Answer support system and answer support method

Abstract

To enable reduction of a burden of answering a check item whose answering method is unknown.SOLUTION: A receiving unit 21 accepts an input of basic data and check items. The basic data is at least one of management data managed for a processing target for which it is determined whether or not a criterion indicated by a guideline is satisfied and image data obtained by photographing the processing target. The check item is an item for determining whether or not the criterion is satisfied. An answer generation unit 22 extracts the basis data serving as the basis of the answer to the check item from the basic data, and generates an answer to the check item based on the basis data.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a technology for assisting in answering check items that check whether a standard is met. [Background technology]

[0002] Evaluations are conducted using guidelines for security, regulations, safety, audits, or the environment, and measures are taken based on the evaluation. In guideline-based evaluations, users input answers to multiple check items to check whether the standards set out in the guidelines are met. However, if users are unfamiliar with the content of the guidelines, they may find it difficult to answer the check items.

[0003] Patent Document 1 describes a method of using responses to other questionnaires to supplement information on responses to similar questions, thereby reducing the burden of responding. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2024-005879 Summary of the Invention [Problem to be solved by the invention]

[0005] The technology described in Patent Document 1 does not complement answer information unless the user has answered similar questions. Therefore, it is not possible to reduce the burden of answering check items that the user has never answered before and does not know how to answer. The present disclosure aims to reduce the burden associated with answering check items that users do not know how to answer. [Means for solving the problem]

[0006] The response support device according to the present disclosure includes: a reception unit that receives input of basic data, which is at least one of management data managed for a processing object for which whether or not it satisfies the criteria set forth in the guidelines and image data of the processing object, and check items, which are items for judging whether or not the criteria are satisfied; an answer generation unit that extracts basis data that serves as a basis for answers to the check items from the basic data accepted by the acceptance unit, and generates answers to the check items based on the basis data; Equipped with. [Effects of the Invention]

[0007] In the present disclosure, basic data and check items are input, evidence data that serves as the basis for answers to the check items is extracted, and answers to the check items based on the evidence data are generated. As a result, answers to check items that are unknown how to answer can be obtained along with the evidence data. Therefore, it is possible to reduce the burden associated with answering check items that are unknown how to answer. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a configuration diagram of a response support device 10 according to a first embodiment. [Figure 2] 3 is a flowchart showing the flow of processing of the response support device 10 according to the first embodiment. [Figure 3] FIG. 2 is an explanatory diagram of the process of the response support device 10 according to the first embodiment. [Figure 4] FIG. 10 is an explanatory diagram of the process of the response support device 10 according to the first modification. [Figure 5] FIG. 10 is a configuration diagram of a response support device 10 according to a second embodiment. [Figure 6] 10 is a flowchart showing the flow of processing of the response support device 10 according to the second embodiment. [Figure 7] FIG. 10 is an explanatory diagram of the process of the response support device 10 according to the second embodiment. [Figure 8] FIG. 10 is a configuration diagram of a response support device 10 according to a third embodiment. [Figure 9] 10 is a flowchart showing the flow of processing of the response support device 10 according to the third embodiment. [Figure 10] FIG. 11 is an explanatory diagram of the process of the response support device 10 according to the third embodiment. [Figure 11] FIG. 10 is a configuration diagram of a response support device 10 according to a fourth embodiment. [Figure 12] 10 is a flowchart showing the flow of processing by the response support device 10 according to the fourth embodiment. [Figure 13] FIG. 10 is an explanatory diagram of the process of the response support device 10 according to the fourth embodiment. [Figure 14] FIG. 10 is a configuration diagram of a response support device 10 according to a fifth embodiment. [Figure 15] 10 is a flowchart showing the flow of processing by the response support device 10 according to the fifth embodiment. [Figure 16] FIG. 13 is an explanatory diagram of the process of the response support device 10 according to the fifth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Embodiment 1 ***Configuration Description*** The configuration of a response support device 10 according to the first embodiment will be described with reference to FIG. The response support device 10 is a computer. The reply support device 10 includes hardware such as a processor 11, a memory 12, a storage 13, and a communication interface 14. The processor 11 is connected to other hardware via signal lines and controls the other hardware.

[0010] The processor 11 is an IC that performs processing. IC stands for Integrated Circuit. Specific examples of the processor 11 include a CPU, a DSP, and a GPU. CPU stands for Central Processing Unit. DSP stands for Digital Signal Processor. GPU stands for Graphics Processing Unit.

[0011] The memory 12 is a storage device that temporarily stores data. Specific examples of the memory 12 include SRAM and DRAM. SRAM stands for Static Random Access Memory. DRAM stands for Dynamic Random Access Memory.

[0012] The storage 13 is a storage device that stores data. A specific example of the storage 13 is an HDD. HDD is an abbreviation for Hard Disk Drive. The storage 13 may also be a portable recording medium such as an SD (registered trademark) memory card, CompactFlash (registered trademark), NAND flash, a flexible disk, an optical disk, a compact disk, a Blu-ray (registered trademark) disk, or a DVD. SD is an abbreviation for Secure Digital. DVD is an abbreviation for Digital Versatile Disk.

[0013] The communication interface 14 is an interface for communicating with external devices. Specific examples of the communication interface 14 include Ethernet (registered trademark), USB, and HDMI (registered trademark) ports. USB stands for Universal Serial Bus. HDMI stands for High-Definition Multimedia Interface.

[0014] The response support device 10 includes, as functional components, a reception unit 21 and a response generation unit 22. The functions of the functional components of the response support device 10 are realized by software. The storage 13 stores a program that realizes the function of each functional component of the response support device 10. This program is read into the memory 12 by the processor 11 and executed by the processor 11. In this way, the function of each functional component of the response support device 10 is realized.

[0015] 1 shows only one processor 11. However, there may be a plurality of processors 11, and the plurality of processors 11 may cooperate to execute programs that realize the respective functions.

[0016] ***Explanation of Operation*** The operation of the response support device 10 according to the first embodiment will be described with reference to FIGS. The operation procedure of the response support device 10 according to the embodiment 1 corresponds to the response support method according to the embodiment 1. Moreover, the program that realizes the operation of the response support device 10 according to the embodiment 1 corresponds to the response support program according to the embodiment 1.

[0017] In the first embodiment, a case where the guidelines define security standards will be described as an example. The guidelines are not limited to security standards, and may also define standards for laws, safety, audits, the environment, and the like. In the first embodiment, the processing target for determining whether or not the standard indicated by the guideline is met is assumed to be a production facility. The processing target is not limited to a production facility, but may be an office or the like.

[0018] The flow of processing by the response support device 10 according to the first embodiment will be described with reference to FIGS. (Step S11: Reception process) The reception unit 21 receives input of basic data 31 from a user terminal. The basic data 31 is at least one of management data managed for the processing target and image data of the processing target. In the first embodiment, the processing target is production equipment. Therefore, the management data is, for example, a setting file for an engineering tool, a management ledger for various assets, or a floor layout of a manufacturing site. The reception unit 21 receives input of one or more check items 32 from an external storage device connected via the communication interface 14. The check items 32 are questions to determine whether or not the criteria set forth in the guidelines are met.

[0019] The basic data 31 is not limited to being input from the user's terminal, but may be acquired by the response support device 10 from an external storage device or the like.

[0020] (Step S12: Answer generation process) The answer generation unit 22 extracts evidence data 33 that serves as the basis for the answers to the check items 32 from the basic data 31 received in step S11. The answer generation unit 22 also generates answers 34 to the check items 32 based on the extracted evidence data 33. Specifically, the answer generation unit 22 inputs the basic data 31 and check items 32 received in step S11 into the answer generation model 35, which is a learning model. Then, the answer generation unit 22 acquires the basis data 33 extracted by the answer generation model 35 and the answer 34 generated by the answer generation model 35 for the basic data 31 and the check items 32. As a result, the answer generation unit 22 extracts the basis data 33 and generates the answer 34. The learning model is what is known as generative AI. AI stands for artificial intelligence. The learning model may be constructed using algorithms such as BERT and GPT, for example. BERT stands for Bidirectional Encoder Representations from Transformers. GPT stands for Generative Pretrained Transformer. The learning model may be constructed by combining multiple algorithms, including these algorithms.

[0021] The answer generation unit 22 may extract the basis data 33 on a rule-based basis based on the basic data 31 and the check items 32, and generate the answer 34. In other words, the answer generation unit 22 may match the basic data 31 and the check items 32 with information stored in a storage device such as the storage 13, and extract the basis data 33 and generate the answer 34.

[0022] (Step S13: Output processing) The answer generation unit 22 outputs the evidence data 33 and the answer 34 obtained in step S12 to the user's terminal and displays them. The user refers to the evidence data 33 and the answer 34 and edits them as necessary to complete the answer 34 to the check item and its evidence.

[0023] Specific examples of the basic data 31, check items 32, evidence data 33, and answers 34 will be described. (Example 1) It is assumed that the basic data 31 is a photograph of a manufacturing site, and the check item 32 is a question as to whether any printed matter containing confidential information has been left in the printer tray. In this case, the evidence data 33 is an image of a printer in a photograph of the manufacturing site, with a circle or the like surrounding the printer. If there is no printed material containing confidential information in the printer tray, the answer 34 will be "Yes." On the other hand, if there is printed material containing confidential information in the printer tray, the answer 34 will be "No." If the printer is not included in the photograph of the manufacturing site, the answer 34 will be "Not applicable."

[0024] (Example 2) Assume that basic data 31 is a photograph of a floor layout or a manufacturing site, and check item 32 is a question about whether physical access to publicly accessible areas within the organization's facilities is monitored by at least one of deploying security guards, using sensor devices, and installing surveillance cameras. In this case, the evidence data 33 is an image of the floor layout or a photograph of the manufacturing site in which the security guard, sensor device, and surveillance camera are surrounded by circles, etc. If at least one of the security guard, sensor device, and surveillance camera is present, the answer 34 will be "Yes." On the other hand, if none of the security guard, sensor device, or surveillance camera is present, the answer 34 will be "No."

[0025] (Example 3) It is assumed that basic data 31 is a photograph of a manufacturing site, and check item 32 is a question as to whether physical access is restricted by placing the equipment in a secure area such as a locked room. In this case, the evidence data 33 is an image of a manufacturing site photograph in which a device such as a PLC is surrounded by a circle or the like. PLC stands for Programmable Logic Controller. If the device such as a PLC is located in a locked room or on a rack, the answer 34 will be "Yes." On the other hand, if the device such as a PLC is not located in a locked room or on a rack, the answer 34 will be "No."

[0026] (Example 4) It is assumed that the basic data 31 is a video of a manufacturing site, and the check item 32 is a question as to whether access to the system and devices is restricted by user IDs or the like. In this case, the evidence data 33 is an image of a PC or other device in a manufacturing site image with a circle or the like surrounding the display. PC stands for Personal Computer. If the display of a PC or other device in the manufacturing site image is always logged on, access is not restricted and anyone can access it, so answer 34 will be "No." In this example, it may be possible to confirm from the evidence data 33 that access is not restricted, but it is difficult to confirm that access is restricted from the evidence data 33. Therefore, if it cannot be confirmed that access is not restricted, the answer 34 will be "not yet" (cannot answer), etc.

[0027] (Example 5) It is assumed that the basic data 31 is a management ledger for various assets, and the check item 32 is a question as to whether ledgers for various assets have been created and whether an inventory has been taken. In this case, the basis data 33 is data that highlights the parts of the management ledger where it is possible to confirm whether an inventory has been taken or not. If it is confirmed that an inventory has been taken, the answer 34 will be "Yes." On the other hand, if it is confirmed that an inventory has not been taken, the answer 34 will be "No."

[0028] Check item 32 may be a question asking whether the various ledgers contain information that identifies the location of the listed devices or software. Check item 32 may also be a question asking whether the various ledgers contain information such as the OS and software version. For these check items 32, the evidence data 33 is data that highlights the portion of the management ledger where the content of these questions is written. Answers 34 are then identified depending on whether or not the information is written.

[0029] In step S11, the receiving unit 21 may receive a plurality of check items 32. In this case, the following two processing methods are possible. (Procedure 1) In step S12, the answer generation unit 22 extracts the basis data 33 for those check items 32 for which the basis data 33 can be extracted from the basic data 31, among the multiple check items 32. Then, the answer generation unit 22 generates answers 34 for the check items 32 from which the basis data 33 was extracted. Check items 32 from which the basis data 33 was not extracted are left unanswered. (Procedure 2) In step S11, the reception unit 21 receives input of a plurality of check items 32 and also receives designation of a check item 32 from among the check items 32 for which an answer 34 is to be generated. The answer generation unit 22 extracts evidence data 33 for the designated check item 32 from among the plurality of check items 32 and generates an answer 34.

[0030] ***Effects of the First Embodiment*** As described above, the response support device 10 according to the first embodiment receives the basic data 31 and the check items 32 as input, extracts the evidence data 33, and generates the response 34 based on the evidence data 33. As a result, for the check items 32 for which it is unclear how to respond, the response 34 can be obtained together with the evidence data 33. This makes it possible to reduce the burden associated with responding to the check items 32 for which it is unclear how to respond.

[0031] ***Other Configurations*** <Variation 1> 4, in step S11, the receiving unit 21 may further receive input of criteria information 36 indicating criteria for determining answers 34 to the check items 32. Then, in step S12, the answer generating unit 22 may extract data corresponding to the criteria indicated by the criteria information 36 as evidence data 33.

[0032] In the case of the above-mentioned Example 1, the reference information 36 is information indicating the following (1) to (3). (1) If the photo shows a printer and there are no printed materials on the tray, or the printed materials are not confidential information (such as confidential or top secret), or there are no printed materials, enter "Yes" in answer 34. (2) If the photo shows a printer and the printed materials on the tray are confidential information (such as confidential or top secret), enter "No" in answer 34. (3) If the photo does not show a printer, enter "Not applicable" in answer 34.

[0033] In the case of the above-mentioned Example 2, the reference information 36 is information indicating the following (1) and (2). (1) If the floor layout or photographs show either security guards, sensor devices, or surveillance cameras, answer 34 as "Yes." (2) If the floor layout or photographs do not show either security guards, sensor devices, or surveillance cameras, answer 34 as "No."

[0034] <Variation 2> In the first embodiment, each functional component is realized by software. However, as a second modification, each functional component may be realized by hardware. The differences between the first embodiment and the second modification will be described below.

[0035] When each functional component is realized by hardware, the reply support device 10 includes an electronic circuit instead of the processor 11, the memory 12, and the storage 13. The electronic circuit is a dedicated circuit for realizing the functions of each functional component, the memory 12, and the storage 13.

[0036] Possible electronic circuits include single circuits, composite circuits, programmed processors, parallel programmed processors, logic ICs, GAs, ASICs, and FPGAs. GA stands for Gate Array. ASIC stands for Application Specific Integrated Circuit. FPGA stands for Field-Programmable Gate Array. Each functional component may be realized by one electronic circuit, or each functional component may be realized by distributing it among a plurality of electronic circuits.

[0037] <Variation 3> As a third modification, some of the functional components may be realized by hardware, and other functional components may be realized by software.

[0038] The processor 11, memory 12, storage 13, and electronic circuitry are collectively referred to as a processing circuit. In other words, the functions of the functional components are realized by the processing circuit.

[0039] Furthermore, the term "unit" in the above description may be read as a "circuit," "step," "procedure," "process," or "processing circuit."

[0040] Embodiment 2 The second embodiment differs from the first embodiment in that an answer 34 to a check item 32 is used as input to generate a question 39 asking about the basis for the answer 34. In the second embodiment, this difference will be explained, and explanations of the same points will be omitted.

[0041] In the first embodiment, the answer 34 is generated by the answer generation model 35. In the second embodiment, the answer 34 is input from the user's terminal. For example, in the method of the first embodiment, there may be a case where the answer 34 is input from the user's terminal for a check item 32 for which the answer generation model 35 was unable to generate an answer 34. There may also be a case where the user inputs the answer 34 for each check item 32 without using the method of the first embodiment.

[0042] ***Configuration Description*** The configuration of the response support device 10 according to the second embodiment will be described with reference to FIG. 1 in that the reply support device 10 includes, as functional components, a validity determination unit 23 and a question generation unit 24. The functions of the question generation unit 24 and the validity determination unit 23 are realized by software or hardware, similar to the other functional components. The response support device 10 does not necessarily have to include the response generation unit 22.

[0043] ***Explanation of Operation*** The flow of processing by the response support device 10 according to the first embodiment will be described with reference to FIGS. (Step S21: Reception process) The reception unit 21 receives input of one or more check items 32 from an external storage device connected via the communication interface 14. The reception unit 21 also receives input of answers 34 to the check items 32 and a rationale input 37 that serves as the rationale for the answers 34 from the user's terminal. Note that there are cases where the rationale input 37 is not input.

[0044] (Step S22: Validity determination process) The validity determination unit 23 determines the validity of the basis input 37 . Specifically, the validity determination unit 23 inputs the check items 32 and the basis input 37 into a validity determination model 38, which is a learning model. The validity determination unit 23 obtains a determination result as to whether the basis input 37 generated by the validity determination model 38 is valid. In this way, the validity determination unit 23 determines the validity of the basis input 37. Alternatively, the validity determination unit 23 may determine the validity of the basis input 37 on a rule basis without using a learning model. The validity determination unit 23 may determine the validity of the basis input 37 by comparing the basis input 37 with a comparison input stored in a storage device such as the storage 13. Here, the comparison input indicates an example of an invalid answer. If the similarity between the comparison input and the basis input 37 is higher than a standard, the validity determination unit 23 determines that the basis input 37 is invalid. If the basis input 37 is not input, the basis input 37 is determined to be invalid.

[0045] If the validity determination unit 23 determines that the basis input 37 is valid, the process ends. On the other hand, if the validity determination unit 23 determines that the basis input 37 is invalid, the process proceeds to step S23.

[0046] (Step S23: Question generation process) The question generator 24 generates a question 39 asking about the basis of the answer 34 from the check items 32 and the answers 34 received in step S21. Specifically, the question generation unit 24 inputs the check items 32 and answers 34 received in step S21 into a question generation model 40, which is a learning model. At this time, the question generation unit 24 may also input a rationale input 37 into the question generation model 40. Then, the question generation unit 24 acquires a question 39 inquiring about the rationale for the answer generated by the question generation model 40. As a result, the question generation unit 24 generates the question 39.

[0047] (Step S24: Output process) The question generator 24 outputs the question sentence 39 obtained in step S23 to the user's terminal to display it. Note that the question generator 24 may also output the question sentence 39 to the user's terminal as audio.

[0048] (Step S25: Response acceptance process) The receiving unit 21 receives an input of a response 41 to the question 39 displayed in step S24, which response 41 indicates the basis for the answer 34. The receiving unit 21 sets the response 41 as the basis input 37 and returns the process to step S22. Then, in step S22, the validity of the basis input 37, which is the response 41, is determined. If it is determined to be invalid, the question 39 is regenerated in step S23. At this time, if the question generation unit 24 determines that the basis input 37, which is the response 41, is invalid, the question generation unit 24 regenerates the question 39 by changing the wording from the question 39 that was the basis of the response 41. Changing the wording does not simply mean changing the sentence, but also includes providing a specific example, etc.

[0049] A specific example will be described. Here, the determination of validity in step S22 includes a case where the basis input 37 is incorrect for the check item 32 and a case where the basis input 37 is insufficient. A case where the basis input is insufficient is a case where an essential element of the basis is missing.

[0050] First, an example in which the basis input 37 is incorrect will be described. Check item 32 is set to "Is the security policy clearly defined, publicized, and implemented?" In step S21, answer 34 is "Yes," and rationale input 37 is "The security policy is XX, publicized, and implemented." In step S22, it is determined to be inappropriate. In step S23, question 39 is generated as "How often do you publicize the security policy?" In step S24, question 39 is output to the user's terminal. In step S25, input of "We publicized it in April" is accepted as response 41 from the user. Then, in step S22, "notified in April" is determined to be invalid because it is not an answer to the frequency, and question 39 is regenerated. In step S23, the wording is changed, and question 39 is generated as "Do you notify everyone every April?" as a question that clarifies the unclear point. In step S24, question 39 is output to the user's terminal. In step S25, input of "Yes" is received as response 41 from the user, and the answer obtained as to the frequency is once a year.

[0051] Next, an example in which the basis input 37 is insufficient will be described. Check item 32 will be "Has the security policy been clarified, disseminated, and implemented?". For this check item 32, answer 34 must include "who," "how often," and "by what means" the policy is disseminated. These are essential elements. In step S21, answer 34 is "Yes," and rationale input 37 is "The security policy is designated as XX, and it is disseminated and put into operation once a year." In step S22, rationale input 37 indicates a rationale corresponding to "how often," but does not indicate a rationale corresponding to "who" and "what means," and is therefore determined to be invalid. In step S23, question 39 is generated, "Who disseminates the security policy?" In step S24, question 39 is output to the user's terminal. In step S25, input of "The department responsible for security has created the security policy, and the president has disseminated it" is accepted as response 41 from the user. Then, in step S22, although there is evidence for "who," there is no evidence for "what means," so it is determined to be invalid, and question 39 is regenerated. In step S23, the wording is changed, and question 39 is generated as "By what means are you making it known?" In step S24, question 39 is output to the user's terminal. In step S25, input of "We are making it known through e-learning" is accepted as response 41 from the user. As a result, the answer "e-learning" is obtained as the means. By repeating questions in this way, the user can answer each question one by one, which has the advantage of providing all the necessary answers without missing anything. In step S23, a question "Who is disseminating the security policy and by what means?" may be generated.

[0052] The required elements may be determined in advance on a rule-based basis for each check item 32, or the validity determination model 38 may check items that are generally considered necessary, such as the 5W1H. When the elements are determined in advance on a rule-based basis, the validity determination model 38 checks whether the response 41 corresponds to that element.

[0053] In addition, if the processing of steps S22 to S25 is repeated several times but the validity is not determined to be valid in step S22, the validity determination unit 23 may determine that a basis for the answer to the check item 32 has not been obtained and terminate the processing.

[0054] ***Effects of the Second Embodiment*** As described above, when a valid basis input 37 cannot be obtained, the response support device 10 according to the second embodiment generates the question 39 asking about the basis of the response 34. This allows the user to understand and input an appropriate basis input 37. As a result, the burden of explaining the response 34 and the basis input 37 to the user is reduced.

[0055] ***Other Configurations*** <Variation 4> In step S21, the receiving unit 21 may further receive input of basic data 31. Then, in step S22, the question generation unit 24 may generate a question 39 using the basic data 31. That is, the question generation unit 24 may input the basic data 31, in addition to the check items 32 and answers 34, to the question generation model 40, and cause the question generation model 40 to generate the question 39.

[0056] For example, suppose check item 32 is a question about whether physical access to publicly accessible areas within the organization's facilities is monitored by at least one of deploying security guards, using sensor devices, and installing surveillance cameras, and basic data 31 is the floor layout. In this case, a question 39 such as "Are there security guards, sensors, or surveillance cameras installed at the location △△ in the ○○ room?" is generated based on the floor layout.

[0057] <Variation 5> In step S21, the receiving unit 21 may further receive input of another check item 32 different from the check item 32 being processed, and an answer 34 and a rationale input 37 for the other check item 32. Then, in step S22, the question generation unit 24 may generate a question 39 using the other check item 32, the answer 34 for the other check item 32, and the rationale input 37. That is, the question generation unit 24 may input the other check item 32, the answer 34 for the other check item 32, and the rationale input 37 for the other check item 32, in addition to the check item 32 and the answer 34, to the question generation model 40, and cause the question generation model 40 to generate the question 39. For example, a question 39 may be generated using the sentences in the reason input 37 for other check items 32 .

[0058] <Variation 6> A sentence may be included in at least one of the answer 34 and the reason input 37. In this case, the question generator 24 generates a question 39 that is suited to the level of the sentence. Specifically, the question generation unit 24 identifies the industry, etc. to which the text creator belongs from the words contained in the text. For example, the question generation unit 24 identifies the industry, etc. to which the text creator belongs by extracting characteristic words specific to the industry, etc. from the text. Then, the question generation unit 24 generates a question 39 according to the industry, etc. "Industry-specific" means a question 39 that uses words used in the industry, etc., or a question 39 that includes examples, etc., of the industry, etc. In this case, the question generation unit 24 inputs the industry, etc., in addition to the check items 32 and answers 34, into the question generation model 40 to generate a question 39 according to the industry, etc.

[0059] Embodiment 3 The third embodiment differs from the first and second embodiments in that it uses a guideline 42 as input to generate an explanation 43 for the check item 32. In the third embodiment, this difference will be explained, and explanations of the same points will be omitted. In the third embodiment, a case where a modification is made to the first embodiment will be described. However, it is also possible to make modifications to the second embodiment.

[0060] In the first embodiment, the answer 34 is generated by the answer generation model 35. In the third embodiment, similar to the second embodiment, a case is assumed in which the user inputs the answer 34.

[0061] ***Configuration Description*** The configuration of the response support device 10 according to the third embodiment will be described with reference to FIG. 1 in that the response support device 10 includes an explanation generation unit 25 as a functional component. The function of the explanation generation unit 25 is realized by software or hardware, similar to the other functional components. The response support device 10 does not necessarily have to include the response generation unit 22.

[0062] ***Explanation of Operation*** The flow of processing by the response support device 10 according to the third embodiment will be described with reference to FIGS. (Step S31: Reception process) The reception unit 21 receives input of the guidelines 42 from the user's terminal. The reception unit 21 also receives input of one or more check items 32 from an external storage device connected via the communication interface 14. The reception unit 21 may also receive input of the guidelines 42 from an external storage device.

[0063] (Step S32: Explanation generation process) The explanation generating unit 25 generates an explanation 43 for the check item 32 from the guideline 42 and the check item 32 received in step S31. Specifically, the explanation generation unit 25 inputs the guidelines 42 and the check items 32 into an explanation generation model 44, which is a learning model. The explanation generation unit 25 acquires the explanation 43 for the check items 32 generated by the explanation generation model 44. As a result, the explanation generation unit 25 generates the explanation 43. The explanation generation model 44 generates the explanation 43 using publicly known information obtainable via a network. For example, suppose check item 32 is "Do security managers and personnel have appropriate skills, and do they improve their skills through security education and training?" In this case, information on the IPA (Information-Technology Promotion Agency) website could be used as publicly known information. The explanation generation unit 25 then accesses the IPA website and presents the portion describing the skills as explanation 43, or presents a summary as explanation 43.

[0064] (Step S33: Output process) The explanation generating unit 25 outputs the explanation 43 obtained in step S32 to the user's terminal and displays it. The user refers to the explanation 43 and inputs the answer 34 to the check item.

[0065] ***Effects of the Third Embodiment*** As described above, the response support device 10 according to the third embodiment generates the explanation 43 for the check item 32. This allows the user to properly understand the content of the check item 32 and to properly input the response 34.

[0066] Embodiment 4 The fourth embodiment differs from the first to third embodiments in that the implementation details 46 are specified by applying countermeasure information 45 indicating the countermeasure to basic data 31. In the fourth embodiment, this difference will be explained, and explanations of the same points will be omitted. In the fourth embodiment, a case where a modification is made to the first embodiment will be described. However, modifications can also be made to the second and third embodiments.

[0067] Answers 34 to each check item 32 are obtained by using the methods of the first to third embodiments, etc. Then, measures to improve the situation are formulated according to the answers 34 to each check item 32. In order to execute the formulated measures, they must be broken down into specific implementation details 46 according to the production equipment, etc. The fourth embodiment supports the work of breaking down the measures into implementation details 46.

[0068] ***Configuration Description*** The configuration of the response support device 10 according to the fourth embodiment will be described with reference to FIG. 1 in that the response support device 10 includes a concretization unit 26 as a functional component. The function of the concretization unit 26 is realized by software or hardware, similar to the other functional components. The response support device 10 does not necessarily have to include the response generation unit 22.

[0069] ***Explanation of Operation*** The flow of processing by the response support device 10 according to the fourth embodiment will be described with reference to FIGS. (Step S41: Reception process) The receiving unit 21 receives, from a user, input of countermeasure information 45 indicating countermeasures to be implemented on a processing target, and basic data 31. Here, the user differs from the user in the first to third embodiments and is assumed to be a person who plans countermeasures, etc.

[0070] (Step S42: instantiation process) The concretization unit 26 applies the measures indicated by the measure information 45 received in step S41 to the basic data 31, thereby concretizing the implementation details 46 of the measures indicated by the measure information 45. Specifically, the concretization unit 26 inputs the countermeasure information 45 and the basic data 31 received in step S41 into a concretization model 47, which is a learning model. The concretization unit 26 acquires implementation details 46 generated by applying the countermeasure indicated by the countermeasure information 45 to the basic data 31 using the concretization model 47. In this way, the concretization unit 26 concretizes the implementation details 46 of the countermeasure indicated by the countermeasure information 45.

[0071] For example, suppose that the measure indicated by the measure information 45 is the installation of a surveillance camera. Also, suppose that the basic data 31 is the floor layout of a manufacturing site. In this case, the concretization unit 26 concretizes the measure indicated by the measure information 45, such as "Install a surveillance camera at position △△ in room XX." Note that the concretization unit 26 may also indicate candidate locations for installing the surveillance camera on the floor layout.

[0072] (Step S43: Output process) The instantiation unit 26 outputs the implementation details 46 instantiated in step S42 to the user's terminal for display.

[0073] ***Effects of the Fourth Embodiment*** As described above, the response support device 10 according to the fourth embodiment embodies the measures indicated by the measure information 45 by applying them to the basic data 31. This makes it easy for the user to easily consider implementing the measures.

[0074] Embodiment 5 The fifth embodiment differs from the fourth embodiment in that measures are prioritized 48. In the fifth embodiment, this difference will be explained, and explanations of the same points will be omitted.

[0075] There are cases where multiple measures are proposed to improve a situation. In such cases, prioritizing the measures 48 in accordance with the user's requests makes it easier for the user to consider them. In the fifth embodiment, prioritizing the measures 48 in accordance with the user's requests, etc.

[0076] ***Configuration Description*** The configuration of the response support device 10 according to the fifth embodiment will be described with reference to FIG. 11 in that the response support device 10 includes a priority setting unit 27 as a functional component. The function of the priority setting unit 27 is realized by software or hardware, similar to the other functional components. The response support device 10 does not necessarily have to include the response generating unit 22 and the concretizing unit 26.

[0077] ***Explanation of Operation*** The flow of processing by the response support device 10 according to the fifth embodiment will be described with reference to FIGS. (Step S51: Reception process) The receiving unit 21 receives input of countermeasure information 45 and conversation data 49 between users regarding the processing target.

[0078] (Step S52: Priority setting process) The priority setting unit 27 sets a priority 48 of the measures indicated by the measure information 45 based on the measures indicated by the measure information 45 received in step S51 and the conversation data 49. Specifically, the priority setting unit 27 inputs the countermeasures indicated by the countermeasure information 45 received in step S51 and the conversation data 49 into a priority determination model 50, which is a learning model. The priority setting unit 27 acquires the priority 48 of each countermeasure generated by the priority determination model 50. In this way, the priority setting unit 27 sets the priority 48 of each countermeasure. For example, suppose that it is determined from a conversation between a user and a user regarding a processing target that the user places importance on availability. In this case, the priority 48 is set so that the priority 48 of a countermeasure against a risk that could lead to an operation stoppage is given a high priority. In addition, when the measures have been concretized by the concretization unit 26 described in embodiment 4, the priority setting unit 27 may input the measures concretized by the concretization unit 26 to the priority determination model 50.

[0079] (Step S53: Output process) The priority setting unit 27 outputs the priority 48 obtained in step S52 to the user's terminal for display.

[0080] ***Effects of the Fifth Embodiment*** As described above, the response support device 10 according to the fifth embodiment sets the priority order 48 of measures based on the conversation data 49 with the user. This makes it easier for the user to consider measures.

[0081] ***Other Configurations*** <Variation 7> In the fifth embodiment, the conversation data 49 is input to the priority determination model 50. However, the concrete model 47 may extract user information such as the user's wishes and restrictions from the conversation data 49 using a rule base or the like, and input the user information to the priority determination model 50 instead of or together with the conversation data 49.

[0082] Various aspects of the present disclosure are summarized below as appendices. (Appendix 1) a reception unit that receives input of basic data, which is at least one of management data managed for a processing object for which whether or not it satisfies the criteria set forth in the guidelines and image data of the processing object, and check items, which are items for judging whether or not the criteria are satisfied; an answer generation unit that extracts basis data that serves as a basis for answers to the check items from the basic data accepted by the acceptance unit, and generates answers to the check items based on the basis data; An answer support device comprising: (Appendix 2) The receiving unit further receives input of criteria information indicating criteria for determining answers to the check items; The answer generation unit extracts data corresponding to the judgment criteria indicated by the reference information as the basis data. 10. The response support device according to claim 1. (Appendix 3) The answer generation unit inputs the basic data and the check items received by the reception unit into an answer generation model, which is a learning model, and obtains the basis data extracted by the answer generation model and the answer generated by the answer generation model, thereby extracting the basis data and generating the answer. 10. The response support device according to claim 1. (Appendix 4) the receiving unit receives input of a plurality of check items, The answer generation unit extracts the basis data for check items for which the basis data can be extracted from the basic data, and generates the answer. (Appendix 5) the receiving unit receives input of a plurality of check items and receives designation of a check item for which an answer is to be generated from among the check items; The answer generation unit extracts the basis data for a specified check item from among the check items and generates the answer. 4. A response support device according to any one of appendices 1 to 3. (Appendix 6) the receiving unit receives input of answers to the check items, The response support device further a question generation unit that generates a question asking about the basis of the answer from the check items and the answers received by the reception unit; 6. The reply support device according to claim 1, comprising: (Appendix 7) The question generation unit inputs the check items and the answers into a question generation model, which is a learning model, and acquires a question that asks about the basis for the answer generated by the question generation model, thereby generating the question. 7. The response support device according to claim 6. (Appendix 8) the receiving unit receives an input of a reason input indicating a reason for the answer; The response support device further a validity determination unit that determines the validity of the reason input Equipped with The question generation unit generates the question sentence when the validity determination unit determines that the basis input is invalid. 8. The response support device according to claim 6 or 7. (Appendix 9) The validity determination unit inputs the basis input into a validity determination model, which is a learning model, and obtains a determination result on whether the basis input is valid or not, generated by the validity determination model, thereby determining the validity of the basis input. 9. The response support device according to claim 8. (Appendix 10) The validity determination unit determines the validity of the basis input by comparing a comparison input stored in a storage device with the basis input. 9. The response support device according to claim 8. (Appendix 11) the receiving unit receives an input of a response to the question sentence acquired by the question generating unit, the response indicating a basis for the answer; the validity determination unit determines the validity of the response; The question generation unit regenerates the question sentence when the validity determination unit determines that the response is invalid. 11. A response support device according to any one of appendices 8 to 10. (Appendix 12) When the validity determination unit determines that the response is invalid, the question generation unit regenerates a question sentence with a different expression from the question sentence that was the basis of the response. 12. The response support device according to claim 11. (Appendix 13) The validity determination unit determines whether the rationale input is valid based on whether the rationale input includes a required element corresponding to the check item. 9. The response support device according to claim 8. (Appendix 14) the receiving unit receives input of basic data, which is at least one of management data managed for a processing object for which it is determined whether or not the processing object satisfies the criteria and image data obtained by capturing the processing object; The question generation unit further generates the question sentence using the basic data. 14. A response support device according to any one of appendixes 6 to 13. (Appendix 15) the receiving unit receives input of other check items different from the check items and answers to the other check items; The question generation unit further generates the question sentence using the other check items and the answers to the other check items. 15. A response support device according to any one of appendices 6 to 14. (Appendix 16) The answer contains a sentence, The question generation unit generates a question sentence that is suited to a level corresponding to the text. 16. A response support device according to any one of appendices 6 to 15. (Appendix 17) the receiving unit receives input of the guideline; The response support device further an explanation generation unit that generates explanations for the check items from the guidelines and the check items received by the reception unit; 17. The reply support device according to any one of appendices 1 to 16, comprising: (Appendix 18) The explanation generation unit inputs the guideline and the check item into an explanation generation model, which is a learning model, and obtains an explanation for the check item generated by the explanation generation model, thereby generating the explanation. 18. The response support device of claim 17. (Appendix 19) the receiving unit receives input of countermeasure information indicating a countermeasure to be implemented on a processing target; The response support device further a concrete implementation unit that applies the countermeasure indicated by the countermeasure information received by the receiving unit to the basic data, thereby concretely implementing the countermeasure indicated by the countermeasure information. 19. The reply support device according to any one of appendices 1 to 18, comprising: (Appendix 20) The concretization unit inputs the countermeasure information and the basic data into a concretization model, which is a learning model, and obtains implementation details generated by applying the countermeasure indicated by the countermeasure information to the basic data using the concretization model, thereby concretizing the implementation details of the countermeasure indicated by the countermeasure information. 19. The response support device of claim 18. (Appendix 21) the countermeasure information indicates a plurality of countermeasures, the receiving unit receives input of conversation data regarding the processing target with a user; The response support device further a priority setting unit that sets priorities of the plurality of measures based on the plurality of measures instantiated by the instantiating unit and the conversation data; 21. The reply support device according to claim 19 or 20, comprising: (Appendix 22) The priority setting unit inputs the plurality of measures instantiated by the instantiating unit and the conversation data into a priority determination model, which is a learning model, and acquires the priorities of the plurality of measures generated by the priority determination model, thereby setting the priorities of the plurality of measures. 22. The response support device according to claim 21. (Appendix 23) The computer receives input of basic data, which is at least one of management data managed for a processing object for which whether or not it satisfies the criteria set forth in the guidelines and image data of the processing object, and check items, which are items for determining whether or not the criteria are satisfied; A response support method in which a computer extracts, from the basic data, evidence data that serves as the basis for responses to the check items, and generates responses to the check items based on the evidence data. (Appendix 24) a reception process for receiving input of basic data, which is at least one of management data managed for a processing object for which whether or not it satisfies the criteria set forth in the guidelines and image data of the processing object, and check items, which are items for determining whether or not the criteria are met; an answer generation process for extracting basis data that serves as a basis for answers to the check items from the basic data received by the reception process, and generating answers to the check items based on the basis data; An answer support program that causes a computer to function as an answer support device that performs the above.

[0083] The embodiments and modifications of the present disclosure have been described above. Some of these embodiments and modifications may be combined and implemented. Also, one or more of them may be implemented partially. Note that the present disclosure is not limited to the above embodiments and modifications, and various modifications are possible as needed. [Explanation of symbols]

[0084] 10 Answer support device, 11 Processor, 12 Memory, 13 Storage, 14 Communication interface, 21 Reception unit, 22 Answer generation unit, 23 Validity determination unit, 24 Question generation unit, 25 Explanation generation unit, 26 Concretization unit, 27 Priority setting unit, 31 Basic data, 32 Check item, 33 Basis data, 34 Answer, 35 Answer generation model, 36 Standard information, 37 Basis input, 38 Validity determination model, 39 Question statement, 40 Question generation model, 41 Response, 42 Guidelines, 43 Explanation statement, 44 Explanation generation model, 45 Countermeasure information, 46 Implementation details, 47 Concretization model, 48 Priority, 49 Conversation data, 50 Priority determination model.

Claims

1. a reception unit that receives input of basic data, which is at least one of management data managed for a processing object for which whether or not it satisfies the criteria set forth in the guidelines and image data of the processing object, and check items, which are items for judging whether or not the criteria are satisfied; an answer generation unit that extracts basis data that serves as a basis for answers to the check items from the basic data accepted by the acceptance unit, and generates answers to the check items based on the basis data; An answer support device comprising:

2. The receiving unit further receives input of criteria information indicating criteria for determining answers to the check items; The answer generation unit extracts data corresponding to the judgment criteria indicated by the reference information as the basis data. The response support device according to claim 1 .

3. The answer generation unit inputs the basic data and the check items received by the reception unit into an answer generation model, which is a learning model, and obtains the basis data extracted by the answer generation model and the answer generated by the answer generation model, thereby extracting the basis data and generating the answer. The response support device according to claim 1 .

4. the receiving unit receives input of a plurality of check items, The response support device according to claim 1 , wherein the response generation unit extracts the basis data for a check item for which the basis data can be extracted from the basic data, and generates the response.

5. the receiving unit receives input of a plurality of check items and receives designation of a check item for which an answer is to be generated from among the check items; The answer generation unit extracts the basis data for a specified check item from among the check items and generates the answer. The response support device according to claim 1 .

6. the receiving unit receives input of answers to the check items, The response support device further a question generation unit that generates a question asking about the basis of the answer from the check items and the answers received by the reception unit; The response support device according to claim 1 , comprising:

7. The question generation unit inputs the check items and the answers into a question generation model, which is a learning model, and acquires a question that asks about the basis for the answer generated by the question generation model, thereby generating the question. The response support device according to claim 6.

8. the receiving unit receives an input of a reason input indicating a reason for the answer; The response support device further a validity determination unit that determines the validity of the reason input Equipped with The question generation unit generates the question sentence when the validity determination unit determines that the basis input is invalid. The response support device according to claim 6.

9. The validity determination unit inputs the basis input into a validity determination model, which is a learning model, and obtains a determination result on whether the basis input is valid or not, generated by the validity determination model, thereby determining the validity of the basis input. The response support device according to claim 8.

10. The validity determination unit determines the validity of the basis input by comparing a comparison input stored in a storage device with the basis input. The response support device according to claim 8.

11. the receiving unit receives an input of a response to the question sentence acquired by the question generating unit, the response indicating a basis for the answer; the validity determination unit determines the validity of the response; The question generation unit regenerates the question sentence when the validity determination unit determines that the response is invalid. The response support device according to claim 8.

12. When the validity determination unit determines that the response is invalid, the question generation unit regenerates a question sentence with a different expression from the question sentence that was the basis of the response. The response support device according to claim 11.

13. The validity determination unit determines whether the rationale input is valid based on whether the rationale input includes a required element corresponding to the check item. The response support device according to claim 8.

14. the receiving unit receives input of basic data, which is at least one of management data managed for a processing object for which it is determined whether or not the processing object satisfies the criteria and image data obtained by capturing the processing object; The question generation unit further generates the question sentence using the basic data. The response support device according to claim 6.

15. the receiving unit receives input of other check items different from the check items and answers to the other check items; The question generation unit further generates the question sentence using the other check items and the answers to the other check items. The response support device according to claim 6.

16. The answer contains a sentence, The question generation unit generates a question sentence that is suited to a level corresponding to the text. The response support device according to claim 6.

17. the receiving unit receives input of the guideline; The response support device further an explanation generation unit that generates explanations for the check items from the guidelines and the check items received by the reception unit; The response support device according to claim 1 , comprising:

18. The explanation generation unit inputs the guideline and the check item into an explanation generation model, which is a learning model, and obtains an explanation for the check item generated by the explanation generation model, thereby generating the explanation. The response support device according to claim 17.

19. the receiving unit receives input of countermeasure information indicating a countermeasure to be implemented on a processing target; The response support device further a concrete implementation unit that applies the countermeasure indicated by the countermeasure information received by the receiving unit to the basic data, thereby concretely implementing the countermeasure indicated by the countermeasure information. The response support device according to claim 1 , comprising:

20. The concretization unit inputs the countermeasure information and the basic data into a concretization model, which is a learning model, and obtains implementation details generated by applying the countermeasure indicated by the countermeasure information to the basic data using the concretization model, thereby concretizing the implementation details of the countermeasure indicated by the countermeasure information.

20. The response support device according to claim 19.

21. the countermeasure information indicates a plurality of countermeasures, the receiving unit receives input of conversation data regarding the processing target with a user; The response support device further a priority setting unit that sets priorities of the plurality of measures based on the plurality of measures instantiated by the instantiating unit and the conversation data; The response support device according to claim 19, comprising:

22. The priority setting unit inputs the plurality of measures instantiated by the instantiating unit and the conversation data into a priority determination model, which is a learning model, and acquires the priorities of the plurality of measures generated by the priority determination model, thereby setting the priorities of the plurality of measures. The response support device according to claim 21.

23. The computer receives input of basic data, which is at least one of management data managed for a processing object for which whether or not it satisfies the criteria set forth in the guidelines and image data of the processing object, and check items, which are items for determining whether or not the criteria are met; A response support method in which a computer extracts, from the basic data, evidence data that serves as the basis for responses to the check items, and generates responses to the check items based on the evidence data.

24. a reception process for receiving input of basic data, which is at least one of management data managed for a processing object for which whether or not it satisfies the criteria set forth in the guidelines and image data of the processing object, and check items, which are items for determining whether or not the criteria are met; an answer generation process for extracting basis data that serves as a basis for answers to the check items from the basic data received by the reception process, and generating answers to the check items based on the basis data; An answer support program that causes a computer to function as an answer support device that performs the above.

Images