Information processing device, care need certification support method, and care need certification support program

Abstract

To support determination of a degree of care need.SOLUTION: An information processing device comprises: a query generation section for generating a query for instructing generation of a query sentence for inquiring an item that becomes a judgment material for determining a degree of care need; a generation control section for generating a query sentence by inputting a query generated by the query generation section to a machine-learned generation model so as to generate a query sentence corresponding to an inputted query; a presentation section for presenting the query sentence generated by the generation model to a subject; and a recording section for recording the subject's response to the query sentence. The information processing device can support a user's decision making.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, a method for supporting nursing care certification, and a program for supporting nursing care certification. [Background technology]

[0002] The long-term care insurance system provides nursing care services according to the level of care required to insured persons who require assistance in daily life. The level of care required for an insured person is determined by a computer-based primary assessment based on the results of a home visit by a certification investigator and the opinion of the person's doctor, followed by a secondary assessment by a nursing care certification committee made by certification committee members.

[0003] Certifying the level of nursing care required requires manpower and time, meaning that the human and time costs of certifying the level of nursing care required are high, and technology to reduce these costs is needed. One example of such technology is the behavior analysis device described in Patent Document 1 below. This behavior analysis device analyzes video images of a person who is the subject of nursing care level certification, and automatically determines the level of nursing care required based on the analysis results. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-25935 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the method for determining the level of care required using the behavioral analysis device described in Patent Document 1 is not an officially recognized method. Therefore, under the current system, it is considered that care services cannot be provided based on the level of care required determined using the above-mentioned method. Furthermore, the above-mentioned cost problem is not limited to cases where a specific indicator such as the level of care required is determined, but is a problem that commonly arises in any case where the degree of care required is determined.

[0006] An exemplary object of the present disclosure is to provide a technology that supports determining the level of care required. [Means for solving the problem]

[0007] An information processing device according to an exemplary aspect of the present disclosure includes a query generation means for generating a query that instructs the generation of a question that asks about matters that can be used to determine the degree of care required, a generation control means for inputting the query generated by the query generation means into a generation model that has been machine-learned to generate a question in response to the input query, thereby generating the question, a presentation means for presenting the question generated by the generation model to a subject, and a recording means for recording the subject's response to the question.

[0008] A method for supporting nursing care certification according to an exemplary aspect of the present disclosure includes a query generation process in which one or more processors generate a query that instructs the generation of a question that asks about matters that can be used to determine the level of nursing care required; a generation control process in which the query generated by the query generation process is input into a generation model that has been machine-learned to generate a question in response to the input query, thereby generating the question; a presentation process in which the question generated by the generation model is presented to a subject; and a recording process in which the subject's response to the question is recorded.

[0009] A nursing care certification support program according to an exemplary aspect of the present disclosure causes a computer to function as a query generation means for generating a query that instructs the generation of a question that asks about matters that can be used to determine the level of nursing care required, a generation control means for inputting the query generated by the query generation means into a generation model that has been machine-learned to generate a question in response to the input query, thereby generating the question, a presentation means for presenting the question generated by the generation model to a subject, and a recording means for recording the subject's response to the question. [Effects of the Invention]

[0010] According to an exemplary aspect of the present disclosure, an exemplary effect is achieved in that it is possible to assist in determining the level of care required. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram illustrating a configuration of an information processing device according to the present disclosure. [Figure 2] 1 is a flow chart showing the flow of a nursing care certification support method according to the present disclosure. [Figure 3] 1 is a diagram illustrating the configuration of a nursing care certification support system according to the present disclosure. [Figure 4] 1 is a diagram illustrating support for a home visit by a care certification support system according to the present disclosure. [Figure 5] FIG. 10 is a block diagram showing a configuration of another information processing device according to the present disclosure. [Figure 6] FIG. 1 is a flowchart showing a flow of processing executed by an information processing device according to the present disclosure. [Figure 7] 10 is a flow diagram of a process in which an information processing device according to the present disclosure generates a report. [Figure 8] 10 is a flow diagram of a process executed by an information processing device according to the present disclosure in a home visit survey or a nursing care certification review committee. [Figure 9] FIG. 1 is a block diagram illustrating a configuration of a computer that functions as an information processing device according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0012] The following are examples of embodiments of the present invention. However, the present invention is not limited to the exemplary embodiments shown below, and various modifications are possible within the scope of the claims. For example, embodiments obtained by appropriately combining the technologies (part or all of the products or methods) employed in the exemplary embodiments shown below may also be included in the scope of the present invention. Furthermore, embodiments obtained by appropriately omitting some of the technologies employed in the exemplary embodiments shown below may also be included in the scope of the present invention. Furthermore, the effects mentioned in the exemplary embodiments shown below are examples of effects expected in the exemplary embodiments, and do not define the scope of the present invention. In other words, embodiments that do not exhibit the effects mentioned in the exemplary embodiments shown below may also be included in the scope of the present invention.

[0013] First Exemplary Embodiment A first exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. This exemplary embodiment is a basic form for each of the exemplary embodiments described below. Note that the scope of application of each technique employed in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technique employed in this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise. Furthermore, each technique shown in the drawings referenced to explain this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise.

[0014] (Configuration of information processing device 1) The configuration of an information processing device 1 according to this exemplary embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of the information processing device 1. As shown in Fig. 1, the information processing device 1 includes a query generation unit 101, a generation control unit 102, a presentation unit 103, and a recording unit 104.

[0015] The query generation unit 101 generates a query that instructs the generation of a question sentence asking about factors that will be used to determine the level of care needed. The factors that will be used to determine the level of care needed may be publicly defined. For example, when the information processing device 1 is used to certify the level of care needed, the factors include the subject's physical function, daily living function, and cognitive function. The query generation method may be determined in advance. The information processing device 1 is not limited to supporting the determination of a specific index, the level of care needed, but can also support any case in determining the level of care needed. For example, the information processing device 1 can support the determination of other indexes that indicate the level of care needed, or can support the determination of the level of care needed as a prerequisite for determining products and services corresponding to the level of care needed.

[0016] The generation control unit 102 inputs the query generated by the query generation unit 101 into the generation model to generate a question. The generation model may be one that has been machine-learned to generate a question in response to the input query. For example, when the query generation unit 101 generates a query written in a natural language, a language model that has been machine-learned to learn the arrangement of components (such as words) of a sentence written in a natural language or the arrangement of sentences in a text may be used as the generation model. Note that the generation model may be a model specialized for generating a question that asks about matters that are used as criteria for determining the level of care required, or may be a general-purpose model that can also generate sentences other than questions.

[0017] The presenting unit 103 presents the question generated by the generative model to the subject. The manner of presentation is not particularly limited. For example, the presenting unit 103 may present the question by displaying and outputting it on a display device, by audibly outputting the question on an audio output device, or by printing and outputting the question on a printing device. Furthermore, the output device used for presentation (for example, the display device, audio output device, or printing device) may be included in the information processing device 1 or may be an external device to the information processing device 1.

[0018] The person to whom the questions are presented may be the person to be certified as needing nursing care, or a person other than the person to be certified who is aware of the person's usual physical and mental state (for example, a person living with the person to be certified).

[0019] The recording unit 104 records the subject's answer to the question sentence presented by the presentation unit 103. The answer may be recorded in any destination, for example, in a storage device provided in the information processing device 1, or in a storage device external to the information processing device 1. The answer may be input as text data or as voice data. The recording unit 104 may record the answer input as voice data as is, or may convert it into text data and record it.

[0020] As described above, the information processing device 1 according to this exemplary embodiment is configured to include a query generation unit 101 that generates a query that instructs the generation of a question that asks about matters that can be used to determine the level of care required, a generation control unit 102 that inputs the query generated by the query generation unit 101 into a generation model that has been machine-learned to generate a question in response to the input query, thereby generating the question, a presentation unit 103 that presents the question generated by the generation model to the subject, and a recording unit 104 that records the subject's response to the question.

[0021] According to the above configuration, information that serves as a basis for determining the level of care required can be collected and recorded without manual intervention by an investigator or the like. Therefore, the information processing device 1 has the effect of being able to support the determination of the level of care required. Furthermore, when the information processing device 1 is used to certify the level of care required, the level of care required can be certified in accordance with a public framework (for example, a primary determination by a computer and a secondary determination by a certification committee) while utilizing the above recorded information. Furthermore, the information processing device 1 can also support the decision-making of users of the information processing device 1, such as investigators or certification committee members.

[0022] (Long-term care certification support program) The functions of the information processing device 1 described above can also be realized by a program. A nursing care certification support program according to this exemplary embodiment causes a computer to function as: a query generation unit that generates a query instructing the generation of a question sentence about matters that serve as criteria for determining the level of nursing care needed; a generation control unit that inputs the query generated by the query generation unit into a generation model trained by machine learning to generate a question sentence in response to the input query, causing the generation of the question sentence; a presentation unit that presents the question sentence generated by the generation model to a subject; and a recording unit that records the subject's response to the question sentence. This nursing care certification support program makes it possible to support the determination of the level of nursing care needed.

[0023] (Flow of nursing care certification support method) The flow of the method for supporting nursing care certification according to this exemplary embodiment will be described with reference to Fig. 2. Fig. 2 is a flow diagram showing the flow of the method for supporting nursing care certification. Note that the entity that executes each step in this method for supporting nursing care certification may be a processor provided in the information processing device 1, or a processor provided in another device, or the entities that execute each step may be processors provided in different devices.

[0024] In S11 (query generation process), at least one processor generates a query that instructs the generation of a question sentence that asks about matters that will be used as information for determining the degree of need for care.

[0025] In S12 (generation control process), at least one processor inputs the query generated in S11 into a generative model that has been machine-learned to generate a question sentence in response to the input query, and causes the model to generate a question sentence.

[0026] In S13 (presentation process), at least one processor presents the question generated by the generative model in S12 to the subject.

[0027] In S14 (recording process), at least one processor records the subject's response to the question presented in S13.

[0028] As described above, the method for supporting nursing care certification according to this exemplary embodiment employs a configuration in which at least one processor includes a query generation process for generating a query that instructs the generation of a question that asks about matters that serve as criteria for determining the level of nursing care needed, a generation control process for inputting the query generated by the query generation process into a generation model that has been machine-learned to generate a question in response to the input query, a presentation process for presenting the question generated by the generation model to a subject, and a recording process for recording the subject's response to the presented question. Therefore, the method for supporting nursing care certification according to this embodiment makes it possible to support the determination of the level of nursing care needed.

[0029] Second Exemplary Embodiment A second exemplary embodiment, which is one example of an embodiment of the present invention, will be described in detail with reference to the drawings. Note that the scope of application of each technique employed in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technique employed in this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical hindrance occurs. Furthermore, each technique shown in each drawing referenced to explain this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical hindrance occurs.

[0030] (Outline of the nursing care certification support system) An overview of a nursing care certification support system according to this exemplary embodiment will be described with reference to FIG. 3. FIG. 3 is a diagram illustrating an overview of a nursing care certification support system 5. The nursing care certification support system 5 is a system that supports nursing care certification and is applicable to the fields of medicine and healthcare. The nursing care certification support system 5 shown in FIG. 3 includes an information processing device 1A, a generative model 2, a storage device 3, and a display device 4. The generative model 2 may be stored in the information processing device 1A or in another device (not shown). The storage device 3 may be built into the information processing device 1A or may be external to the information processing device 1A. Although an example of supporting nursing care certification will be described below, the nursing care certification support system 5 can also support various cases in determining the level of nursing care required. Therefore, in the following description, "items that serve as criteria for determining the level of nursing care required" can be read as "items that serve as criteria for determining the level of nursing care required."

[0031] The information processing device 1A generates a query that instructs the generation of a question that asks about matters that will be used as criteria for determining the level of care required, and inputs the generated query into the generation model 2 to generate the question. The information processing device 1A then presents the question generated by the generation model 2 to a subject P1 who is the subject of certification of the level of care required by displaying it on the display device 4, and records the subject P1's answer to the presented question in the storage device 3. Note that the answer to the presented question may be provided by a person other than the subject of certification who is aware of the usual physical and mental state of the subject of certification of the level of care required.

[0032] The generative model 2 is a trained model that has been machine-learned to generate a question in response to an input query. In this exemplary embodiment, an example will be described in which the generative model 2 is a language model that has been machine-learned to determine the arrangement of components (such as words) in a sentence and the arrangement of sentences in a piece of writing. By using the generative model 2, which is a language model, a text-format question can be generated from a text-format query. Note that the generative model 2 is not limited to a language model and can be any model that is appropriate for the format of the input query and the format of the question to be output. For example, if a question including an image is to be generated, a generative model trained to generate an image can be applied as the generative model 2.

[0033] In the example of Fig. 3, a question for subject P1, "Please tell me what happened today," is displayed on the display device 4. This question is generated by the generative model 2 based on a query generated by the information processing device 1A, and is a question asking about the subject P1's daily life, which will be used as information for determining the level of care required. The subject P1 inputs an answer to this question into the information processing device 1A, and the information processing device 1A records the input answer in the storage device 3.

[0034] The answer can be input by any method. For example, the answer of the subject P1 may be input via a microphone (not shown) or an input device (not shown) such as a keyboard. When the answer is input by voice, the information processing device 1A may record the input answer as voice data or may convert it into text data and record it.

[0035] As described above, the nursing care certification support system 5 can collect and record the answers of the subject P1, which serve as information for determining the level of nursing care required by the subject P1, without the intervention of an investigator. This reduces the effort required to collect information that serves as information for determining the level of nursing care required. Furthermore, the nursing care certification support system 5 can collect and record the answers of the subject P1 without the intervention of an investigator, making it possible to determine an appropriate level of nursing care required based on the subject's honest answers.

[0036] The nursing care certification support system 5 can also support a home visit survey for the certification of the level of nursing care required. This will be explained based on Fig. 4. Fig. 4 is a diagram explaining the support for a home visit survey by the nursing care certification support system 5. The home visit survey is a survey in which an investigator visits the home or other location of a person who is the subject of nursing care level certification and interviews the person about their physical and mental condition.

[0037] In the example of FIG. 4, investigators P3 and P4 visit the home of subject P1 and his / her cohabitant P2 to conduct an interview survey for the purpose of nursing care level certification. More specifically, investigator P3 asks subject P1 whether he / she has recently forgotten to turn off the fire. In response to this question, subject P1 replies that he / she has not. The answer to the above question from investigator P3 is also displayed on display device 4. The answer displayed on display device 4 was generated by information processing device 1A and indicates that the fire was forgotten to be turned off once.

[0038] As will be described in detail later, the answers displayed on the display device 4 are generated by the generative model 2 based on the answers of the subject P1 recorded in the storage device 3. Therefore, by referring to both the actual answers of the subject P1 and the answers displayed on the display device 4, the investigators P3 and P4 can accurately judge the state of the subject P1.

[0039] For example, if the answers of subject P1 obtained through a face-to-face interview match the answers displayed on the display device 4, investigators P3 and P4 can determine that the answers of subject P1 obtained through a face-to-face interview are reliable. On the other hand, as in the example of Figure 4, if the answers of subject P1 obtained through a face-to-face interview differ from the answers displayed on the display device 4, it can be determined that subject P1 may have given an answer that is different from the facts due to a memory error or vanity.

[0040] The home visit survey for the certification of the level of care required is conducted in an unusual situation where the investigator visits the home of the subject P1, and the time is limited, so it can be difficult to accurately grasp the subject P1's usual condition only from the statements of the subject P1 and the person living with P2. For example, in the example of Figure 4, if the answers of the subject P1 are different from the facts, the certification result may be lower than the actual condition if the care required certification support system 5 is not used.

[0041] In this regard, the nursing care certification support system 5 makes it possible to determine an appropriate level of nursing care required by utilizing the answers of the subject P1 that are collected without the intervention of an investigator and recorded in the storage device 3. Furthermore, since the answers of the subject P1 recorded in the storage device 3 can be utilized to conduct a home visit survey, the physical and mental state of the subject P1 can be more accurately reflected in the results of the home visit survey.

[0042] Furthermore, the nursing care certification review board, which makes the final decision on the certification of the level of nursing care required, can also use the nursing care certification support system 5 in the same manner as in the case of the above-mentioned home visit survey. That is, at the nursing care certification review board, the certification committee member who decides the level of nursing care required simply asks the nursing care certification support system 5 questions about matters they want to confirm about the subject P1. The nursing care certification support system 5 can cause the generative model 2 to generate answers to the above questions based on the answers of the subject P1 recorded in the storage device 3, and present the generated answers to the certification committee member.

[0043] Furthermore, as will be described in detail later, the nursing care certification support system 5 can also generate a report summarizing the results of interviews with the subject P1. The nursing care certification review board can use this report to determine the level of nursing care required, making it possible to determine an appropriate level of nursing care required.

[0044] As described above, the nursing care certification support system 5 can support the determination of the level of nursing care required in accordance with the current nursing care certification system. It is sufficient that the nursing care certification support system 5 at least supports the collection of information that will be used to determine the level of nursing care required for the subject P1. Furthermore, as described above, the system may support each of the three stages: interviewing the subject P1 about their usual physical and mental condition before the home visit, interviewing them during the home visit, and the nursing care certification review committee. This can contribute to improving the efficiency and accuracy of the entire nursing care certification process.

[0045] (Configuration of information processing device 1A) The configuration of an information processing device 1A according to this exemplary embodiment will be described with reference to Fig. 5. Fig. 5 is a block diagram showing the configuration of the information processing device 1A. Note that the information processing device 1A may be a device whose main function is to support nursing care certification, or may be a general-purpose device having other functions as well. Furthermore, the information processing device 1A may be a stationary device or a portable device.

[0046] As shown in the figure, information processing device 1A includes a control unit 10A that controls all the units of information processing device 1A, and a storage unit 11A that stores various data used by information processing device 1A. Information processing device 1A also includes a communication unit 12A that enables information processing device 1A to communicate with other devices, an input unit 13A that accepts input to information processing device 1A, and an output unit 14A that enables information processing device 1A to output data. Control unit 10A includes a query generation unit 101A, a generation control unit 102A, a presentation unit 103A, a recording unit 104A, an acquisition unit 105A, a contradiction detection unit 106A, an investigation report generation unit 107A, and an output control unit 108A.

[0047] The query generation unit 101A generates a query that instructs the generation of a question sentence asking about matters that will be used as information for determining the level of care required, similar to the query generation unit 101 in exemplary embodiment 1. A method for generating a query by the query generation unit 101A will be described later in the section "Query Generation Method."

[0048] Similar to the generation control unit 102 in exemplary embodiment 1, the generation control unit 102A inputs the query generated by the query generation unit 101A into the generation model 2 that has been machine-learned to generate a question sentence in response to the input query, thereby generating a question sentence. As described above, since the generation model 2 is a language model, the generation control unit 102A inputs the query written in natural language into the generation model 2, causing it to generate a question sentence written in natural language.

[0049] The generative model 2 may be a language model that has been machine-learned from questions posed by an investigator conducting an investigation to certify the level of nursing care required and / or questions posed by a certification committee member at a nursing care certification review board. By using such a generative model 2, the information processing device 1A can obtain, in addition to the effects of the information processing device 1, the effect of being able to present questions similar to those posed by an investigator or a certification committee member.

[0050] The presentation unit 103A presents the question generated by the generative model 2 to the subject, similar to the presentation unit 103 in the first exemplary embodiment. The presentation manner of the answer is arbitrary. For example, the presentation unit 103A may present the answer to the subject by displaying the answer on the display device 4 shown in FIGS. 3 and 4. The presentation unit 103A may also cause the output unit 14A to output the answer.

[0051] The recording unit 104A records the subject's answers to the questions, similar to the recording unit 104 in the first exemplary embodiment. The answer may be recorded in any desired location. For example, the recording unit 104A may record the answer in the storage device 3 shown in FIGS. 3 and 4, or may record the answer in the storage unit 11A.

[0052] The acquisition unit 105A acquires the subject's answer to the question presented by the presentation unit 103A. The method of acquiring the answer is not particularly limited. For example, the acquisition unit 105A may acquire the answer input via the input unit 13A, or may acquire the answer input via an input device external to the information processing device 1A. As described above, the answer may be input by voice or text. Similarly, the acquisition unit 105A can also acquire questions from investigators conducting a home visit survey and certification committee members of a long-term care certification review board.

[0053] The contradiction detection unit 106A detects contradictions in the subject's answers to the questions presented by the presentation unit 103. By including the contradiction detection unit 106A, the information processing device 1A can obtain, in addition to the effects of the information processing device 1, an effect that it is possible to obtain information useful for determining the subject's cognitive function, i.e., whether or not the subject has given contradictory answers.

[0054] Here, "inconsistency" means inconsistency. The method for detecting inconsistency is not particularly limited. For example, the inconsistency detection unit 106A may determine that an inconsistency has been detected when the content of the subject's answer is inconsistent with objective facts. As a specific example, the inconsistency detection unit 106A may determine that an inconsistency has been detected when the date of answer given to a question asking about the subject's date of birth does not match the subject's date of birth. The subject's date of birth may be input in advance by someone who is known to have no cognitive problems, such as a cohabitant of the subject.

[0055] Furthermore, the contradiction detection unit 106A may detect a contradiction using data supporting the subject's behavior, such as an image of the subject, information detected by a portable device or wearable device carried by the subject (e.g., location information, call history, vital data such as heart rate, activity level, etc.), and detection values ​​of various sensors (e.g., human presence sensors) provided within the subject's range of movement. For example, the contradiction detection unit 106A may determine that a contradiction has been detected when an analysis result of an image of the subject indicates that the subject has fallen, but the subject's answer to a question about whether he or she has fallen is that he or she has not fallen.

[0056] Furthermore, the contradiction detection unit 106A may detect contradictions in the answers to multiple questions posed to the subject. For example, when a question identical to a previously asked question is asked again, the contradiction detection unit 106A may determine that a contradiction has been detected if the answers to the questions do not match. The method for determining whether the answers match is not particularly limited. For example, the contradiction detection unit 106A may generate a query that includes multiple answers to be determined as to whether the answers match, and request an answer as to whether the answers match, and input the query to the language model 2. This allows the contradiction detection unit 106A to determine whether the input answers match, based on the output of the language model 2.

[0057] The survey report generation unit 107A generates a survey report showing the results of an interview survey for certifying the level of care required, using the responses of the subject recorded in the storage device 3. By being equipped with the survey report generation unit 107A, the information processing device 1A can obtain the effect of eliminating or reducing the effort required by the survey committee to generate a survey report, in addition to the effect achieved by the information processing device 1. Details of the method for generating a survey report will be explained later in the section "Method for generating a survey report."

[0058] The output control unit 108A causes the output device to output the investigation report generated by the investigation report generation unit 107A. The investigation report may be output as a display or as a printout. The output device that outputs the investigation report may be included in the information processing device 1A or may be an external device to the information processing device 1A.

[0059] (How to generate a query) A method for generating a query by the query generation unit 101A will be described below. The query generation unit 101A can apply any method as long as it is possible to generate a query that instructs the generation of a question sentence that asks about matters that are used as criteria for determining the level of care required.

[0060] For example, the query generation unit 101A may generate a query that includes a document that explains survey items for determining the level of care required and instructs the user to generate a question based on the document. This provides the effect of being able to generate a question sentence according to the survey items and explanations described in the document, in addition to the effect provided by the information processing device 1.

[0061] For example, the query generation unit 101A may generate a query including a document stating that a survey is to be conducted on cognitive function, which is one of the survey items for determining the level of care required, such as whether the subject can communicate their intentions and whether they can correctly answer questions about their date of birth and age. In this case, the query generation unit 101A may generate a query including, for example, a sentence saying, "Please generate a question to check the cognitive function of the subject based on the following document." and the above document. The sentence and document included in this query may be stored in advance in the storage unit 11A or the like.

[0062] Furthermore, the query generation unit 101A can generate a query that generates a question for each survey item by performing the same process for each survey item. Note that it is not essential to generate a query for each survey item, and the query generation unit 101A may generate a query that instructs the generation of questions for multiple survey items. Furthermore, when the volume of a document is large, the query generation unit 101A may generate a query that instructs the generation of questions by summarizing the contents of the document.

[0063] Furthermore, the query generation unit 101A may generate a query including information about the subject. This allows a question appropriate for the subject to be generated. For example, the query generation unit 101A may generate a query including the subject's age, sex, medical history, etc. This allows a question to be generated taking into account the subject's age, sex, medical history, etc.

[0064] Furthermore, it is preferable that the query generation unit 101A generates a query that clearly indicates that the question to be generated is a question for a survey to determine the level of care required. For example, the query generation unit 101A may generate a query including a sentence such as "Please generate questions for a survey to determine the level of care required" or "As an investigator to determine the level of care required, please generate questions to be presented to the subject."

[0065] Furthermore, the query generation unit 101A may generate a query including a specific example of the question to be generated. This allows a question based on the specific example to be generated. For example, the query generation unit 101A may generate a query including a sentence such as, "Please generate a question by referring to the following specific example. Specific example: Regarding cognitive function, have you ever been out and not been able to come home?". Note that two or more specific examples may be included. Furthermore, the specific examples may include questions actually asked by investigators or certification committee members.

[0066] Alternatively, the presenting unit 103A may present the generated questions to the subject sequentially at time intervals, rather than presenting them all at once, and the acquiring unit 105A may sequentially acquire answers to the presented questions. In this case, the generation control unit 102A may input the subject's answer to the presented question into the generative model 2, and generate a new question related to the question. This provides the effect of enabling the generation of a new question based on the subject's answer, in addition to the effect provided by the information processing device 1.

[0067] For example, the query generation unit 101A may generate a query that includes the subject's answer and instructs the generation of a question based on the answer. The generation control unit 102A can input this query to the generation model 2 to generate a new question based on the subject's answer. For example, by inputting a query to the generation model 2 that instructs the generation of a question based on the answer that the subject forgot to turn off the fire, in response to a question asking whether the subject forgot to turn off the fire, it becomes possible to generate a question asking about the frequency of forgetting to turn off the fire. Furthermore, for example, if the content of the question is not conveyed to the subject, it becomes possible to generate a question expressed in simpler terms, or if the subject is upset by the content of the question, it becomes possible to generate a question posed in a different way.

[0068] In the interview survey to determine the level of care required, there are a wide range of items that need to be confirmed with the subject, so if questions about all of the items are presented at once, it is expected that the burden on the subject who has to answer will be large.

[0069] Therefore, it is preferable that the presentation unit 103A distributes the timing of presenting the generated multiple questions within a period until the time when the interview survey is to be ended. This provides the effect of reducing the burden on the subject who answers the questions, in addition to the effect achieved by the information processing device 1. Here, dispersing the timing of presenting the questions within a period means determining the timing of presenting the questions so that the timing of presenting the questions is even or nearly even within the period, or determining the timing of presenting the questions so that the timing of presenting the questions is not concentrated at one time. Note that the presentation unit 103A may determine the timing of presenting each question in advance, or may present the next question only if a predetermined time or more has elapsed since the previous question was presented.

[0070] The time when the interview should end may be specified in advance. For example, the day before the visit survey is to be conducted may be specified as the time when the interview should end. In this case, the timings for presenting the multiple questions are distributed over the period from the time of the specification to the day before the visit survey is to be conducted. The presentation timing may be determined taking into account the schedule of the subject, etc.

[0071] Furthermore, if there is sufficient time before the interview is to be terminated, the presentation unit 103A may present the same question (or questions with the same content) multiple times at different times. In this case, the contradiction detection unit 106A can determine whether there are any contradictions in the answers to the same question.

[0072] The query generation unit 101A may also generate a query that specifies a period for presenting questions and instructs the generation of a set of questions to be asked over that period. For example, if an interview survey needs to be completed within one month, the query generation unit 101A may generate a query that instructs the generation of a set of questions that will enable comprehensive confirmation of each survey item described in the input document within one month. This makes it possible to generate questions with content and number appropriate for the period for presenting questions.

[0073] (How to generate a report) The following describes how the survey report generator 107A generates a survey report. Generally, the format of a survey report is predetermined. Therefore, the survey report generator 107A can generate a survey report by extracting necessary information from the subject's responses recorded in the storage device 3 according to the predetermined format. For example, if the survey report includes an item for describing the subject's physical functions, the survey report generator 107A can extract answers to questions about physical functions from the subject's responses recorded in the storage device 3 and enter them into the item.

[0074] Furthermore, the survey report generation unit 107A may generate the survey report using the generative model 2. For example, the survey report generation unit 107A may generate a query that instructs the extraction of information to be entered into each input field of the survey report from each response of the subject recorded in the storage device 3. The survey report generation unit 107A can then input the generated query into the generative model 2 to identify the information to be entered into each input field of the survey report and generate the survey report. Note that the generation of the query may be executed by the query generation unit 101A, and the input of the query into the generative model 2 may be executed by the generation control unit 102A.

[0075] The survey report generated by the survey report generation unit 107A may be used as one of the materials for determining whether or not a care need is required, or an investigator may create a formal survey report after the home visit by referring to the survey report generated by the survey report generation unit 107A. In the latter case, the survey report generation unit 107A may generate information that will be useful in the home visit and present the information to the investigator via the output control unit 108A or the presentation unit 103A.

[0076] An example of information useful in a home visit survey is the progress of confirmation for each of multiple survey items. For example, among the three survey items, physical function, daily living function, and cognitive function, suppose that a sufficient number of questions about daily living functions are not presented, or the number of responses to the questions about daily living functions is insufficient. In such cases, the survey report generation unit 107A may generate information indicating that the progress of confirmation for daily living functions is low or that daily living functions should be confirmed thoroughly during the home visit survey. This information may be expressed in text or as an image such as a graph. For example, the survey report generation unit 107A may cause the presentation unit 103A to present a radar chart showing the progress of confirmation for each of the multiple survey items. This allows the progress of confirmation for each of the multiple survey items to be recognized at a glance.

[0077] It should be noted that which question corresponds to which survey item may be determined by the generation model 2 or another language model. Furthermore, when a question sentence is generated for each survey item, by associating each generated question sentence with the corresponding survey item and recording the association, it becomes possible to identify the correspondence between the question sentence and the survey item.

[0078] Furthermore, when there is a bias in the degree of progress of confirmation for each survey item, the query generation unit 101A may generate a query that prompts the degree of progress of confirmation for each survey item to fall within an appropriate range when generating a query that instructs the generation of a question from the next time onward. For example, the query generation unit 101A may generate a query that requests the generation of a question that allows the confirmation of each survey item to be evenly distributed, or may generate a query that requests the generation of a question that focuses on the survey item with a low degree of progress.

[0079] Furthermore, the information processing device 1A may include a model updating unit that updates the generative model 2 so that questions of an appropriate quantity and quality are generated for survey items that have not been sufficiently confirmed. For example, the model updating unit may update the generative model 2 by fine-tuning the generative model 2 using example questions for survey items that have not been sufficiently confirmed as training data. The training data may be input by a user of the information processing device 1A, for example.

[0080] (How to generate answers to questions) This section explains how to generate answers to questions from investigators in home visits and from certification committee members at nursing care certification review boards. Note that the method for generating answers is the same whether the questioner is an investigator or a certification committee member, so the following explains an example in which the questioner is an investigator.

[0081] The question from the investigator is acquired by the acquisition unit 105A. When the acquisition unit 105A acquires the question, the query generation unit 101A generates a query that instructs the generation of an answer (which can also be called an answer sentence) to the acquired question. More specifically, the query generation unit 101A generates a query that instructs the generation of an answer to the question acquired by the acquisition unit 105A, based on the answer recorded in the storage device 3 by an interview survey conducted by the care need certification support system 5.

[0082] For example, query generating unit 101A may generate a query including each answer read from storage device 3, the question acquired by acquiring unit 105A, and a standard phrase of a generation instruction specifying a generation method. The standard phrase may be, for example, a sentence such as "Please generate an answer to the input question based on each input answer."

[0083] The generation control unit 102A inputs the query generated in this manner into the generation model 2. This allows the generation model 2 to generate an answer to a question from an investigator that is based on the answer recorded in the storage device 3. Note that the answer to the question from the investigator may be generated by a language model other than the generation model 2 (one that has been machine-learned to generate an answer to a question).

[0084] As described above, the generation control unit 102A causes a machine-learned language model to generate answers to questions, which are answers to questions posed by an investigator conducting an investigation for certifying the level of care required or a certification committee member at a care certification review board, based on recorded answers. This provides the effect of making it possible to obtain information that is useful for the investigation by the investigator or for certification at a care certification review board, in addition to the effect provided by the information processing device 1.

[0085] For example, as explained based on Fig. 4, the on-site investigation committee can request answers to the investigator's questions from both the subject and the information processing device 1A. This allows the investigator to take into account the answers from both the subject and the information processing device 1A and produce reasonable investigation results.

[0086] Furthermore, in the past, nursing care certification examination boards where the subject is not present could only certify the level of nursing care required based on submitted documents such as survey reports and doctor's opinions, and if any unclear points were found in those submitted documents, it was difficult to confirm those unclear points with the subject. In this regard, by using the nursing care certification support system 5, answers to questions about unclear points that are based on recorded answers can be presented, making it possible to resolve any unclear points and make a proper certification.

[0087] (Other uses of answers to questions) The answers recorded in the storage device 3 can also be used for purposes other than certifying the level of care required. For example, the information processing device 1A may be provided with a recommendation unit that determines a product or service to recommend to a subject based on the subject's answers recorded in the storage device 3. It is sufficient to determine in advance what product or service to recommend for what answer. For example, it may be predetermined that for an answer that the subject has difficulty moving their joints, a doctor, clinic, etc. with extensive experience in treating such symptoms should be recommended. In this case, the recommendation unit determines to recommend the above-mentioned doctor, clinic, etc. to the subject who has answered that the subject has difficulty moving their joints. The product or service determined by the recommendation unit may be presented to the subject by the presentation unit 103A.

[0088] Furthermore, the information processing device 1A may provide the subject's answers recorded in the storage device 3 to medical professionals such as the subject's family doctor, on the condition that the subject's consent is obtained. This allows medical professionals to understand the subject's usual condition, which can help provide appropriate and smooth medical services.

[0089] Furthermore, the information processing device 1A may continue to present questions to the subject and collect and record responses even after the home visit survey is completed. In this case, the information processing device 1A may be equipped with a condition change detection unit that detects changes in the subject's physical or mental condition based on changes over time in the content of the recorded responses. This reduces the possibility of overlooking changes in the subject's physical or mental condition.

[0090] For example, the status change detection unit may determine whether the status has changed by comparing a group of answers given by the subject to a group of questions with common content that are presented at predetermined intervals. Whether the content of the answers has changed can be determined, for example, by comparing the feature vectors of each answer, or by using the generative model 2 or another language model. Furthermore, when the status change detection unit detects a change in status, the notification unit 103A may notify the subject of the change. Furthermore, the recipient of the notification of the change in status is not limited to the subject, but may also be, for example, the subject's family, medical professionals, or care providers.

[0091] (Processing flow: before the visit) The flow of processing executed by the information processing device 1A will be described with reference to Fig. 6. Fig. 6 is a flow diagram showing the flow of processing executed by the information processing device 1A. This processing is performed, for example, before a home visit by an investigator. The flow of Fig. 6 includes each step of the nursing care certification support method according to this exemplary embodiment.

[0092] In S21 (query generation process), the query generation unit 101A generates a query that instructs the generation of a question sentence asking about matters that will be used as information for determining the level of care required. The method for generating a query is as described above, and therefore will not be described again here.

[0093] In S22 (generation control process), the generation control unit 102A inputs the query generated in S21 into the generation model 2 that has been machine-learned to generate a question sentence in response to the input query, and causes the generation control unit 102A to generate a question sentence. Note that in S22, the generation control unit 102A may cause the generation model 2 to generate multiple questions sentences.

[0094] In S23 (presentation process), the presenting unit 103A presents the question generated by the generative model 2 to the subject. Note that, if multiple questions are generated in S22, the presenting unit 103A may present a part of the generated questions. In this case, if a determination of NO is made in S26 (described later), the process transitions to S23, and the presenting unit 103A presents another part of the multiple generated questions. By repeating such processes, the multiple generated questions can be presented sequentially to the subject. Note also that the process of S23 does not necessarily have to be performed immediately after the processes of S21 and S22. For example, the presenting unit 103A may present a question generated in advance by the query generation unit 101A and the generation control unit 102A when the subject performs an input operation to the information processing device 1A to request presentation of a question.

[0095] In S24, the acquisition unit 105A acquires the subject's answer to the question presented in S23. As described above, the acquisition unit 105A may acquire the answer input via the input unit 13A, or may acquire the answer input via an input device external to the information processing device 1A. The answer may be input by voice or text.

[0096] In S25, contradiction detection unit 106A determines whether there is a contradiction in the answer acquired in S24. The method for determining whether there is a contradiction is as described above, and therefore the description will not be repeated here. Furthermore, the timing for determining whether there is a contradiction is not limited to the example in FIG. 6. For example, the determination of whether there is a contradiction may be made after the processing of S26 or after the processing of S27.

[0097] In S26, the query generation unit 101A determines whether to end the questioning of the subject. The condition for ending the questioning may be determined in advance. For example, the end condition may be that the number of questions for each survey item has reached a predetermined number. Furthermore, taking into account the subject's fatigue, the end condition may be that a predetermined time (e.g., 15 minutes) or more has elapsed since the start of the process in FIG. 6, or that the total number of questions presented has reached a predetermined number (e.g., 10).

[0098] If the determination in S26 is YES, the process proceeds to S27. On the other hand, if the determination in S26 is NO, the process returns to S21, where the query generation unit 101A generates a query that instructs the generation of a new question. At this time, the query generation unit 101A may generate a query that includes the answer acquired in S24. This makes it possible to generate a new question based on the answer of the subject.

[0099] In S27 (recording process), the recording unit 104A records the answer acquired in S24, i.e., the subject's answer to the question presented in S23. This ends the process in Fig. 6. Note that the answer may be recorded each time an answer is acquired.

[0100] (Process flow: Creation of investigation report) The flow of processing by which the information processing device 1A generates a report will be described with reference to Fig. 7. Fig. 7 is a flow diagram of processing by which the information processing device 1A generates a report.

[0101] In S31, the acquisition unit 105A acquires the answers of the subject recorded in the storage device 3 (more precisely, a plurality of answers recorded during the interview survey conducted by the nursing care need certification support system 5).

[0102] In S32, query generation unit 101A generates a query that instructs the user to summarize the answers acquired in S31. For example, query generation unit 101A may generate a query that includes a template that instructs the user to summarize the answers, such as "Please summarize the answers below," and each answer acquired in S31. Furthermore, for example, query generation unit 101A may also generate a query that includes each item to be filled in the survey form in the query and instructs the user to generate a summary to be entered in each item.

[0103] In S33, the generation control unit 102A causes the generation model 2 to generate summaries of multiple answers to multiple question sentences. Specifically, the generation control unit 102A inputs the query generated in S32 to the generation model 2, which then generates a summary. Note that the generation of the summary may be performed by a language model other than the generation model 2.

[0104] In S34, the survey report generation unit 107A uses the summaries generated in S33 to generate a survey report summarizing the results of the interview survey conducted by the nursing care certification support system 5. For example, when summaries of the subject's responses to each item to be entered in the survey report are generated, the survey report generation unit 107A can generate a survey report by inputting those summaries into the corresponding items.

[0105] In S35, the output control unit 108A outputs the investigation report generated in S34. This ends the processing in Figure 7. Note that the information processing device 1A may be configured to accept corrections made by the investigator to the investigation report output in S35.

[0106] (Processing flow: Visiting investigation / Long-term care certification review board) The flow of processing executed by the information processing device 1A in a home visit survey or a nursing care certification review board will be explained with reference to Fig. 8. Fig. 8 is a flow diagram of processing executed by the information processing device 1A in a home visit survey or a nursing care certification review board. The processing in a home visit survey will be explained below. Note that the processing in a nursing care certification review board is similar except that the questioner and the person to whom the answer to the question is presented are replaced by a certification committee member instead of an investigator.

[0107] In S41, the acquisition unit 105A acquires a question from an investigator. The question from the investigator may be input by voice or may be input as text. If the question is input by voice, the acquisition unit 105A converts the input voice into text.

[0108] In S42, the query generation unit 101A generates a query that instructs the generation of an answer to the question acquired in S41. More specifically, the query generation unit 101A generates a query that instructs the generation of an answer to the question acquired in S41 based on the answer recorded in the storage device 3 by the interview survey conducted by the care need certification support system 5.

[0109] In S43, the generation control unit 102A causes the generation model 2 to generate an answer to the question acquired in S41. Specifically, the generation control unit 102A inputs the query generated in S42 to the generation model 2, which then generates an answer. Note that the generation of an answer to a question from an investigator may be performed by a language model other than the generation model 2.

[0110] In S44, the presentation unit 103A presents the answer generated in S43 to the investigator. This ends the processing in Fig. 8. The answer may be presented in S44 by audio output or by display output on the display device 4 or the like.

[0111] Furthermore, the information processing device 1A may accept a further question related to the answer presented in S44. In this case, the processes of S41 to S44 are repeated again. Then, in the process of S42 from the second time onwards, the query generation unit 101A may generate a query including the answer presented in S44. This makes it possible to generate an answer that takes into account the context of the dialogue, in which a new question has been asked in response to the previously presented answer.

[0112] [Modification] The execution entity of each process described in the above exemplary embodiment is arbitrary and is not limited to the above examples. For example, functions similar to those of the information processing devices 1 and 1A can be realized by a plurality of devices that can communicate with each other. Furthermore, the execution entity of each process shown in the flowcharts of Figures 6 to 8 may be one device (which can also be called a processor) or multiple devices (which can also be called processors).

[0113] [Software implementation example] Some or all of the functions of the information processing devices 1 and 1A may be realized by hardware such as an integrated circuit (IC chip) or by software. Note that the "functions" referred to here include functions realized by the blocks shown in Figures 1 and 5, as well as functions of the model update unit (model update means), recommendation unit (recommendation means), status change detection unit (status change detection means), etc.

[0114] When some or all of the functions of the information processing devices 1 and 1A are realized by software, the information processing devices 1 and 1A are realized, for example, by a computer that executes instructions of a program, which is software that realizes each function. An example of such a computer (hereinafter referred to as computer C) is shown in Fig. 9. Fig. 9 is a block diagram showing the hardware configuration of computer C that functions as information processing device 1 or 1A.

[0115] The computer C includes at least one processor C1 and at least one memory C2. The memory C2 stores a program (care need certification support program) P for causing the computer C to operate as the information processing device 1 or 1A. In the computer C, the processor C1 reads and executes the program P from the memory C2, thereby realizing each function of the information processing device 1 or 1A.

[0116] The processor C1 may be, for example, a central processing unit (CPU), a graphic processing unit (GPU), a digital signal processor (DSP), a micro processing unit (MPU), a floating point number processing unit (FPU), a physics processing unit (PPU), a tensor processing unit (TPU), a quantum processor, a microcontroller, or a combination thereof. The memory C2 may be, for example, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or a combination thereof.

[0117] The computer C may further include a RAM (Random Access Memory) for expanding the program P during execution and for temporarily storing various data. The computer C may also include a communication interface for transmitting and receiving data to and from other devices. The computer C may also include an input / output interface for connecting input / output devices such as a keyboard, mouse, display, and printer.

[0118] Furthermore, the program P can be recorded on a non-transitory tangible recording medium M that can be read by the computer C. Such a recording medium M can be, for example, a tape, a disk, a card, a semiconductor memory, or a programmable logic circuit. The computer C can acquire the program P via such a recording medium M. The program P can also be transmitted via a transmission medium. Such a transmission medium can be, for example, a communication network or broadcast waves. The computer C can also acquire the program P via such a transmission medium.

[0119] Furthermore, each of the above functions of the information processing devices 1 and 1A may be realized by a single processor provided in a single computer, by multiple processors provided in a single computer working together, or by multiple processors provided in each of multiple computers working together. Furthermore, the programs for causing the information processing devices 1 and 1A to realize each of the above functions may be stored in a single memory provided in a single computer, or may be distributed and stored in multiple memories provided in a single computer, or may be distributed and stored in multiple memories provided in each of multiple computers.

[0120] [Additional Notes] This disclosure includes the techniques described in the following appendices. However, the present invention is not limited to the techniques described in the following appendices, and various modifications are possible within the scope of the claims.

[0121] [Appendix A] (Appendix A1) An information processing device comprising: a query generation means for generating a query that instructs the generation of a question sentence that asks about matters that are used as criteria for determining the level of care required; a generation control means for inputting the query generated by the query generation means into a generation model that has been machine-learned to generate a question sentence in response to the input query, thereby generating the question sentence; a presentation means for presenting the question sentence generated by the generation model to a subject; and a recording means for recording the subject's response to the question sentence.

[0122] (Appendix A2) The information processing device described in Appendix A1, wherein the generative model is a language model that has been machine-learned to generate either or both of questions posed by investigators conducting surveys for certifying the level of nursing care required and questions posed by certification committee members at a nursing care certification review board.

[0123] (Appendix A3) The information processing device according to claim 1 or 2, wherein the generation control means inputs the subject's answer to the presented question into the generation model to generate a new question related to the question.

[0124] (Appendix A4) An information processing device described in any of Appendices A1 to A3, comprising a survey report generation means for generating a survey report showing the results of an interview survey for certifying the level of care required using the recorded responses of the subject.

[0125] (Appendix A5) The information processing device described in any of Appendices A1 to A4, wherein the query generation means generates a query that includes a document explaining survey items for determining the level of care required and instructs the device to generate a question based on the document.

[0126] (Appendix A6) The information processing device according to any one of appendices A1 to A5, wherein the presenting means distributes the timings of presenting the plurality of questions within a period until the interview is to be ended.

[0127] (Appendix A7) The information processing device according to any one of appendices A1 to A6, further comprising a contradiction detection means for detecting a contradiction in the subject's answer to the presented question.

[0128] (Appendix A8) The information processing device described in Appendices A1 to A7, wherein the generation control means causes a language model trained by machine learning to generate answers to questions based on the recorded answers to questions posed by an investigator conducting an investigation for certifying the level of nursing care required or a certification committee member at a nursing care certification review board.

[0129] [Appendix B] (Appendix B1) A method for supporting nursing care certification, comprising: a query generation process in which one or more processors generate a query that instructs the generation of a question that asks about matters that will be used as criteria for determining the level of nursing care required; a generation control process in which the query generated by the query generation process is input into a generation model that has been machine-learned to generate a question in response to the input query, thereby generating the question; a presentation process in which the question generated by the generation model is presented to a subject; and a recording process in which the subject's answers to the question are recorded.

[0130] (Appendix B2) A method for supporting nursing care certification as described in Appendix B1, in which the generative model is a language model that has been machine-learned to generate either or both of questions posed by investigators conducting surveys to certify the level of nursing care required and questions posed by certification committee members at a nursing care certification review board.

[0131] (Appendix B3) A method for supporting nursing care certification as described in Appendix B1 or B2, in which, in the generation control process, the at least one processor inputs the subject's answer to the presented question into the generation model to generate a new question related to the question.

[0132] (Appendix B4) A method of supporting nursing care certification described in any of Appendices B1 to B3, including a survey report generation process in which at least one processor generates a survey report showing the results of an interview survey for certifying the level of nursing care required using the recorded responses of the subject.

[0133] (Appendix B5) A method for supporting nursing care certification described in any of Appendices B1 to B4, wherein, in the query generation process, the at least one processor generates a query that includes a document describing survey items for determining the level of nursing care required and instructs the processor to generate questions based on the document.

[0134] (Appendix B6) A method for supporting nursing care certification as described in any of Appendices B1 to B5, wherein, in the presentation process, the at least one processor distributes the timing of presenting the multiple questions within a period up to the time when the interview survey should be terminated.

[0135] (Appendix B7) A method for supporting nursing care certification as described in any one of Appendices B1 to B6, wherein the at least one processor includes a contradiction detection process for detecting contradictions in the subject's answers to the presented question.

[0136] (Appendix B8) A method for supporting nursing care certification described in any of Appendices B1 to B7, wherein the at least one processor performs a process of generating answers to questions from an investigator conducting an investigation for certifying the level of nursing care required or a certification committee member at a nursing care certification review board, based on the recorded answers, using a language model trained on machine learning to generate answers to questions.

[0137] [Appendix C] (Appendix C1) A nursing care certification support program that causes a computer to function as a query generation means that generates a query that instructs the generation of questions that ask about matters that will be used to determine the level of nursing care required, a generation control means that inputs the query generated by the query generation means into a generation model that has been machine-learned to generate questions in response to the input query, thereby generating questions, a presentation means that presents the questions generated by the generation model to a subject, and a recording means that records the subject's answers to the questions.

[0138] (Appendix C2) The nursing care certification support program described in Appendix C1, in which the generative model is a language model that has been machine-learned to generate either or both of questions posed by investigators conducting surveys to certify the level of nursing care required and questions posed by certification committee members at the nursing care certification review board.

[0139] (Appendix C3) The nursing care certification support program described in Appendix C1 or C2, wherein the generation control means inputs the subject's answer to the presented question into the generation model to generate a new question related to the question.

[0140] (Appendix C4) A nursing care certification support program described in any of Appendices C1 to C3, which causes the computer to function as a survey report generation means that generates a survey report showing the results of an interview survey for certifying the level of nursing care required using the recorded responses of the subject.

[0141] (Appendix C5) A nursing care certification support program described in any of Appendices C1 to C4, wherein the query generation means generates a query that includes a document explaining survey items for determining the level of nursing care required and instructs the program to generate questions based on the document.

[0142] (Appendix C6) A nursing care certification support program described in any one of Appendices C1 to C5, wherein the presentation means distributes the timing of presenting the multiple questions within a period up to the time when the interview survey should be completed.

[0143] (Appendix C7) A nursing care certification support program described in any of Appendices C1 to C6, which causes the computer to function as a contradiction detection means for detecting contradictions in the subject's answers to the presented questions.

[0144] (Appendix C8) A nursing care certification support program described in any of Appendices C1 to C2, in which the generation control means generates answers to questions from an investigator conducting an investigation to certify the level of nursing care required or a certification committee member at a nursing care certification review board, based on the recorded answers, using a language model trained by machine learning to generate answers to questions.

[0145] [Appendix D] (Appendix D1) An information processing device comprising at least one processor, the at least one processor executing a query generation process that generates a query that instructs the generation of a question sentence that asks about matters that will be used as criteria for determining the level of care required; a generation control process that inputs the query generated by the query generation process into a generation model that has been machine-learned to generate a question sentence in response to the input query, thereby generating the question sentence; a presentation process that presents the question sentence generated by the generation model to a subject; and a recording process that records the subject's response to the question sentence.

[0146] The information processing device may further include a memory, and the memory may store a program for causing the at least one processor to execute each of the processes.

[0147] (Appendix D2) The information processing device described in Appendix D1, wherein the generative model is a language model that has been machine-learned to generate either or both of questions posed by investigators conducting surveys to certify the level of nursing care required and questions posed by certification committee members at nursing care certification review boards.

[0148] (Appendix D3) The information processing device described in Appendix D1 or D2, wherein in the generation control process, the at least one processor inputs the subject's answer to the presented question into the generation model to generate a new question related to the question.

[0149] (Appendix D4) An information processing device described in any of Appendices D1 to D3, wherein the at least one processor executes a survey report generation process to generate a survey report showing the results of an interview survey for certification of the level of care required using the recorded responses of the subject.

[0150] (Appendix D5) An information processing device described in any of Appendices D1 to D4, wherein in the query generation process, the at least one processor generates a query that includes a document describing survey items for determining the level of care required and instructs the device to generate questions based on the document.

[0151] (Appendix D6) An information processing device described in any of Appendices D1 to D5, wherein in the presentation process, the at least one processor distributes the presentation timing of the multiple question sentences within a period until the interview survey is to be ended.

[0152] (Appendix D7) The information processing device according to any one of appendices D1 to D6, wherein the at least one processor executes a contradiction detection process to detect contradictions in the subject's answers to the presented question.

[0153] (Appendix D8) An information processing device described in any of Appendices D1 to D8, wherein the at least one processor performs a process of generating answers to questions from an investigator conducting a survey for certifying the level of care required or a certification committee member at a care certification review board, based on the recorded answers, using a machine-learned language model to generate answers to questions.

[0154] [Appendix E] (Appendix E1) A non-temporary recording medium that records a nursing care certification support program that causes a computer to perform the following: a query generation process that generates a query that instructs a computer to generate questions that ask about matters that will be used as criteria for determining the level of nursing care required; a generation control process that inputs the query generated by the query generation process into a generation model that has been machine-learned to generate questions in response to the input query, thereby generating the questions; a presentation process that presents the questions generated by the generation model to the subject; and a recording process that records the subject's answers to the questions. [Explanation of symbols]

[0155] 1. Information processing equipment 101 Query generation unit (acceptance means) 102 Generation control unit (generation control means) 103 Presentation unit (presentation means) 104 Recording unit (recording means) 1A Information processing equipment 101A query generation unit (reception means) 102A Generation control unit (generation control means) 103A Presentation unit (presentation means) 104A Recording unit (recording means) 106A contradiction detection unit (conflict detection means) 107A Investigation report generation unit (investigation report generation means) 2. Generative Model

Claims

1. a query generation means for generating a query that instructs the generation of a question sentence asking about matters that will be used as criteria for determining the level of care required; a generation control means for inputting the query generated by the query generation means into a generation model that has been machine-learned to generate a question sentence in response to the input query, and causing the generation model to generate a question sentence; a presentation means for presenting the question generated by the generative model to a subject; and a recording means for recording the subject's response to the question.

2. The information processing device according to claim 1, wherein the generative model is a language model that has been machine-learned from either or both of questions posed by investigators conducting surveys for certifying the level of nursing care required and questions posed by certification committee members at a nursing care certification review board.

3. The information processing device according to claim 1 , wherein the generation control means inputs the subject's answer to the presented question into the generation model to generate a new question related to the question.

4. 3. The information processing device according to claim 1, further comprising a survey report generation means for generating a survey report showing the results of an interview for certification of a level of care required using the recorded responses of the subject.

5. 3. The information processing device according to claim 1, wherein the query generation means generates a query that includes a document explaining survey items for determining a level of care required, and instructs the device to generate a question based on the document.

6. 3. The information processing apparatus according to claim 1, wherein the presenting means distributes the timings of presenting the plurality of questions within a period up to the time when the interview survey is to be ended.

7. 3. The information processing apparatus according to claim 1, further comprising a contradiction detection unit that detects a contradiction in the subject's answer to the presented question.

8. The information processing device described in claim 1 or 2, wherein the generation control means generates an answer based on the recorded answer to a question from an investigator conducting a survey for certifying the level of care required or a certification committee member at a care certification review board, using a language model trained by machine learning to generate an answer to a question.

9. one or more processors a query generation process for generating a query that instructs the generation of a question sentence asking about matters that will be used as criteria for determining the level of care required; a generation control process in which the query generated by the query generation process is input to a generative model that has been machine-learned to generate a question sentence in response to the input query, and the query generated by the query generation process is generated; a presentation process of presenting the question generated by the generative model to a subject; A method for supporting nursing care certification, including a recording process for recording the subject's answers to the questions.

10. Computer, a query generation means for generating a query that instructs the generation of a question sentence asking about matters that will be used as criteria for determining the degree of care required; a generation control means for inputting the query generated by the query generation means into a generation model trained by machine learning to generate a question sentence in response to the input query, and causing the generation model to generate a question sentence; a presentation means for presenting the question generated by the generative model to a subject; and A nursing care certification support program that functions as a recording means for recording the subject's answers to the questions.

Images