Information processing method and information processing program

Abstract

To provide a method for information processing or the like allowing for selecting an appropriate message for a reaction of a care receiver in a dialogue with the care receiver.SOLUTION: A method for information processing is adapted for a computer to output a message to a care receiver based on an interactive scenario with the care receiver, acquire an answer and nonlinguistic information from the care receiver, selects a message matched with the answer and nonlinguistic information acquired based on the interactive scenario, and outputs the selected message addressed to the care receiver.SELECTED DRAWING: Figure 18

Description

[Technical Field]

[0001] The present invention relates to an information processing method for outputting a message based on a dialogue scenario. [Background technology]

[0002] Nursing care services are available to care recipients who have difficulty living independently. The nursing care services needed for each care recipient should be selected based on the remaining abilities of each individual care recipient. Depending on the remaining abilities of the care recipient, multiple care services may be combined. The content and duration of each care service, as well as the frequency of care services, must also be determined. A plan outlining the types of nursing care services to be provided to a care recipient is called a care plan. Care plans are created by care managers in consultation with the care recipient, taking into account the remaining abilities and living environment of the care recipient. Because the remaining abilities of care recipients may change over time, care plans must be updated accordingly. Therefore, care managers are required to meet with the care recipient at least once a month to determine whether the care plan needs to be updated.

[0003] Like those requiring care, those requiring support who have some difficulty living independently can receive preventive care services. Care plans for those requiring support are also created by care managers. It is desirable for care managers to regularly communicate with those requiring support. In this specification, those who receive some kind of service in their daily lives, such as those requiring care and support, are referred to as care recipients. Furthermore, while a care manager generally refers to a care manager, in this specification it is not limited to this and refers to anyone who communicates with those receiving care.

[0004] Meanwhile, an information processing device capable of acquiring information about a care recipient by using IT technology has been proposed (for example, Patent Document 1). The information processing device described in Patent Document 1 can proceed with a voice dialogue with a care recipient based on a dialogue scenario, and acquire information about the care recipient from the care recipient during the dialogue. The information about the care recipient includes, for example, meal times and wake-up times. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-234443 Summary of the Invention [Problem to be solved by the invention]

[0006] In the prior art described in Patent Document 1, an information processing device can converse with a care recipient based on a dialogue scenario on behalf of a care manager. However, in order for the care manager to obtain the care recipient information he or she needs, the dialogue scenario must be devised. This is because simply focusing on the care recipient's response may not lead to a dialogue that proceeds as expected. For example, even if a response to a question is positive, it may only be superficial, and the care recipient may actually have negative feelings. Therefore, in order to conduct a dialogue, it is necessary to consider not only the response but also the care recipient's emotions, which can be inferred from non-verbal information, in order to proceed with the dialogue.

[0007] The present invention has been made in view of the above circumstances, and its purpose is to provide an information processing method and the like that is capable of selecting an appropriate message in response to the care recipient's reaction during a conversation. [Means for solving the problem]

[0008] An information processing method according to one aspect of the present application is characterized in that a computer outputs a message to a care recipient based on a dialogue scenario with the care recipient, acquires the care recipient's responses and non-verbal information, selects a message corresponding to the acquired responses and non-verbal information based on the dialogue scenario, and outputs the selected message to the care recipient. [Effects of the Invention]

[0009] According to one aspect of the present invention, it is possible to select an appropriate message in response to the reaction of the care recipient in a dialogue. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is an explanatory diagram illustrating an example of the configuration of a dialogue system. [Figure 2] FIG. 2 is a block diagram showing an example of a hardware configuration of the dialogue device. [Figure 3] FIG. 2 is a block diagram illustrating an example of a hardware configuration of an editing apparatus. [Figure 4] FIG. 2 is a block diagram illustrating an example of a database configuration. [Figure 5] FIG. 2 is a block diagram illustrating an example of a hardware configuration of the interactive terminal. [Figure 6] FIG. 2 is a block diagram showing an example of the hardware configuration of a professional terminal. [Figure 7] FIG. 10 is an explanatory diagram illustrating an example of a user DB. [Figure 8] FIG. 10 is an explanatory diagram showing an example of a professional DB. [Figure 9] FIG. 1 is an explanatory diagram showing an example of a scenario. [Figure 10] FIG. 2 is an explanatory diagram illustrating an example of a scenario DB. [Figure 11] FIG. 10 is an explanatory diagram illustrating an example of a preset DB. [Figure 12] FIG. 10 is an explanatory diagram illustrating an example of a dialogue attribute DB. [Figure 13] FIG. 10 is an explanatory diagram showing an example of a dialogue content DB. [Figure 14] FIG. 10 is an explanatory diagram showing an example of a dialogue evaluation DB. [Figure 15] FIG. 10 is an explanatory diagram showing an example of a preset status DB. [Figure 16] FIG. 10 is an explanatory diagram showing an example of a pattern DB. [Figure 17] This is a flowchart showing the workflow for creating and updating a scenario. [Figure 18] 10 is a flowchart illustrating an example of a procedure for interactive processing. [Figure 19] 10 is a flowchart illustrating an example of a procedure for a pre-evaluation process. [Figure 20] 10 is a flowchart illustrating an example of a procedure for update processing. [Figure 21] FIG. 10 is an explanatory diagram showing an example of a pre-evaluation screen. [Figure 22] FIG. 10 is an explanatory diagram showing an example of a scenario update screen. [Figure 23] FIG. 10 is an explanatory diagram illustrating an example of an incentive DB. DETAILED DESCRIPTION OF THE INVENTION

[0011] An embodiment will be described below with reference to the drawings. Fig. 1 is an explanatory diagram showing an example of the configuration of an interactive system. The interactive system 100 includes a service providing cluster 10, an interactive terminal 4, and a professional terminal 5. The service providing cluster 10, the interactive terminal 4, and the professional terminal 5 are connected to each other via a network N so that they can communicate with each other.

[0012] The service provider cluster 10 is a computer system that connects multiple computers and makes services available as if they were a single computer. The services provided by the service provider cluster 10 will be described later. The dialogue terminal 4 is a terminal that dialogues with the care recipient. The professional terminal 5 is a terminal used by care managers and the like.

[0013] First, an overview of the work using the dialogue system 100 will be explained. The dialogue terminal 4 is installed near the care recipient. The dialogue terminal 4 dialogues with the care recipient periodically, for example, once a day. The content of the dialogue (images, audio) is sent to the service provider cluster 10 via the network. The care manager uses the professional terminal 5 to view the content of the dialogue of the care recipient under his / her care, thereby understanding the condition of the care recipient (physical condition, cognitive function state). If necessary, the care manager visits the care recipient and updates the care plan.

[0014] The dialogue terminal 4 conducts a dialogue based on a pre-created scenario. The data structure of the scenario is, for example, a tree structure. In the scenario, a message spoken from the dialogue terminal 4 to the care recipient corresponds to a node, and multiple possible responses from the care recipient correspond to links. The tree-structured scenario defines the next message spoken by the dialogue terminal 4 depending on the care recipient's response. Furthermore, the purpose of the dialogue conducted by the dialogue system 100 is to understand the condition of the care recipient. Therefore, detailed consideration is required, such as the development of the conversation and the selection of terms to be used in the dialogue, in order to obtain an appropriate response from the care recipient. In the dialogue system 100, before starting to use the created scenario, the care manager reviews its content and edits it as necessary (referred to as "pre-evaluation"). Furthermore, after accumulating the content of the dialogue between the dialogue terminal 4 and the care recipient using the scenario (referred to as "post-evaluation"), the care manager watches the dialogue and updates the scenario in use as appropriate. This reflects the care manager's experiential knowledge in the scenario, improving the quality of the scenario. As a result, it becomes possible to obtain the same information from the dialogue at the dialogue terminal 4 as when the care manager visits and talks with the patient.

[0015] Next, each component of the dialogue system 100 will be described. The service providing cluster 10 includes a dialogue device 1, an editing device 2, and a database 3. The dialogue device 1 controls a dialogue terminal 4 to execute a dialogue with a care recipient. The editing device 2 provides an editing function for a dialogue scenario. The database 3 stores various data used by the dialogue system 100, such as a dialogue scenario. In the following description, the dialogue device 1, the editing device 2, and the database 3 will be described as being configured as a single computer such as a server computer, a workstation, or a PC (Personal Computer), but this is not limited to this. The dialogue device 1, the editing device 2, and the database 3 may also be configured as a cluster or multicomputer consisting of multiple computers, a virtual machine virtually constructed by software, or a quantum computer. The functions of the dialogue device 1 and the editing device 2 may be provided by a cloud service, or the database stored in the database 3 may be constructed in cloud storage.

[0016] 2 is a block diagram showing an example of the hardware configuration of the dialogue device 1. The dialogue device 1 includes a control unit 11, a main memory unit 12, an auxiliary memory unit 13, a communication unit 14, and a reading unit 15. The control unit 11, the main memory unit 12, the auxiliary memory unit 13, the communication unit 14, and the reading unit 15 are connected by a bus B.

[0017] The control unit 11 has one or more arithmetic processing devices such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit), etc. The control unit 11 reads and executes a control program 1P (program, program product) stored in the auxiliary storage unit 13, thereby performing various information processing, control processing, etc. related to the interactive device 1.

[0018] The main memory unit 12 is a static random access memory (SRAM), a dynamic random access memory (DRAM), a flash memory, etc. The main memory unit 12 mainly temporarily stores data required for the control unit 11 to execute arithmetic processing.

[0019] The auxiliary storage unit 13 is a hard disk or a solid state drive (SSD) or the like, and stores a control program 1P required for the control unit 11 to execute processing.

[0020] The communication unit 14 communicates with the interactive terminal 4 via the network N. In addition, the control unit 11 may use the communication unit 14 to download the control program 1P from another computer via the network N or the like and store it in the auxiliary storage unit 13.

[0021] The reading unit 15 reads the portable storage medium 1a including a CD (Compact Disc)-ROM and a DVD (Digital Versatile Disc)-ROM. The control unit 11 may read the control program 1P from the portable storage medium 1a via the reading unit 15 and store it in the auxiliary storage unit 13. The control unit 11 may also read the control program 1P from the semiconductor memory 1b.

[0022] 3 is a block diagram showing an example of the hardware configuration of the editing device 2. The editing device 2 includes a control unit 21, a main memory unit 22, an auxiliary memory unit 23, a communication unit 24, and a reading unit 25. The control unit 21, the main memory unit 22, the auxiliary memory unit 23, the communication unit 24, and the reading unit 25 are connected by a bus B. The configurations of the control unit 21, the main memory unit 22, the auxiliary memory unit 23, the communication unit 24, and the reading unit 25 are similar to those of the control unit 11, the main memory unit 12, the auxiliary memory unit 13, the communication unit 14, and the reading unit 15 of the dialogue device 1, and therefore description thereof will be omitted.

[0023] 4 is a block diagram showing an example of the database configuration. The database 3 stores a user DB 31, a professional DB 32, a scenario DB 33, a preset DB 34, a dialogue attribute DB 35, a dialogue content DB 36, a dialogue evaluation DB 37, a preset situation DB 38, and a pattern DB 39.

[0024] 5 is a block diagram showing an example of the hardware configuration of the dialogue terminal 4. The dialogue terminal 4 includes a control unit 41, a main memory unit 42, an auxiliary memory unit 43, a communication unit 44, an input unit 45, a display unit 46, an imaging unit 47, an audio acquisition unit 48, and an audio output unit 49. The control unit 41, the main memory unit 42, the auxiliary memory unit 43, the communication unit 44, the input unit 45, the display unit 46, the imaging unit 47, the audio acquisition unit 48, and the audio output unit 49 are connected by a bus B.

[0025] The control unit 41 has one or more arithmetic processing units such as a CPU, an MPU, a GPU, etc. The control unit 41 reads and executes a control program 4P stored in the auxiliary storage unit 43, thereby realizing functional units (an output unit, an acquisition unit, a selection unit, etc.) that provide various functions.

[0026] The main memory unit 42 is an SRAM, a DRAM, a flash memory, etc. The main memory unit 42 mainly temporarily stores data required for the control unit 41 to execute arithmetic processing.

[0027] The auxiliary storage unit 43 is a hard disk, an SSD, or the like, and stores various data necessary for the control unit 41 to execute processing. The auxiliary storage unit 43 may be an external storage device connected to the interactive terminal 4.

[0028] The communication unit 44 communicates with the interactive device 1 via the network N. In addition, the control unit 41 may use the communication unit 44 to download the control program 4P from another computer via the network N or the like and store it in the auxiliary storage unit 43.

[0029] The input unit 45 is a keyboard and a mouse. The display unit 46 includes a liquid crystal display panel or the like. The display unit 46 displays dialogue messages and the like output by the dialogue device 1. The input unit 45 and the display unit 46 may be integrated to form a touch panel display.

[0030] The imaging unit 47 includes, for example, a CCD camera or a CMOS camera, and obtains image data by photoelectrically converting an optical signal input via the CCD or CMOS.

[0031] The audio acquisition unit 48 includes a microphone to acquire audio, converts the acquired audio into a digital audio signal, and outputs it to the bus B. The audio output unit 49 includes an audio speaker, converts the digital audio signal into audio, and outputs it.

[0032] The dialogue terminal 4 can be configured as a smartphone or a tablet computer. The dialogue terminal 4 may be incorporated into a stuffed animal or the like so that the care recipient can easily talk to it during dialogue.

[0033] 6 is a block diagram showing an example of the hardware configuration of a professional terminal 5. The professional terminal 5 includes a control unit 51, a main memory unit 52, an auxiliary memory unit 53, a communication unit 54, an input unit 55, a display unit 56, a voice acquisition unit 57, and a voice output unit 58. The control unit 51, the main memory unit 52, the auxiliary memory unit 53, the communication unit 54, the input unit 55, the display unit 56, the voice acquisition unit 57, and the voice output unit 58 are connected by a bus B.

[0034] The control unit 51 has one or more arithmetic processing units such as a CPU, an MPU, a GPU, etc. The control unit 51 provides various functions by reading and executing a control program 5P stored in the auxiliary storage unit 53.

[0035] The main memory unit 52 is an SRAM, a DRAM, a flash memory, etc. The main memory unit 52 mainly temporarily stores data required for the control unit 51 to execute arithmetic processing.

[0036] The auxiliary storage unit 53 is a hard disk or SSD, etc., and stores various data necessary for the control unit 51 to execute processing. The auxiliary storage unit 53 may be an external storage device connected to the professional terminal 5.

[0037] The communication unit 54 communicates with the editing device 2 via the network N. In addition, the control unit 51 may use the communication unit 54 to download the control program 5P from another computer via the network N or the like and store it in the auxiliary storage unit 53.

[0038] The input unit 55 is a keyboard and a mouse. The display unit 56 includes a liquid crystal display panel or the like. The display unit 56 displays dialogue messages and the like output by the editing device 2. The input unit 55 and the display unit 56 may be integrated to form a touch panel display.

[0039] The audio acquisition unit 57 includes a microphone to acquire audio, converts the acquired audio into a digital audio signal, and outputs it to the bus B. The audio output unit 58 includes an audio speaker, converts the digital audio signal into audio, and outputs it.

[0040] FIG. 7 is an explanatory diagram showing an example of a user DB. The user DB 31 stores information about care recipients. The user DB 31 includes a user ID column, a name column, an age column, a grade column, a living arrangement column, and an address column. The user ID column stores a user ID that can uniquely identify a care recipient. The name column stores the name of the care recipient. The age column stores the age of the care recipient. The grade column stores the level of care or support required by the care recipient. The living arrangement column stores the living arrangement of the care recipient. For example, for a user who lives at home, the living arrangement is "home." Furthermore, for users who live in various facilities, the living arrangement column stores the type of facility. The type of facility is a special nursing home for the elderly, a nursing care health facility for the elderly, a long-term care medical facility, a low-cost nursing home, a care house, etc. The address column stores the user's address.

[0041] Figure 8 is an explanatory diagram showing an example of a professional DB. The professional DB 32 stores information on professionals such as care managers. Professional information does not refer to occupational information, but rather to information on each individual who uses the dialogue system 100 as a professional. The professional DB 32 includes a professional ID column, a name column, an affiliation column, an editing authority column, and a responsible user column. The professional ID column stores a professional ID that uniquely identifies a professional. The name column stores the name of the professional. The affiliation column stores the institution to which the professional belongs. The editing authority column stores whether or not the individual has the authority to edit or update a scenario. A manager with several years of experience is authorized to edit or update a scenario. For example, editing authority is granted to a chief care manager who has obtained the certification of a certified care manager or a chief long-term care manager from the Japan Care Management Association. Editing or updating authority may also be granted to professionals other than certified care managers or chief care managers who have equivalent capabilities. In this specification, certified care managers and the like who have editing authority are referred to as senior professionals (secondary workers), and care managers and the like who do not have editing authority are referred to as general professionals (first workers). Senior professionals and general professionals are collectively referred to as workers. The assigned user column stores the user ID of the user (care recipient) who is in charge of the professional. Note that in this specification, the terms edit and update are used differently for convenience, but they have the same meaning. In the above, only certain professionals have editing authority, but editing authority may be granted to both general professionals and senior professionals.

[0042] FIG. 9 is an explanatory diagram showing an example of a scenario. As mentioned above, a scenario has a tree structure including nodes and links. In the example shown in FIG. 9, there are two types of nodes. The two types of nodes are unit nodes and main state nodes. A unit node is a node similar to a general root node. A unit node is a node that stores the initial spoken message when starting a dialogue on a specific theme (called a "question theme"). A tree structure with a unit node as its root defines a scenario based on a certain question theme. A unit node differs from a general root node in that it can be a child node of a main state node. When transitioning from one question theme to another, a link is established from the main state node to the unit node. Specific examples include a case where multiple question themes are set in a single dialogue, or a case where a dialogue is started on a certain question theme but it is preferable to switch to a dialogue on another question theme based on the user's response.

[0043] FIG. 10 is an explanatory diagram showing an example of a scenario DB. The scenario DB 33 is an example in which the structure of the scenario shown in FIG. 9 is expressed in a relational database. The scenario DB 33 includes a node ID column, a type column, an identification text column, a theme column, a message column, and a child node column. The node ID column stores a node ID that can uniquely identify a node. The type column stores the type of node. In this case, the type is a unit or a main state. The identification text column stores text that professionals and others refer to to make each scenario identifiable. The theme column stores question themes. Question themes are assigned to unit nodes. The message column stores the text of a message to be spoken to the care recipient. The child node column stores information about the child node. The child node column includes a yes / no column, an emotion column, a special note column, and an ID column. The yes / no column, emotion column, and special note column store the conditions for transitioning to a child node. The yes / no column stores the user's conditions for yes / no for a message. The emotion column stores the user's emotional conditions for a message. The special note column stores any noteworthy reactions as the user's response to a message. The ID column stores the node ID of the node to transition to. For example, at node ID = 2, the user is asked the question, "Do you exercise or engage in other activities at least twice a week?" Six patterns of user response are assumed. If the user answers affirmatively, has normal emotions, and is determined to have no noteworthy reactions, the node transitions to the node specified by the ID column corresponding to the affirmative / negative column = Yes, emotion column = normal, and special note column = nothing. In this case, the node transitions to node ID = 5.

[0044] FIG. 11 is an explanatory diagram showing an example of a preset DB. The preset DB 34 stores preset scenarios, which are combinations of multiple scenarios aimed at dialogues on multiple question themes. A preset scenario is also simply called a preset. The preset DB 34 includes an ID column, a preset name column, a theme name column, a question theme 1 column, a question theme 2 column, a question theme 3 column, a question theme 4 column, a question theme 5 column, a remarks column, a description column, a creator column, and a creation date column. The ID column stores an ID that can uniquely identify a preset. The preset name column stores the name of the preset. The theme name column stores the theme name (subject) of the dialogue using the preset. The question theme 1 column stores the first question theme. The question theme 2 column stores the second question theme. The question theme 3 column stores the third question theme. The question theme 4 column stores the fourth question theme. The question theme 5 column stores the fifth question theme. If there are fewer than five question themes set in the preset, no question theme is set in the question theme 5 column, etc. The remarks column stores remarks about the preset. The description column stores an explanation of the preset's purpose, etc. The creator column stores information about the creator of the preset. The creation date column stores the creation date of the preset. The edit or update date when editing or updating may also be stored in the creation date column. As shown in Figure 11, scenarios corresponding to each preset are stored in the scenario DB 33. Note that in Figure 11, the question theme 1 to question theme 5 columns only show the question theme, but the ID of the unit node attached to the first spoken message when starting a dialogue for each question theme may also be stored. In this case, the combination of scenarios for each question theme forms the overall scenario for each preset. A transition message is required when switching question themes, and this is stored separately in the database 3. Examples of transition messages include, "Let me ask you a question on a different topic here," or "Let me change the topic and continue the conversation."

[0045] 12 is an explanatory diagram showing an example of a dialogue attribute DB. The dialogue attribute DB 35 stores dialogue meta information. The dialogue attribute DB 35 includes a log ID column, a preset column, a type column, a start column, an end column, and a partner column. The log ID column stores the log ID that identifies the dialogue data. The preset column stores the ID of the preset used during the dialogue. The type column stores the type of dialogue. "Pre" indicates that the dialogue is the subject of pre-evaluation of the preset. "Post" indicates that the preset is a dialogue for the purpose of post-evaluation. The start column stores the start date and time of the dialogue. The end column stores the end date and time of the dialogue. The partner column stores information about the other party in the dialogue. The partner column stores a distinction column and an ID column. The distinction column stores the distinction of the other party. "User" indicates that the other party in the dialogue was a user (care recipient). "Professional" indicates that the other party in the dialogue was a professional. "Digital Twin" indicates that the other party in the dialogue was a digital twin. "Digital twin" generally refers to technology that collects information that exists in real (actual or physical) space and reproduces it in cyber (virtual) space. Because it is like copying the entire real-world environment into virtual space, it is called a "digital twin" with the meaning of "digital twin." In this specification, "digital twin" refers to a virtual person (care recipient model) that reproduces a care recipient. The ID column stores an ID that identifies the other party in the conversation. If the other party is a user, the ID column stores the user ID. If the other party is a professional, the ID column stores the professional ID. If the other party is a digital twin, the ID column stores identification information that identifies the digital twin.

[0046] FIG. 13 is an explanatory diagram showing an example of a dialogue content DB. The dialogue content DB 36 stores the content of dialogues that have already taken place. The dialogue content DB 36 includes a log ID string, a sequence number string, a system utterance string, and a user response string. The log ID string stores a log ID that identifies the dialogue. The sequence number string stores sequence numbers that indicate the sequence of the spoken messages or user responses included in the dialogue. A dialogue can be reproduced by arranging the spoken messages or user responses in ascending order of sequence numbers. The system utterance string stores information about messages uttered by the dialogue terminal 4. The system utterance string includes a hierarchy string, an ID string, and a text string. The hierarchy string stores the hierarchy of the utterance. A hierarchy is the unit or main state described above. The ID string stores the node ID of the unit or main state. The text string stores the content output as a spoken message as text. The user response string stores the user's response to the spoken message. The user response string includes an affirmative / negative string, an emotion string, a special note string, and a text string. The yes / no column stores whether the user responded positively or negatively. YES indicates a positive response from the user, and NO indicates a negative response from the user. The emotion column stores the user's emotion (non-verbal information) at the time of the response. Emotions are categorized, for example, as "normal," "joy," "surprise," "anger," or "other." The special note column stores the content of a notable response when the user responds. Typically, the special note column stores "nothing in particular." The text column stores the words uttered by the user as text. The user's response is determined based on the user's speech, movements, facial expressions, etc. collected from the dialogue terminal 4. The determination method is publicly known, so a description thereof will be omitted. Non-verbal information obtained as the user's reaction may include facial expression, complexion, corners of the mouth, eye movement, blinking, wrinkles between the eyebrows, or gaze direction, tone, pitch, tempo, rhythm, or sound quality (smooth, husky, etc.) of voice, or body movement (nodding, fidgeting, etc.), gestures (body movements), mannerisms (crossing arms, touching the tip of the nose, stroking hair, etc.), or breathing (pitch, depth, etc.), which may be determined from the voice or image at the time of response and stored in the dialogue content DB 36. Alternatively, these non-verbal information may be combined to determine the user's emotions.

[0047] FIG. 14 is an explanatory diagram showing an example of a dialogue evaluation DB. Evaluations for the dialogue evaluation DB 37 are stored. The dialogue evaluation DB 37 includes an evaluation ID column, a preset column, a log ID column, a classification column, an evaluator column, an evaluation column, a total evaluation column, and an individual evaluation column. The evaluation ID column stores an evaluation ID that uniquely identifies an evaluation. The preset column stores the ID of the preset used during the dialogue. The log ID column stores the log ID that identifies the content of the dialogue. The classification column stores the classification of the dialogue. "Pre" indicates that the dialogue took place for pre-evaluation of the preset. "Post" indicates that the dialogue took place for the purpose of post-evaluation. The evaluator column stores the ID of the professional who performed the evaluation. The evaluation column stores the evaluation of the entire dialogue. The evaluation is, for example, on a five-point scale, and the evaluation column stores the numerical value of the evaluation. The total evaluation column stores a numerical value indicating the overall evaluation. Multiple options are provided to select as the overall evaluation, and the total evaluation column stores a numerical value indicating the option selected by the evaluator. The options are, for example, "just the right length," "good flow of speech," "stiff expression," "roundabout," etc. The individual evaluation column stores the evaluation of each message issued by the dialogue terminal 4. The individual evaluation column includes a sequence number column and a comment column. The sequence number column stores a sequence number that identifies the message. The comment column stores a comment entered as an evaluation of the message.

[0048] FIG. 15 is an explanatory diagram showing an example of a preset status DB. The preset status DB 38 stores the status of a preset. The preset status DB 38 includes an ID column and a latest column. The ID column stores an ID that identifies the preset. The latest column stores the status of the preset. The stored statuses include, for example, creating, pre-interaction, pre-evaluation, pre-evaluation completed, editing, post-evaluation, updating, update completed, and in use. "Creating" indicates that a new scenario is being created. "Pre-interaction" indicates that a dialogue with the digital twin is waiting or in progress to pre-evaluate the new scenario. "Pre-evaluation" indicates that the dialogue with the digital twin regarding the new scenario has been completed and pre-evaluation is being performed. "Pre-evaluation completed" indicates that the pre-evaluation of the new scenario has ended. "Editing" indicates that the new scenario is being edited based on the pre-evaluation. "Post-evaluation" indicates that a dialogue using the new scenario is being performed with the care recipient for evaluation. Assuming that the new scenario may have defects, the care recipient to be interacted with during the post-evaluation must be carefully selected. Additionally, in case an unexpected situation occurs during the conversation, such as the care recipient expressing intense emotions, it is desirable to have an experienced professional present at the conversation so that any unexpected events can be collected immediately. "Updating" indicates that a new scenario is being updated based on the results of the post-evaluation. "Updated" indicates that the new scenario has been updated. "In use" indicates that the new scenario has been completed and is being put into practical use.

[0049] FIG. 16 is an explanatory diagram showing an example of a pattern DB. The pattern DB 39 stores type data of care recipients. The type data is used when classifying care recipients. The type data is also used as attribute data for digital twins. The pattern DB 39 includes a pattern ID column, a level of care required column, a living arrangement column, an age group column, and a personality type column. The pattern ID column stores a pattern ID that identifies the type. The level of care required column stores the level of care or support required by the care recipient. For example, the level of care required column stores levels 1 to 5, or levels 1 or 2. The living arrangement column stores the living arrangement of the care recipient. It stores information about the building in which the care recipient lives. For a general residence, the living arrangement column stores detached house, apartment, condominium, etc. For elderly care facilities, the following types are stored: public facilities such as special nursing homes for the elderly, nursing homes for the elderly, nursing care facilities for the elderly, nursing care hospitals, or care houses; and private facilities such as paid nursing homes with care, paid residential nursing homes, paid health care homes for the elderly, serviced housing for the elderly, group homes, or condominiums for the elderly. Note that "nursing home" in Figure 16 is a common name for nursing care facilities for the elderly. The "living style" column can also store the common name of the facility, but only one commonly established name should be used. The age group column stores age. The personality type column stores the personality type of the care recipient. For example, we use the personality types proposed by American psychologist Suzanne Reichardt. Type 1 is the mature type. Mature care recipients are content with accepting the fact of aging. Type 2 is the armchair type. Armchair care recipients are free from responsibilities and seeking a comfortable life. Type 3 is the self-protective type. Self-defensive care recipients try to cope with anxiety about aging by putting up a strong defensive system. The fourth type is the blame-resentment type. Care recipients of the blame-resentment type blame others instead of accepting the fact of aging. They have no hobbies and are particularly anxious and afraid of death. The fifth type is the self-loathing type. Care recipients of the self-loathing type look back on the past and blame themselves. They have little interest in others, are isolated, and see death as a release from an unbearable existence and are not afraid. From the five types, indicate the type that the care recipient fits into and memorize the number, and the personality type column.

[0050] Next, the flow of scenario creation and updating in the dialogue system 100 will be described. Figure 17 is a flowchart showing the workflow for scenario creation and updating. First, a new scenario is created (step S1). The professional selects questions to ask for each theme, such as diet, physical condition, activity, and excretion. Questions about the selected items are considered. The care recipient's responses to the questions are anticipated. Responses are anticipated not only for responses that provide answers to the questions, but also for responses that do not provide answers. In the case of responses that do not provide answers, additional or supplementary questions are considered. The order in which the questions will be asked is determined, and the new scenario is completed. Once the scenarios for each theme are completed, new preset scenarios including multiple themes are created. The created new scenarios are stored in the scenario DB 33, and the new preset scenarios are stored in the preset DB 34. A pre-evaluation of the created new scenarios is performed (step S2). In the pre-evaluation, the dialogue system 100 uses the created scenarios to interact with the digital twin. The content of the interaction with the digital twin is stored in the interaction content DB 36. The dialogue with the digital twin is used to identify areas for improvement in the new scenario. The dialogue partner may not be a digital twin, but a professional. When the dialogue partner is a professional, improvements to the new scenario are identified through dialogue using the scenario. The new scenario is edited based on the improvements (step S3), and the new scenario to be used in operation is tentatively completed. A post-evaluation of the new scenario is performed (step S4). Unlike the pre-evaluation, the post-evaluation is performed by conducting a dialogue with the care recipient using the new scenario. In the post-evaluation, it is desirable to conduct dialogue with multiple care recipients. After sufficient post-evaluation has been performed, the scenario is updated (step S5). The new scenario is then completed and becomes the scenario to be used in actual operation. The post-evaluation (step S4) and scenario update (step S5) may be performed repeatedly. The pre-evaluation (step S2) and scenario editing (step S3) may also be performed repeatedly. Furthermore, after the scenario update (step S5), a dialogue with the digital twin may be performed to confirm whether the update is appropriate.

[0051] Next, information processing performed by the dialogue system 100 will be described. First, dialogue processing will be described. Dialogue processing is processing executed when the dialogue device dialogues with the care recipient via the dialogue terminal 4. When a dialogue is to be conducted, it is necessary to set the ID of the preset scenario to be used in the dialogue. The ID is set in the dialogue terminal 4 before the dialogue starts. A professional such as a care manager sets the ID at the request of the care recipient's family or the like. A visiting care worker (home helper) who visits the care recipient daily may set the ID at the time of the visit based on the professional's request. Also, instead of setting it in the dialogue terminal 4, it may be stored in the dialogue device 1 or the database 3. In this case, the ID of the preset scenario, the user ID of the care recipient who is the dialogue partner, and the scheduled date or time of the dialogue are associated and stored in the auxiliary storage unit 13 of the dialogue device 1 or the database 3. A preset scenario can be selected by referring to the contents of the theme name column and the description column of the preset DB 34.

[0052] 18 is a flowchart showing an example of the procedure for dialogue processing. The control unit 41 of the dialogue terminal 4 transmits settings to the dialogue device 1 (step S21). The settings include a user ID, etc. If an ID of a preset scenario to be used in the dialogue is set, the ID is also included. The control unit 11 of the dialogue device 1 receives the settings and stores them in the dialogue attribute DB 35 (step S22). The control unit 11 acquires the preset scenario to be used in the dialogue (step S23). If the ID of the preset scenario has not been received from the dialogue terminal 4, the control unit 11 acquires the ID from the auxiliary storage unit 13 or the database 3. The control unit 11 selects a message (step S24). The control unit 11 transmits the selected message to the dialogue terminal 4 (step S25). The control unit 41 of the dialogue terminal 4 receives the message (step S26). The control unit 41 outputs the message as voice from the voice output unit 49 (step S27). The control unit 41 acquires the care recipient's response as video and audio from the imaging unit 47 and the audio acquisition unit 48, and transmits them to the interactive device 1 (step S28). The control unit 11 of the interactive device 1 receives the care recipient's response (step S29). The control unit 11 analyzes the care recipient's acceptance or rejection of the question and the care recipient's emotions at the time of response from the response image and response audio (step S30). As a result of the emotion analysis, the control unit 11 determines which of the six patterns the care recipient's response falls into. The control unit 11 consults the scenario based on the determined pattern and determines whether or not to end the dialogue (step S31). If the control unit 11 determines not to end the dialogue (NO in step S31), the process proceeds to step S24. The control unit 11 consults the scenario and selects the next message (step S24). The control unit 11 executes step S25 and subsequent steps. When the control unit 11 determines to end the dialogue (YES in step S31), it stores a record of the dialogue in the dialogue attribute DB 35 and the dialogue content DB 36 (step S32). The dialogue content may be stored in the dialogue content DB 36 as needed during the dialogue. For example, every time a response is received from the care recipient to a message, it is stored in the dialogue content DB 36. The control unit 11 ends the process.

[0053] The dialogue process shown in Figure 18 is the process when dialogue with a care recipient, but the process is similar when dialogue with a digital twin in the pre-evaluation of a scenario. Dialogue with a digital twin is carried out by setting the preset scenario to be used and the ID of the digital twin with which the dialogue will be conducted. The digital twin's response can be determined by acquiring video and audio and determining whether or not it agrees or disagrees and its emotions, but it is also possible to acquire the agreement and emotions directly from the computer running the digital twin.

[0054] Pre-evaluation will now be described. As described above, pre-evaluation involves evaluating a scenario by referring to the results of an interaction with a digital twin. Figure 19 is a flowchart showing an example of the procedure for pre-evaluation processing. Pre-evaluation is performed by a professional such as a care manager using the professional terminal 5. The professional instructs the professional terminal 5 to perform pre-evaluation. This triggers the control unit 51 of the professional terminal 5 to request a list of scenarios awaiting pre-evaluation from the editing device 2 (step S51). The control unit 21 of the editing device 2 receives the request (step S52). The control unit 21 creates a list of preset scenarios to be pre-evaluated and sends it to the professional terminal 5 (step S53). The control unit 21 references the preset status DB 38 and acquires the IDs of preset scenarios whose latest column value is "pre-evaluation in progress." Based on the acquired ID, the control unit 21 searches the preset DB 34, acquires information on preset scenarios to be pre-evaluated, and creates a list. The control unit 51 of the professional terminal 5 receives the list and displays it on the display unit 56 (step S54). The professional selects a scenario to be evaluated. The control unit 51 accepts the selection (step S55). The control unit 51 transmits the selection information to the editing device 2 (step S56). The control unit 21 of the editing device 2 receives the selection information (step S57). The selection information includes the ID of the preset scenario. The control unit 21 acquires the dialogue history from the dialogue attribute DB 35 and the dialogue content DB 36 (step S58). The control unit 21 creates a screen displaying the dialogue history and transmits it to the professional terminal 5 (step S59). The control unit 51 of the professional terminal 5 receives the screen and displays it on the display unit 56 (step S60). In the pre-evaluation, the content of the dialogue with the digital twin is displayed like the display screen of a chat application or a message application. That is, the message output by the dialogue device and the digital twin's response to that message are displayed in text. The professional refers to the displayed dialogue history and inputs an evaluation. The control unit 51 accepts the input (step S61). Step S61 may be executed multiple times depending on the number of operations performed by the professional. When the professional has finished the evaluation, he / she inputs "Complete." The control unit 51 accepts the evaluation completion (step S62). The control unit 51 transmits the input contents (evaluation) of the professional to the editing device 2 (step S63).The control unit 21 of the editing device 2 receives the evaluation (step S64), stores the evaluation in the dialogue evaluation DB 37 (step S65), and ends the process.

[0055] It is desirable that the pre-assessment be carried out by multiple professionals. The number of professionals who have made the assessment can be tallied in the dialogue assessment DB 37, so once a predetermined number of professionals have made the assessment, the pre-assessment is completed, and the value of the latest column in the preset status DB is rewritten to "pre-assessment completed."

[0056] Scenario updating will now be described. As described above, scenario updating refers to updating a scenario by referring to the results of dialogue with the care recipient. The timing for completing the post-evaluation and starting a scenario update may be determined appropriately based on the number of dialogues, the number of care recipients with whom dialogues have been conducted, the period of time elapsed since use began, etc. Alternatively, the conditions for starting a scenario update may be stored in the database 3, and a scenario that satisfies the conditions may be searched for in a daily batch, or whether the conditions are met when a scenario is used may be determined, and the update may be started when the target scenario is found.

[0057] FIG. 20 is a flowchart showing an example of the update processing procedure. Scenario updates are performed by professionals such as care managers, but preferably by certified care managers or head care managers with ample experience and knowledge. The professional instructs the professional terminal 5 to update the scenario. This triggers the control unit 51 of the professional terminal 5 to request a list of scenarios to be updated from the editing device 2 (step S81). The control unit 21 of the editing device 2 receives the request (step S82). The control unit 21 creates a list of scenarios to be updated and sends it to the professional terminal 5 (step S83). The control unit 21 references the preset status DB 38 and acquires the ID of a preset scenario whose latest column value is "updating." Based on the acquired ID, the control unit 21 searches the preset DB 34, acquires information on the preset scenarios to be updated, and creates a list. The control unit 51 of the professional terminal 5 receives the list and displays it on the display unit 56 (step S84). The professional selects a scenario to update. The control unit 51 accepts the selection (step S85). The control unit 51 transmits the selection information to the editing device 2 (step S86). The control unit 21 of the editing device 2 receives the selection information (step S87). The selection information includes the ID of the preset scenario. The control unit 21 acquires the dialogue history from the dialogue attribute DB 35 and the dialogue content DB 36 (step S88). The control unit 21 tally up the responses of the care recipients in the units or main states that make up the scenario to be evaluated (step S89). The control unit 21 creates an editing screen including the tallying results and transmits it to the professional terminal 5 (step S90). The control unit 51 of the professional terminal 5 receives the editing screen (step S91). The control unit 51 displays the editing screen on the display unit 56 (step S92). The editing screen displays the scenario as a tree. Each link indicating a care recipient's response displays the percentage of care recipients who made that response among all care recipients who interacted in the post-evaluation. The professional refers to the displayed dialogue history to edit the scenario, etc. The control unit 51 accepts operation input (step S93). The control unit 51 determines whether the received input is a change in the profile conditions (step S94). The proportion displayed for each link can be limited to the attributes (user profile) of the care receiver.Here, the change in profile conditions refers to a change in the attributes of the care recipients to be the subject of the percentage display. When the control unit 51 determines that the received input is a change in profile conditions (YES in step S94), it transmits the changed conditions to the editing device 2 (step S95). The control unit 21 of the editing device 2 receives the conditions and changes the conditions of the care recipients to be the subject of response aggregation (step S96). The control unit 21 moves the process to step S89. For example, when a professional inputs a designated level of care need, the care recipients with the designated level of care need are counted as all care recipients, and the percentage of care recipients who made each response is aggregated. When the control unit 51 determines that the received input is not a change in profile conditions (NO in step S94), it determines whether the received input is an update completion (step S97). When the control unit 51 determines that the received input is not an update completion (NO in step S97), it moves the process to step S92 and displays the editing screen with the content updated according to the input. If the control unit 51 determines that the received input indicates that the update is complete (YES in step S97), it transmits the updated content (correction proposal) to the editing device 2 (step S98). The control unit 21 of the editing device 2 receives the updated content (step S99). The control unit 21 stores the updated content in the scenario DB 33 (step S100). The update process ends. In the update process, the updated content may be stored as a new version of the scenario rather than overwriting the scenario DB 33. The scenario DB 33 stores both the scenario before the update and the scenario after the update.

[0058] Next, the screens displayed on the professional terminal 5 will be described. FIG. 21 is an explanatory diagram showing an example of a pre-evaluation screen. The pre-evaluation screen d01 is a screen used when performing a pre-evaluation of a preset scenario. The pre-evaluation screen d01 shown in FIG. 21 shows an example in which a professional performs a pre-evaluation while engaging in a dialogue. The pre-evaluation screen d01 includes a scenario list d011, a control area d012, an outgoing message d013, a reply message d014, an evaluation value field d015, a general evaluation list d016, and a comment field d017. The scenario list d011 displays a list of scenarios awaiting pre-evaluation. A single scenario is selected using a check box. The control area d012 is operated to start, stop, etc. the dialogue. The outgoing message d013 is a message from the dialogue system 100. The reply message d014 is a message entered by the professional. The message is entered by voice or text. The evaluation value field d015 is used to enter an evaluation of the entire preset scenario on a five-point scale. The overall evaluation list d016 displays options for expressing the overall evaluation in a subjective manner. Professionals check the appropriate options. In the comment field d017, they input correction comments for individual messages. In Figure 21, in response to the message "I see. That's good," from the dialogue system 100, the comment "It seems like you're evaluating it. 'Excellent' is better than 'good'" is entered. When displaying a scenario that has undergone pre-evaluation on a screen similar to the pre-evaluation screen d01 shown in Figure 21, the names of the professionals who entered messages, selected overall evaluations, or entered correction comments may be displayed. Furthermore, correction comments entered by general professionals may be accessible to all professionals, regardless of region or professional level. Furthermore, correction instructions, correction suggestions, and comments are acquired from professionals. Instructions may represent decisions, suggestions may literally represent proposals, and comments may include wishes, questions, criticisms, impressions, encouragement, praise, etc.

[0059] The pre-assessment screen d01 when a dialogue for pre-assessment is conducted with a digital twin is the same as that shown in FIG. 21. In this case, the control area d012 is not displayed. Also, all outgoing messages d013 and reply messages d014 from the beginning to the end of the dialogue are displayed. The professional enters an overall comment and comments on the outgoing messages. In addition, in the post-assessment, after conducting a dialogue with the care recipient, the care manager may evaluate the scenario for the dialogue using a screen similar to the pre-assessment screen d01.

[0060] FIG. 22 is an explanatory diagram showing an example of a scenario update screen. The scenario update screen d02 is a screen used when updating a scenario after a post-evaluation of a preset scenario. The scenario update screen d02 includes a scenario list d021, evaluation results d022, a profile specification field d023, a dialogue history display d024, and a type selection field d025. The scenario list d021 displays a list of scenarios waiting to be updated. The evaluation results d022 display the number of dialogues performed using the preset scenario and evaluations from general professionals. Here, general professionals refer to care managers and others who do not have the authority to update scenarios. For evaluations from general professionals, a screen obtained by modifying the above-mentioned pre-evaluation screen d01 for post-evaluation is used. Alternatively, the scenario update screen d02 may be modified to create a screen that does not allow scenario updates but allows comments to be added to the scenario and evaluation values to be entered, and this screen may be used. The profile specification field d023 is used when you want to narrow down the statistics (aggregated values) displayed in the dialogue history display d024 by the profile of the care recipient, and check the profile you want to narrow down. The dialogue history display d024 displays the flow of preset dialogues in a tree format and the response rate of the care recipient. The dialogue history display d024 includes an emotion icon d0241, an answer rate d0242, and an update icon d0243. The emotion icon d0241 represents the most common emotional reaction shown by the care recipient as a result of the dialogue. The answer rate d0242 represents the rate of responses from the care recipient by pattern as a percentage and link thickness as a result of the dialogue. The update icon d0243 is used to select the target node when replacing the message (question) of each node (unit or main state) or modifying the expression. The type selection field d025 is used to search for and list the preset scenarios to be updated by theme or target person. When displaying an updated preset scenario on a screen similar to the scenario update screen d02 shown in Figure 22, it may be possible to make it clear which staff member's correction instructions, suggestions, and comments were the basis for the update.

[0061] This embodiment has the following advantages. Dialogue using a scenario makes it possible to select appropriate messages in response to the reactions of the care recipient. Furthermore, when creating a scenario, a general care manager conducts a preliminary evaluation, so that even professionals who are not capable of creating a scenario can be involved. Since the scenario can be updated by referring to the history of actual dialogue, it is possible to create a highly refined scenario.

[0062] It is desirable to create multiple preset scenarios per theme, rather than just one. This is because the appropriate way to proceed with the dialogue varies depending on the level of care required, living arrangement, age group, or personality type of the care recipient. Therefore, for each theme, a maximum of as many scenarios as there are patterns of care recipients defined in the pattern DB 39 should be prepared. If the professional specifies the purpose of the dialogue and the pattern (attributes) of the care recipient before the dialogue, the dialogue device 1 can select the appropriate preset scenario.

[0063] (Incentive granting function) In order to motivate general professionals who do not have the authority to edit scenarios to actively participate in creating scenarios, incentives may be given for tasks such as pre-evaluations. FIG. 23 is an explanatory diagram showing an example of an incentive DB. The incentive DB 3A includes a professional ID column, a pre-evaluation column, a comment column, an adoption column, and a post-evaluation column. The professional ID column stores the professional ID of the professional to whom an incentive is given. The pre-evaluation column stores the incentive for the pre-evaluation. The comment column stores the incentive for providing a comment in the pre-evaluation. The adoption column stores the incentive for adopting a comment provided in the pre-evaluation when editing the scenario. Whether or not to grant an incentive is input by the senior professional who performs the editing when editing the scenario. The post-evaluation column stores the incentive for participating in the post-evaluation. Note that these incentives may also be given to senior professionals. The values stored in the incentive DB 3A may not be the incentives themselves, but may instead store the number of times pre-evaluations have been participated in, the number of comments provided, the number of comments adopted, and the number of times post-evaluations have been participated in. Incentives may also be given differently depending on the content. For example, participating in a pre-evaluation may be worth 1 point, leaving a comment may be worth 2 points, adopting a comment may be worth 5 points, and participating in a post-evaluation may be worth 3 points. Professionals who accumulate a certain number of points may be awarded a prize. Points may also be given monetary value and may be used to pay for goods or services.

[0064] After updating the scenario, the improvement in the functionality of the scenario before and after the revision may be measured. For example, the ease with which the care recipient can respond may be evaluated based on the speed at which the care recipient can respond and their facial expressions. If it is confirmed that the revision has improved the functionality of the scenario, an additional incentive may be given to the professionals involved in the update.

[0065] The present invention has the effect of enabling the caregiver to recognize the areas that need correction, or to receive evaluations from the care recipient, thereby motivating the caregiver to continually revise the dialogue scenario and finding a sense of accomplishment in the work.

[0066] The technical features (constituent elements) described in each embodiment can be combined with each other, and by combining them, new technical features can be formed. The embodiments disclosed herein are to be considered as illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0067] 100 Dialogue Systems 10 Serving Clusters 1. Interactive device 11 Control section 12 Main memory 13 Auxiliary storage 14 Communications Department 15 Reading unit 16 Display 1P control program 1a Portable storage media 1b semiconductor memory 2 Editing equipment 21 Control section 22 Main memory 23 Auxiliary storage 24 Communications Department 25 Reading unit 3 Database 31 User DB 32 Professional DB 33 Scenario DB 34 Preset DB 35 Dialogue Attribute DB 36 Dialogue Content DB 37 Dialogue Evaluation DB 38 Preset Status DB 39 Pattern DB 3A Incentive DB 4 Interactive terminal 41 Control Unit 42 Main memory 43 Auxiliary storage section 44 Communications Department 45 Input section 46 Display section 47 Imaging unit 48 Voice Acquisition Unit 49 Audio output section 4P control program 5 Professional terminals 51 Control section 52 Main memory 53 Auxiliary storage section 54 Communications Department 55 Input section 56 Display section 57 Voice Acquisition Unit 58 Audio output section 5P control program B Bus N Network

Claims

1. The computer Acquire a dialogue scenario with the care recipient, Reading a care recipient model corresponding to the care recipient from a plurality of care recipient models constructed in the digital twin; outputting, to a caregiver, a first dialogue history including a message based on the dialogue scenario, and a response and non-verbal information from the care receiver model to the message; Editing the dialogue scenario based on the correction instructions, correction suggestions, or comments for the dialogue scenario received from the employee; outputting, to a caregiver, a second dialogue history including a message based on the edited dialogue scenario, and the care recipient's response to the message and non-verbal information; updating the edited dialogue scenario based on correction instructions, correction suggestions, or comments for the edited dialogue scenario received from the employee; outputting a message to the care recipient based on the updated dialogue scenario; Acquire the care recipient's response and non-verbal information; selecting a message according to the acquired answer and non-verbal information based on the updated dialogue scenario; The selected message is output to the care recipient. An information processing method comprising:

2. Based on the non-verbal information, an emotion analysis of the care recipient is performed, and a message corresponding to the answer and the emotion of the care recipient is selected.

2. The information processing method according to claim 1,

3. Acquire attributes of the care recipient; A plurality of dialogue scenarios are prepared, and the dialogue scenario to be used is selected based on the acquired attributes; The message is selected using the obtained dialogue scenario.

3. The information processing method according to claim 1 or 2.

4. Get the topic of the dialogue, A dialogue scenario corresponding to each of a plurality of subjects is prepared, and the dialogue scenario to be used is acquired based on the acquired subject; The message is selected using the obtained dialogue scenario.

4. The information processing method according to claim 1, wherein the information processing method further comprises:

5. outputting to a second employee any instruction, suggestion, or comment for modifying the dialogue scenario received from the first employee; acquiring a correction instruction, a correction suggestion, or a comment for the dialogue scenario received from the second worker; The dialogue scenario is updated based on the acquired correction instruction, correction suggestion, or comment.

5. The information processing method according to claim 1, wherein the information processing method comprises:

6. Identifying a portion of the dialogue scenario that should be corrected based on the correction instruction, the correction suggestion, or the comment; The correction instruction, the correction suggestion or the comment, and the part to be corrected are output in association with each other.

6. The information processing method according to claim 1, wherein the information processing method further comprises:

7. A predetermined incentive is given to the worker who created the correction instruction, the correction suggestion, or the comment.

7. The information processing method according to claim 5 or 6.

8. A computer comprising: outputting a message to the care recipient based on a dialogue scenario with the care recipient; Acquire the care recipient's response and non-verbal information; selecting a message according to the acquired answer and non-verbal information based on the dialogue scenario; outputting the selected message to the care recipient; storing the message, and a dialogue history including the care recipient's response to the message and non-verbal information in a storage unit; outputting the dialogue history stored in the storage unit to a first employee; acquiring a correction instruction, a correction suggestion, or a comment on the dialogue scenario received from the first worker; outputting the correction instruction, the correction suggestion, or the comment received from the first worker to a second worker; acquiring a correction instruction, a correction suggestion, or a comment for the dialogue scenario received from the second worker; updating the dialogue scenario based on the correction instruction, correction suggestion, or comment received from the second worker; A predetermined incentive is given to the first worker or the second worker who created the correction instruction, the correction suggestion, or the comment. An information processing method comprising:

9. On the computer, Acquire a dialogue scenario with the care recipient, Reading a care recipient model corresponding to the care recipient from a plurality of care recipient models constructed in the digital twin; outputting, to a caregiver, a first dialogue history including a message based on the dialogue scenario, and a response and non-verbal information from the care receiver model to the message; Editing the dialogue scenario based on the correction instructions, correction suggestions, or comments for the dialogue scenario received from the employee; outputting, to a caregiver, a second dialogue history including a message based on the edited dialogue scenario, and the care recipient's response to the message and non-verbal information; updating the edited dialogue scenario based on correction instructions, correction suggestions, or comments for the edited dialogue scenario received from the employee; outputting a message to the care recipient based on the updated dialogue scenario; Acquire the care recipient's response and non-verbal information; Selecting a message according to the acquired answer and non-verbal information based on the updated dialogue scenario; The selected message is output to the care recipient. An information processing program characterized by causing a process to be performed.

Images