172] According to the embodiment, the assurance proposal information is created in the order of viewpoints and recommended audit procedures, and the audit procedures are created by analyzing the viewpoints that were analyzed earlier. Therefore, accurate assurance proposal information can be generated using the accumulated analysis results, and assurance display information that displays the accurate assurance proposal information can be generated.

[0173] According to the embodiment, the advice proposal information is generated in the order of the solution domain and the tentative advice plan, and the tentative advice plan is generated by analyzing the previously analyzed solution domain. Therefore, accurate advice proposal information can be generated using the accumulated analysis results, and advice display information that displays the accurate advice proposal information can be generated.

[0174] According to the embodiment, accurate information regarding guarantees or advice can be extracted from the analysis target information input by the user in various states according to clear perspective information including at least one of a compliance approach, an internal control evaluation approach, a risk-based approach, a strategic alignment approach, and a social evaluation approach, and suggested information for accurate guarantees or advice can be generated based on the extracted accurate information, and display information for guarantees or advice can be generated that displays suggested information for accurate guarantees or advice.

[0175] According to an embodiment, in addition to the purpose, mission, vision, values, strategy, business plan, corporate governance, organizational structure, internal policies, regulations, internal rules, etc. of the target organization, accurate information regarding assurance or advice can be extracted from the information to be analyzed input by the user in various states in accordance with clear organizational information including at least one of the internal audit regulations of the internal audit function of the target organization, past audit papers, audit reports, findings (including judgment criteria, factual explanations of events and situations including audit evidence, root causes, manifested problems / potential risks and impacts), improvement proposals (measures to prevent recurrence), follow-up information on the introduction of measures to prevent recurrence, and past advice proposals, and accurate assurance or advice proposal information can be generated based on the extracted accurate information, and assurance or advice display information that displays the accurate assurance or advice proposal information can be generated.

[0176] According to the embodiment, in addition to proposal instructions, proposal information is generated by analyzing the information to be analyzed according to purpose and scope information including input of the major stakeholders to be analyzed, input of the results and / or objectives expected by the major stakeholders, and input of the physical scope. Therefore, proposal information can be generated as appropriate assurance or advice while taking into consideration appropriate purpose and scope information, and display information can be generated as appropriate assurance or advice that displays the proposal information as appropriate assurance or advice.

[0177] According to the embodiment, in addition to the proposed instructions, proposed information is generated by analyzing the information to be analyzed according to the important risks and risk assessment, so that proposed information can be generated as appropriate guarantees or advice while taking into consideration appropriate risks, and display information can be generated as appropriate guarantees or advice that displays the proposed information as appropriate guarantees or advice.

[0178] (Variation) In the embodiment, as an example, the case where the suggestion instruction is selected from the warranty suggestion instruction and the advice suggestion instruction has been described. However, the suggestion instruction may be selected from one or more other suggestion instructions in addition to the warranty suggestion instruction and the advice suggestion instruction.

[0179] In the embodiment, as an example, a case has been described in which the viewpoint information includes at least one of the compliance approach, the internal leadership assessment approach, the risk-based approach, the strategic alignment approach, and the social assessment approach. However, the viewpoint information may include at least one of one or more other viewpoint information in addition to the compliance approach, the internal leadership assessment approach, the risk-based approach, the strategic alignment approach, and the social assessment approach.

[0180] In the embodiment, as an example, the case has been described in which the information to be analyzed is included in text information input into the input area A1 of the proposal instruction reception image A. However, the information to be analyzed may also be an electronic file such as an image or text file imported by the user.

[0181] For example, if the electronic file is an image, the analysis information receiving unit P1 may perform character recognition on the text included in the image using OCR, thereby extracting text information. For example, the analysis information receiving unit P1 may extract text information included in a text file.

[0182] In the embodiment, as an example, a case has been described in which the trained model 30 is GPT-4. However, the trained model 30 is not limited to this, and may be Gemini, a large-scale language model, or the like.

[0183] In the embodiment, as an example, the trained model 30 is trained by supervised learning. However, the trained model 30 is not limited to this, and may be trained by both supervised learning and reinforcement learning techniques.

[0184] In the embodiments, guarantees include assurances such as audits, inspections, and certifications such as ISO. Advice in the embodiments includes advisory and consulting.

[0185] That is, the information processing system 1 includes an analysis information receiving unit P1 that receives input of analysis target information to be analyzed, a proposal instruction unit P3 that receives proposal instructions from a user, which are either guarantee proposal instructions for analyzing the analysis target information and making a proposal regarding guarantee to the organization, or advice proposal instructions for analyzing the analysis target information and making a proposal regarding advice to the organization, and a display generation unit P17 that generates display information that displays the proposal information generated by analyzing the analysis target information in accordance with the proposal instructions.

[0186] The information processing method also accepts input of analysis target information to be analyzed, accepts a proposal instruction from the user which is either a guarantee proposal instruction for analyzing the analysis target information and making a proposal regarding guarantee to the organization, or an advice proposal instruction for analyzing the analysis target information and making a proposal regarding advice to the organization, and generates display information that displays the proposal information generated by analyzing the analysis target information in accordance with the proposal instruction.

[0187] The information processing program also causes the computer to execute code for an analysis information receiving unit P1 that receives input of analysis target information to be analyzed, code for a proposal instruction unit P3 that receives a proposal instruction from a user, which is either a guarantee proposal instruction for analyzing the analysis target information and making a proposal regarding guarantee to the organization, or an advice proposal instruction for analyzing the analysis target information and making a proposal regarding advice to the organization, and code for a display generation unit P17 that generates display information that displays the proposal information generated by analyzing the analysis target information in accordance with the proposal instruction.

[0188] As described above, the embodiments of the present invention have been described, but the embodiments are not limited to these, and various changes, modifications, etc. are possible within the scope that does not change the gist of the present invention.

[0189] Furthermore, the order of execution of each step of each procedure in the embodiments may be changed, multiple steps may be executed simultaneously, or the steps may be executed in a different order each time, as long as this does not contradict the nature of the steps. Furthermore, all or part of the steps of each procedure in the present embodiment may be realized by hardware or by a program. [Explanation of symbols]

[0190] P1...analysis information reception section, P2...proposal instruction switching section, P3...proposal instruction section, P4...purpose / scope reception section, P5...risk information reception section, P6...risk assessment section, P9...perspective information reception section, P15...organization information reception section, P16...proposal generation section, P17...display generation section

Claims

1. A purpose / scope receiving unit that receives purpose / scope information including input of key stakeholders to be analyzed, input of the results and / or objectives expected by said key stakeholders, and input of physical scope, and inputs said purpose / scope information into a large-scale language model so that it can be identified in natural language; a risk information receiving unit that receives risk information including a setting of a risk category including at least one of an external environment risk, an internal environment risk, and an operational risk, an input of text information of the risk according to the risk category, and a setting of a risk level according to the risk category, and inputs the risk information into the large-scale language model so that the risk information can be identified in natural language; a risk assessment unit that instructs the large-scale language model to analyze the purpose / scope information and the risk information to generate a risk assessment including a significant risk, a basic assessment, and an interdependency assessment; an analysis information receiving unit that receives input of analysis target information to be analyzed and inputs the analysis target information into the large-scale language model so that the analysis target information can be identified by natural language; a proposal instruction unit that receives from a user a proposal instruction, which is either a guarantee proposal instruction for analyzing the analysis target information and making a proposal regarding guarantee to an organization, or an advice proposal instruction for analyzing the analysis target information and making a proposal regarding advice to the organization, and inputs the proposal instruction into the large-scale language model so that it can be identified by natural language; a proposal generation unit that instructs the large-scale language model to analyze the analysis target information in accordance with the proposal instruction, the purpose / scope information, the significant risks, and the risk assessment to generate proposal information; a display generation unit that generates display information by obtaining the proposed information from the large-scale language model and displaying the display information; An information processing system comprising:

2. The proposal generation unit: If the proposal instruction is the guarantee proposal instruction, An instruction to analyze the analysis target information in accordance with the assurance proposal instruction, the objective and scope information, the significant risks, and the risk assessment to generate a focus point; and An instruction to analyze the analysis target information and the viewpoints in accordance with the assurance proposal instruction, the objectives and scope information, the significant risks, and the risk assessment, and generate recommended audit procedures. on the large-scale language model, The proposed information includes the points of focus and the recommended audit procedures; The information processing system according to claim 1 .

3. The proposal generation unit: When the suggestion instruction is the advice suggestion instruction, An instruction to analyze the analysis target information in accordance with the advice suggestion instruction, the purpose and scope information, the significant risks, and the risk assessment to generate a focus point; and an instruction to analyze the analysis target information and the viewpoints in accordance with the advice proposal instruction, the objectives and scope information, the significant risks, and the risk assessment to generate a tentative advice plan; on the large-scale language model, The proposal information includes the viewpoint and the interim advice plan. The information processing system according to claim 1 .

4. A suggestion instruction switching unit is provided, the proposal instruction switching unit switches the proposal instruction to either the guarantee proposal instruction or the advice proposal instruction in response to the input of the proposal instruction received from the user; The information processing system according to claim 1 .

5. A viewpoint information receiving unit is provided, the viewpoint information receiving unit receives viewpoint information including at least one of a compliance approach, an internal control evaluation approach, a risk-based approach, a strategic alignment approach, and a social evaluation approach, and inputs the viewpoint information into the large-scale language model so that the viewpoint information can be identified by natural language; the proposal generation unit issues an instruction to the large-scale language model to analyze the analysis target information in accordance with the proposal instruction, the purpose / scope information, the significant risk, the risk assessment, and the viewpoint information, and generate the proposal information. The information processing system according to claim 1 .

6. An organization information receiving unit is provided, the organizational information receiving unit receives organizational information including at least one of the organization's regulations, the proposed information of past guarantees, and the proposed information of past advice from a knowledge base stored in a storage unit, and inputs the organizational information into the large-scale language model so that the organizational information can be identified by natural language; the proposal generation unit issues an instruction to the large-scale language model to analyze the analysis target information in accordance with the proposal instruction, the purpose / scope information, the significant risks, the risk assessment, and the organizational information, and generate the proposal information. The information processing system according to claim 1 .

7. An information processing method in which a computer executes multiple processes, comprising: The computer A process of receiving purpose and scope information including input of key stakeholders to be analyzed, input of the results and / or objectives expected by the key stakeholders, and input of physical scope, and inputting the purpose and scope information into a large-scale language model so that it can be specified in natural language; a process of receiving risk information including a setting of risk categories including at least one of external environmental risks, internal environmental risks, and operational risks, an input of text information of risks according to the risk categories, and a setting of risk levels according to the risk categories, and inputting the risk information into the large-scale language model so that the risk information can be identified in natural language; instructing the large-scale language model to analyze the purpose / scope information and the risk information to generate a risk assessment including a key risk, a baseline assessment, and an interdependency assessment; a process of accepting input of analysis target information to be analyzed and inputting the analysis target information into the large-scale language model so that the analysis target information can be identified by natural language; a process of receiving from a user a proposal instruction, which is either an assurance proposal instruction for analyzing the analysis target information and making a proposal regarding assurance to an organization, or an advice proposal instruction for analyzing the analysis target information and making a proposal regarding advice to the organization, and inputting the proposal instruction into the large-scale language model so that it can be identified by natural language; a process of issuing an instruction to the large-scale language model to analyze the analysis target information in accordance with the proposal instruction, the purpose / scope information, the significant risks, and the risk assessment to generate proposal information; generating display information by obtaining the proposed information from the large-scale language model and displaying the information; An information processing method that performs the above.

8. Code for a purpose / scope receiving unit that receives purpose / scope information including input of key stakeholders to be analyzed, input of the outcomes and / or objectives expected by said key stakeholders, and input of physical scope, and inputs said purpose / scope information into a large-scale language model so that it can be identified using natural language; a code for a risk information receiving unit that receives risk information including a setting of a risk category including at least one of an external environment risk, an internal environment risk, and an operational risk, an input of text information of the risk according to the risk category, and a setting of a risk level according to the risk category, and inputs the risk information into the large-scale language model so that the risk information can be identified by natural language; code for a risk assessment unit that instructs the large-scale language model to analyze the purpose / scope information and the risk information to generate a risk assessment including significant risks, a baseline assessment, and an interdependency assessment; code for an analysis information receiving unit that receives input of analysis target information to be analyzed and inputs the analysis target information into the large-scale language model so that the analysis target information can be identified by natural language; code for a proposal instruction unit that receives from a user a proposal instruction, which is either an assurance proposal instruction for analyzing the analysis target information and making a proposal regarding assurance to an organization, or an advice proposal instruction for analyzing the analysis target information and making a proposal regarding advice to the organization, and inputs the proposal instruction into the large-scale language model so that it can be identified by natural language; code for a proposal generation unit that instructs the large-scale language model to analyze the analysis target information in accordance with the proposal instruction, the purpose / scope information, the significant risks, and the risk assessment to generate proposal information; a display generator for generating display information for displaying the suggestion information obtained from the large-scale language model; An information processing program that causes a computer to execute the above.

Images