Information processing system, information processing method, and program

Abstract

To provide a technique for generating a question for obtaining a message with expressions customized according to at least one piece of information regarding age, gender, race or ethnicity, and region of a user from a machine learning system.SOLUTION: An information processing system comprising one or more processors is provided, the one or more processors being configured to receive at least one piece of information regarding age, gender, race or ethnicity, and region of a user using a lifestyle improvement support service, and generate a question for instructing a machine learning model trained at least on expressions corresponding to information to change the expressions to those appropriate for users identified by the received information.SELECTED DRAWING: Figure 6

Description

[Technical field]

[0001] The present disclosure relates to an information processing system, an information processing method, and a program. [Background technology]

[0002] Currently, U.S. companies are showing increasing interest in services aimed at helping prevent, diagnose and treat lifestyle-related diseases. Summary of the Invention [Problem to be solved by the invention]

[0003] Providing appropriate information and support according to the user's situation is important for establishing healthy lifestyle habits. However, with current services, a message is prepared for each piece of information to be conveyed. However, even with the same text, the impression received by users may differ depending on their age and gender, and this difference in impression may affect their motivation to improve their lifestyle habits.

[0004] One aspect of the present disclosure provides a technique for generating a query to obtain a message from a machine learning system whose wording is customized according to at least one piece of information related to a user's age, gender, race or ethnicity, and region. [Means for solving the problem]

[0005] As one embodiment of the present disclosure, there is provided an information processing system having one or more processors, which accepts at least one piece of information related to the age, sex, race or ethnicity, and region of a user who uses a lifestyle habit improvement support service, and generates a question for instructing a machine learning model that has at least learned an expression according to the information to change the expression to an expression suitable for the user group identified by the accepted information.

[0006] As another aspect of the present disclosure, there is provided an information processing method in which a computer executes a process of accepting at least one piece of information related to the age, sex, race or ethnicity, and region of a user who uses a lifestyle habit improvement support service, and a process of generating a question sentence for instructing a machine learning model that has at least learned an expression corresponding to the information to change the expression to an expression suitable for a user group identified by the accepted information.

[0007] As another aspect of the present disclosure, a program is provided for causing a computer to perform the following functions: accepting at least one piece of information relating to the age, sex, race or ethnicity, and region of a user who uses a lifestyle habit improvement support service; and generating a question for instructing a machine learning model that has learned at least an expression according to the information to change the expression to an expression suitable for a user group identified by the accepted information. Effect of the Invention

[0008] According to one aspect of the present disclosure, a question can be generated to obtain a message from a machine learning system whose wording is customized according to at least one piece of information related to the user's age, gender, race or ethnicity, and region. [Brief description of the drawings]

[0009] [Figure 1] FIG. 2 is a diagram illustrating an example of the configuration of an information processing system assumed in the first embodiment. [Diagram 2] FIG. 2 is a diagram illustrating a hardware configuration of the support system. [Diagram 3] FIG. 2 is a diagram illustrating the hardware configuration of a user terminal. [Figure 4] 1 is a sequence diagram illustrating an overview of a support service provided through an information processing system. [Diagram 5] FIG. 11 is a sequence diagram illustrating a procedure for creating a message whose expression is customized for each user. [Figure 6] FIG. 11 is a diagram illustrating an example of a question generated by the question generation system. [Figure 7]13 is an example of an answer sentence output by the question answering system. [Figure 8] 13 is another output example of an answer sentence output by the question answering system. [Figure 9] 13 is another output example of an answer sentence output by the question answering system. [Figure 10] FIG. 13 is a diagram illustrating an example of sending a customized message. [Figure 11] 13 is another output example of an answer sentence output by the question answering system. [Figure 12] 13 is another output example of an answer sentence output by the question answering system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. <Embodiment 1>

[0011] <System configuration> FIG. 1 is a diagram illustrating an example of the configuration of an information processing system 1 assumed in the first embodiment. The information processing system 1 shown in FIG. 1 is composed of a lifestyle habit improvement support system (hereinafter referred to as the "support system") 10 that provides a lifestyle habit improvement support service (hereinafter referred to as the "support service"), a question answering system 20 that accepts questions in natural language and outputs answers to them, a user terminal 30 operated by a user who uses the support service, and a network N that connects these terminals.

[0012] The support system 10 in this embodiment is a system that distributes an application that supports improvement of lifestyle habits (hereinafter referred to as an “support application”), stores user data input through the support application, and provides information that supports improvement of lifestyle habits. The support system 10 in this embodiment is operated by a company, organization, etc., where the user works. However, the support system 10 may be operated by an entity, etc., that is entrusted by the company, organization, etc., where the user works.

[0013] The assistance application in this embodiment is installed in the user terminal 30 and used to record the user's daily life. The record of daily life includes, for example, blood pressure value, weight, pulse rate, behavior history, and feelings and moods. The support system 10 in this embodiment has a message generation system 11 and a question generation system 12 in relation to the function of providing information to support improvement of lifestyle habits.

[0014] The message generation system 11 is a computer system that generates messages addressed to individual users. In this embodiment, the message generation system 11 is configured with a single server. However, the message generation system 11 may be configured with multiple servers that cooperate with each other.

[0015] The messages here include, for example, email addressed to an email address, SNS (=Social Networking Service) addressed to an account, and SMS (=Short Message Service) addressed to a mobile phone number. The message generation system 11 generates various messages (hereinafter referred to as "first messages") that contribute to improving lifestyle habits, based on user data.

[0016] The message generation system 11 transmits to the target user a message (hereinafter referred to as a "second message") in which the expression of the first message is changed according to at least one of the age, sex, race or ethnicity, and region of the destination user. The expression here is, for example, a tone or manner. In the case of this embodiment, the age, sex, race or ethnicity, and region are all registered in advance in the support system 10. The region may be, for example, the user's place of residence or place of work, or may be specified by the administrative division in which these are included. The region may be a country, a region, or the like. Incidentally, the number of regions is not limited to one, and multiple regions may be registered. For example, two regions, the place of residence and the place of work, may be registered. The second message here is obtained by providing the question generated by the question generation system 12 and the first message to the question answering system 20.

[0017] Question generation system 12 is a computer system that generates questions to be given to question answering system 20. In this embodiment, question generation system 12 is configured with a single server. However, question generation system 12 may be configured with multiple servers that cooperate with each other. In this embodiment, the question generation system 12 generates a question for instructing a change of a message to an expression suitable for a user group specified by the age, sex, race or ethnicity, and region of the user who is the destination of the message. Note that information on the user's age, sex, race or ethnicity, and region is provided to the question generation system 12 from the message generation system 11 as a question.

[0018] The question answering system 20 is a computer system that provides a so-called natural language processing service. The question answering system 20 is not limited to a dedicated AI chatbot system, and may be a general-purpose AI chatbot system. The question answering system 20 in this embodiment has a function of using a machine learning model that has learned expressions according to age, gender, race or ethnicity, and region, and changing the expression of the first message provided from the message generation system 11 to an expression according to the age, gender, race or ethnicity, and region of the user specified in the question text, and outputting it as an answer text.

[0019] However, the machine learning model does not need to learn expressions for all combinations of age, gender, race or ethnicity, and region. That is, the machine learning model may be a model that has learned expressions according to any one of age, sex, race or ethnicity, and region, or any combination of these. For example, the machine learning model may be a model that has learned expressions according to a combination of age and sex. For example, the machine learning model may be a model that has learned expressions according to race or ethnicity. For example, the machine learning model may be a model that has learned expressions according to the region in which the user lives or works.

[0020] It should be noted that the machine learning model does not have to be a model that has learned only expressions corresponding to any one of age, sex, race or ethnicity, and region, or any combination of these. For example, the machine learning model may have learned various conversations, etc., in addition to the expressions described above.

[0021] The question may be generated according to the machine learning model used to generate the message, because, for example, instructing a machine learning model that has learned expressions according to combinations of age and gender to change expressions according to race or ethnicity will not produce the desired results. However, in that case, multiple machine learning models with different learned information may be used. For example, if there is a machine learning model A that has learned expressions according to combinations of age and gender, and a machine learning model B that has learned expressions according to region, the first message and a question (instructing a change in expression according to a combination of age, gender, and region) may be given to machine learning model A to obtain the first second message, and then the first second message and a question (instructing a change to an expression according to age, gender, and region) may be given to machine learning model B to obtain the second second message.

[0022] Alternatively, two questions may be prepared according to a plurality of machine learning models, and a second message may be generated that targets each question. For example, if there is a machine learning model A that has learned expressions according to a combination of age and gender, and a machine learning model B that has learned expressions according to region, the first message and a question (a question instructing the model to change to an expression according to age and gender) may be given to machine learning model A to obtain the first second message, and then the first second message and a question (a question instructing the model to change to an expression according to region) may be given to machine learning model B to obtain the second second message.

[0023] The user terminal 30 is a computer terminal in which a support application is installed. The user terminal 30 in this embodiment is configured, for example, by a smartphone, a tablet computer, a notebook computer, or a desktop computer. The user terminal 30 is capable of communicating with the support system 10 through the network N. The user terminal 30 uploads user data input through a support application to the support system 10, and receives support services from the support system 10. The network N is, for example, a LAN (Local area network), the Internet, or a mobile communication system (4G, 5G).

[0024] <System hardware configuration> 2 is a diagram illustrating a hardware configuration of the support system 10. Note that the message generation system 11 and the question generation system 12 constituting the support system 10 also have the same hardware configuration as the support system 10.

[0025] The support system 10 has a processor 101, a ROM (=Read Only Memory) 102 in which a BIOS (=Basic Input Output System) and the like are stored, a RAM (=Random Access Memory) 103 used as a work area for the processor 101, an auxiliary storage device 104, and a communication interface 105. Each device is connected via a bus and other signal lines 106.

[0026] The processor 101 is a device that realizes various functions by executing a program. In this embodiment, the support application and the OS (=Operating System) are collectively called a program. The processor 101, the ROM 102, and the RAM 103 function as a computer. The auxiliary storage device 104 is configured, for example, by a hard disk device or semiconductor storage. The communication interface 105 is an interface for communicating with other servers and terminals via the network N. The communication interface 105 supports Ethernet (registered trademark), Wi-Fi (registered trademark), mobile communication systems, and other communication standards.

[0027] <Data stored in the auxiliary storage device> <Support system> The auxiliary storage device 104 of the assistance system 10 stores the assistance application as well as user data collected through the application. The user data includes, for example, the user's attributes (gender, date of birth, age, height, weight), blood pressure value, activity record, mood record, physical condition record, medical examination history, operation history of the lifestyle habit improvement support app, biological characteristics, psychological characteristics, social characteristics, habits, goal achievement status, etc. However, the user data does not need to be all of this information, and may be only a part of it.

[0028] The blood pressure values ​​are, for example, systolic blood pressure (i.e., the highest blood pressure value) and diastolic blood pressure (i.e., the lowest blood pressure value). The activity record is, for example, a usage history of a medical application, a medication record, a food record, a smoking record, a drinking record, and an exercise record. The activity record is an example of information about an activity. The record of taking medicine is, for example, the type and number of medicines taken, and the time when the medicine was taken. The record of taking medicine is an example of information regarding taking medicine. The food record is, for example, the contents of the food eaten and the amount of salt ingested. The food record is an example of information related to food.

[0029] The smoking record is, for example, whether or not a person smoked, the date and time of smoking, and the number of cigarettes smoked. The smoking record is an example of information related to smoking. The drinking record is, for example, the date and time of drinking, the amount of alcohol consumed, and the type of alcohol. The drinking record is an example of information related to drinking. The exercise record includes, for example, records related to sports, as well as records of daily activities such as walking, shopping, cleaning, etc. The exercise record is an example of information related to exercise. The mood record is, for example, a subjective mood. The mood record is an example of information related to mood.

[0030] The record of physical condition is, for example, a subjective physical condition or symptoms. The record of physical condition is an example of information related to physical condition. The medical examination history is, for example, the start date of treatment, the date of examination, the content of treatment, and advice from doctors and other medical personnel. The medical examination history is an example of information related to medical examinations.

[0031] The operation history of the support application is, for example, a history of operations related to the start-up operation, the input operation of blood pressure values, and the input operation of reviewing the support application. Biological characteristics include, for example, the presence or absence of other diseases, the presence or absence of injuries undergoing treatment, the presence or absence of knee or leg pain, the presence or absence of experience of treatment for hypertension, the number of years since hypertension was diagnosed, the strength of seasonings used in home cooking, and the amount of food eaten. Psychological characteristics include, for example, expectations for support apps, willingness to acquire knowledge and information that will contribute to improving lifestyle habits, whether reducing salt intake is difficult or whether one believes one's sense of taste cannot be changed, and psychological resistance to leaving food on one's plate.

[0032] Social characteristics include, for example, wake-up time, bedtime, type of work (e.g., shift work, day shift, night shift), days of the week worked, start time of work, time home from work, regular days off, and whether or not there is heating in the changing room. Habits include, for example, exercise habits, the habit of weighing oneself, the habit of checking calorie counts on food labels, the habit of choosing low-fat foods, the habit of not consuming caffeine after 4 p.m., the habit of eating and drinking after 10 p.m., the habit of skipping breakfast, the habit of snacking, the habit of taking a bath one hour before going to bed, the habit of stretching or massaging before going to bed, and the habit of sleeping six hours or more. The goal achievement status is, for example, the relationship between the user's status and a goal set for each user by a doctor or the like. The above-mentioned information can be classified into subjective information and objective information.

[0033] <Message Generation System> The auxiliary storage device 104 of the message generation system 11 stores a template of a first message to be sent to the user to support the improvement of lifestyle habits, for example. The template of the first message is prepared according to the timing of sending the message, the purpose of notifying the user of the message, and the content of the message. The templates here include, for example, a template used for a question to the user, a template used for an answer to the user, and a template used for providing information or advice to the user.

[0034] <Question Generation System> The auxiliary storage device 104 of the question generation system 12 stores, for example, a template of a question sentence to be given to the question answering system 20. In the case of this embodiment, the age and gender columns of the template are left blank.

[0035] <Hardware configuration of user terminal> FIG. 3 is a diagram illustrating the hardware configuration of the user terminal 30. As shown in FIG. 3 includes a processor 301, a ROM 302 storing a BIOS and the like, a RAM 303 used as a work area for the processor 301, an auxiliary storage device 304, a communication interface 305, a display 306, and an input reception device 307. Each device is connected via a bus or other signal lines 308. The user terminal 30 may also be provided with a camera, a receiver for GPS (Global Positioning System) radio waves, or the like.

[0036] The communication interface 305 is an interface for communicating with other servers or terminals or with peripheral devices through the network N. The communication interface 305 supports Ethernet (registered trademark), Wi-Fi (registered trademark), Bluetooth (registered trademark), USB (=Universal Serial Bus), mobile communication systems, and other communication standards.

[0037] The display 306 is, for example, a liquid crystal display or an organic EL (Electro Luminescence) display. The input reception device 307 is, for example, a mouse or a keyboard. In addition, when the user terminal 30 is a notebook computer, a tablet computer, or a smartphone, a capacitive touch sensor having transparency that does not interfere with the visibility of the image displayed on the display 306 may be used as the input reception device 307. A device that combines a touch sensor and a display is called a touch panel.

[0038] <Processing sequence executed in the information processing system> <Summary of support services> FIG. 4 is a sequence diagram illustrating an outline of the support service provided through the information processing system 1. As shown in FIG. The processing sequence shown in Fig. 4 is realized through the execution of a program by the processor 101 (see Fig. 2) constituting the support system 10 and the processor 301 (see Fig. 3) constituting the user terminal 30. Note that the symbol S shown in the figure represents a step.

[0039] [Step 1] In the case of Fig. 4, the user's registration information is transmitted from the user terminal 30 to the support system 10. For example, the measured values ​​of blood pressure and pulse rate, information on medicines taken, etc. are transmitted as the registration information. In the present embodiment, the input and transmission of the registration information is realized through the support app.

[0040] [Step 2] The support system 10 transmits learning contents, questions about lifestyle habits, and the like to the user terminal 30. The learning content includes knowledge and information that contributes to improving lifestyle habits. The learning content is prepared, for example, for each disease. In the present embodiment, the disease is assumed to be hypertension. The learning content may be text-based, still image-based, moving image-based, or audio-based. The questions about lifestyle habits are questions about the lifestyle habits and preferences of each user. In this embodiment, questions are prepared for six categories, for example, diet, exercise, weight loss, sleep, stress, and alcohol.

[0041] [Step 3] The user terminal 30 transmits an answer to a question posed by a user to the support system 10. Note that the answer to the question is input by the user, for example, by selecting an option.

[0042] [Step 4] The support system 10 diagnoses the lifestyle habits of the target user based on the received answers. In the case of this embodiment, for example, the scores of the target user are calculated for the six categories described above. In the case of this embodiment, the three categories with the lowest scores are extracted as noteworthy categories that are likely to contribute to improving lifestyle habits. The extracted noteworthy categories are notified to the user as the diagnosis results.

[0043] [Step 5] The support system 10 notifies the user terminal 30 of recommended actions and learning contents related to the attention category. The recommended actions here are actions that contribute to the improvement of the attention category. However, it is not necessary to notify one recommended action for one attention category, and two or more recommended actions may be notified for one attention category. In the present embodiment, the support system 10 determines the recommended behavior to be notified to the user, but a mechanism may be adopted in which the user himself selects the behavior to be practiced.

[0044] [STEP 6] The user terminal 30 records the practice results of the recommended behavior and the measurement results such as blood pressure values ​​in the auxiliary storage device 304 (see FIG. 3). In this embodiment, the practice results of the recommended behavior and the measurement values ​​are input through a support application. The practice results and the measurement results are recorded in the auxiliary storage device 304 as user data.

[0045] [STEP 7] The user terminal 30 transmits the recorded practice results and the like to the support system 10. The practice results and the like may be transmitted each time user data is input, or may be transmitted when the user accesses the support system 10.

[0046] [STEP 8] The support system 10 periodically updates the recommended actions. In the case of this embodiment, at least a part of the actions recommended to the user is changed every week. For example, two of the three recommended actions are replaced with new recommended actions. The new recommended actions are selected from actions that have not been recommended in the past as much as possible.

[0047] [STEP 9] The assistance system 10 notifies the user terminal 30 of the updated recommended actions, learning contents, and the like. [STEP 10] The user terminal 30 records the results of practicing the recommended behavior and the measurement results such as blood pressure values ​​in the auxiliary storage device 304 (see FIG. 3). [STEP 11] The user terminal 30 transmits the recorded practice results, etc. to the support system 10.

[0048] Thereafter, the processing operations corresponding to steps 8 to 11 are executed once a week. In this embodiment, the processes from step 2 onward are repeatedly executed once a month. That is, the attention category is reviewed once a month, and the notification and implementation of the recommended actions for improving the reviewed attention category are repeated.

[0049] <Customize your message> Fig. 5 is a sequence diagram for explaining the procedure for creating a message whose expression is customized for each user. Note that the question generation system 12 shown in Fig. 5 is capable of generating a question sentence according to one or more pieces of information on age, sex, race or ethnicity, and region, and the question answering system 20 is capable of changing the expression according to one or more pieces of information on age, sex, race or ethnicity, and region.

[0050] [STEP 21] First, the message generation system 11 provides the question generation system 12 with one or more pieces of information regarding the age, sex, race or ethnicity, and region of the user to whom the message is to be sent.

[0051] [STEP 22] For example, when information on a user's age and gender is received, the question generation system 12 generates a question that instructs the user to change the expression to one appropriate for the user group identified by the received age and gender, and returns the question to the message generation system 11. FIG. 6 is a diagram showing an example of a question generated by the question generation system 12. As shown in FIG. The example shown in Figure 6 is a case where the user information is given as "Age: 70" and "Gender: Female." In the case of Figure 6, the question "You are an AI assistant that changes tone and voice in writing targeting demographics specified. Age: 70, Gender: Female" is generated.

[0052] [STEP 23] The message generation system 11 that receives the question sends the question and the first message to the question answering system 20. The first message here is, for example, learning content, a question about lifestyle habits, or information to support improving lifestyle habits. These are examples of information about lifestyle habits.

[0053] [STEP 24] The question answering system 20 generates an answer sentence by modifying the expression of the first message in response to the question sentence, and transmits the answer sentence to the message generation system 11. Fig. 7 is an example of an answer sentence output by the question answering system 20. The answer sentence shown in Fig. 7 is a text in which the wording of the first message has been changed to suit a 70-year-old woman. Note that the main changes are shown in bold.

[0054] Fig. 8 is another output example of an answer sentence output by the question answering system 20. The answer sentence shown in Fig. 8 is also a text obtained by changing the wording of the first message to be suitable for a 70-year-old woman. Note that the first message in Fig. 8 is a message when the lifestyle habit diagnosis result is not good. Fig. 9 is another output example of an answer sentence output by the question answering system 20. The answer sentence shown in Fig. 9 is also a sentence in which the wording of the first message is changed to be suitable for a 70-year-old woman. Note that the first message in Fig. 9 is a message when the situation of practicing the recommended behavior is not good.

[0055] [STEP 25] The message generation system 11, which has received the answer text from the question answering system 20, transmits a second message having the answer text as the main text or a second message to which the answer text is attached to the user terminal 30. This realizes transmission of a message whose expression is customized for each user who uses the lifestyle habit improvement support service. FIG. 10 is a diagram illustrating an example of sending a customized message. In Fig. 10, the same reference numerals are used to denote parts corresponding to those in Fig. 1. In the case of Fig. 10, attention is focused on the relationship between the lifestyle habit improvement supporting system 10, which is the sender of the message, and the user terminal 30.

[0056] 10, it is assumed that the contents of the first message transmitted to users A, B, and Z are the same. In other words, it is assumed that the original message is common to users A, B, and Z. However, the support system 10 in this embodiment sends to user A a message whose expression is customized for a 30-year-old male, sends to user B a message whose expression is customized for a 50-year-old male, and sends to user Z a message whose expression is customized for a 70-year-old female.

[0057] <Other examples of question generation> In the above description, a case has been described in which the question answering system 20 (see FIG. 5) is instructed to change the expression to one appropriate for the age and sex, but the present invention is not limited to this. For example, in step 21, information about the race or ethnicity of the user may be provided to question generation system 12. In this case, in step 22, a question instructing a change to an expression according to the race or ethnicity is generated and returned to message generation system 11. For example, the question returned may be "You are an AI assistant that changes tone and voice in writing targeting demographics specified. Race / Ethnicity Japanese." Fig. 11 is another output example of a reply sentence output by the question answering system 20. The reply sentence shown in Fig. 11 is a sentence in which the wording of the first message has been modified to suit Japanese people.

[0058] Also, for example, in step 21, information about the region registered for the user may be provided to the question generation system 12. In this case, in step 22, a question instructing a change to an expression appropriate for the region is generated and returned to the message generation system 11. For example, the question returned may be "You are an AI assistant that changes tone and voice in writing targeting demographics specified. Region with a lot of sunshine." FIG. 12 is another output example of an answer sentence output by the question answering system 20. The answer sentence shown in FIG. 12 is a text in which the expression of the first message is changed for a user living in an area with a lot of sunshine. In the case of the answer sentence shown in FIG. 12, the expression is changed to a slightly more cheerful expression. However, the expression according to the area with a lot of sunshine depends on the teacher data used for learning by the machine learning model, etc. Therefore, even if the content of the question is the same, the expression is changed to a different one if a different machine learning model is used.

[0059] <Summary> According to the information processing system 1 (see FIG. 1) of this embodiment, a message with a customized expression for each user can be sent. As a result, the message is more receptive to and sympathetic to than when a message with a uniform expression is sent regardless of the user's age or sex, and it becomes easier to obtain the effect of improving lifestyle habits.

[0060] <Other embodiments> (1) Although the embodiments of the present disclosure have been described above, the technical scope of the present disclosure is not limited to the scope described in the above-mentioned embodiments. It is clear from the claims that the technical scope of the present disclosure also includes various modifications or improvements to the above-mentioned embodiments.

[0061] (2) The processor in the above-described embodiments refers to a processor in a broad sense, and includes general-purpose processors (e.g., CPU (=Central Processing Unit)) as well as dedicated processors (e.g., GPU (=Graphical Processing Unit), ASIC (=Application Specific Integrated Circuit), FPGA (=Field Programmable Gate Array), programmable logic devices, etc.). In addition, the operations of the processor in each of the above-mentioned embodiments may be executed by multiple processors in cooperation with each other, and the order of execution of each operation in the processor is not limited to the order described in each of the above-mentioned embodiments, and may be changed individually.

[0062] (3) In the above-described embodiment, the operator of the support system 10 (see FIG. 1) is assumed to be a company or organization where the user works, but the operator may also be an entity that operates a public or private health insurance system or a medical institution that provides medical services.

[0063] (4) In the above-described embodiment, the company or organization that operates the support system 10 also operates the message generation system 11 and the question generation system 12. However, the operator of the message generation system 11 and the operator of the question generation system 12 may be different.

[0064] (5) In the above embodiment, the support system 10 is composed of the message generating system 11 and the question generating system 12. However, the message generating system 11 and the question generating system 12 may be integrated into one system.

[0065] (6) The support service in the above-described embodiment is intended to support the prevention, diagnosis, and treatment of hypertension, but can also be applied to other diseases. For example, the technology can be applied to customizing messages in services that support the improvement of lifestyle habits related to nicotine addiction, insomnia, depression, diabetes, alcoholism, and obesity.

[0066] Incidentally, in the case of a user with nicotine addiction, the following user data is stored: carbon monoxide (CO) in the breath, nicotine concentration in saliva, smoking date and time, number of cigarettes smoked or number of times, user's cognitive information (way of thinking, values), exercise (number of steps, length of exercise, distance traveled), other illnesses (diabetes), medication, blood pressure, heart rate, pulse, current physical condition (headache, irritability, feeling good, nausea, etc.), and all or part of the answers to the assumed causes. These pieces of information are examples of information related to smoking.

[0067] For a user with insomnia, the user data may include, for example, all or part of the following: time of going to bed, time of waking up, sleep duration, sleep efficiency, number and content of meals, exercise record, and medication record. Time of going to bed, time of waking up, sleep duration, and sleep efficiency are examples of information related to sleep. In addition, in the case of a user suffering from depression, the user data stored may include, for example, all or part of a record of mood, a record of behavior, a record of medication, a record of sleep, and a record of appetite.

[0068] For a diabetic user, the user data may include, for example, the number and type of meals, exercise records, medication records, weight, blood pressure, and blood glucose levels. Blood glucose levels are an example of information related to blood glucose levels. In the case of a user with alcoholism, for example, a record of medication, whether or not the user has drunk alcohol, and all or part of the details of the drinking are stored as user data.

[0069] In addition, for obese users, the user data stored may include, for example, target weight, start date of treatment, number of days of treatment, appropriate calorie intake, health history (mood), weight history, calorie intake history, date and time of calorie intake, medication history (dosage taken, date and time of medication taken), therapy history (therapy, content of advice, whether or not the user performed it), medical history, blood data (AST (aspartate aminotransferase) / ALT (alanine aminotransferase)) and all or part of the CT value.

[0070] (7) In the above embodiment, 70 years old is given as an example of the age given in step 21, and ages 30, 50, and 70 are given as examples in step 25, but these examples do not mean that the user's age is specified in increments of 10 years, but rather that actual ages such as 33 and 54 are used. However, the user's age may also be specified in increments of 5 years or 10 years.

[0071] <Summary> The disclosed examples described in the above-mentioned embodiments are shown below. (((1))) An information processing system having one or more processors, which accepts at least one piece of information relating to the age, sex, race or ethnicity, and region of a user who uses a lifestyle habit improvement support service, and generates a question for instructing a machine learning model that has at least learned expressions according to the information to change the expressions to ones appropriate for the user group identified by the accepted information. According to this information processing system, a question can be generated to obtain a message from a machine learning system whose expression is customized according to at least one piece of information related to the user's age, gender, race or ethnicity, and region.

[0072] (((2))) The information processing system described in (((1))), wherein the processor provides a first message addressed to a user and a question text to the machine learning model, and sends to the user a second message having an answer text from the machine learning model as the text body, or a second message with the answer text attached. According to this information processing system, a message can be sent whose expression is customized according to at least one piece of information related to the user's age, sex, race or ethnicity, and region.

[0073] (((3))) The information processing system described in (((1))), wherein the processor provides a first message corresponding to information on the user's lifestyle habits and a question to the machine learning model, and sends to the user a second message having an answer from the machine learning model as the main text, or a second message with the answer attached. According to this information processing system, a message can be sent whose expression is customized according to at least one piece of information related to the user's age, sex, race or ethnicity, and region.

[0074] (((4))) The information processing system according to (((3))), wherein the processor generates the first message in response to a diagnosis result of the user's lifestyle habits. According to this information processing system, a message can be sent whose expression is customized according to at least one piece of information related to the user's age, sex, race or ethnicity, and region.

[0075] (((5))) The information processing system according to (((3))), wherein the processor generates the first message in response to a behavior history of the user in response to a diagnosis result of the user's lifestyle habits. According to this information processing system, a message can be sent whose expression is customized according to at least one piece of information related to the user's age, sex, race or ethnicity, and region.

[0076] (((6))) An information processing method in which a computer executes a process of accepting at least one piece of information relating to the age, sex, race or ethnicity, and region of a user who uses a lifestyle habit improvement support service, and a process of generating a question sentence for instructing a machine learning model that has at least learned an expression according to the information to change the expression to an expression suitable for the user group identified by the accepted information. According to this information processing method, a question can be generated to obtain a message from a machine learning system whose expression is customized according to at least one piece of information related to the user's age, gender, race or ethnicity, and region. In addition, in this information processing method, it is possible to combine the techniques corresponding to (((2))) to (((5))).

[0077] (((7))) A program for enabling a computer to perform the following functions: accepting at least one piece of information relating to the age, sex, race or ethnicity, and region of a user who uses a lifestyle improvement support service; and generating a question for instructing a machine learning model that has at least learned expressions according to the information to change the expressions to be appropriate for the user group identified by the accepted information. The program can generate questions to obtain messages from a machine learning system whose wording is customized according to at least one piece of information about the user's age, gender, race or ethnicity, and region. In this program, it is also possible to combine technologies corresponding to (((2))) to (((5))). [Explanation of symbols]

[0078] 1...information processing system, 10...lifestyle improvement support system, 11...message generation system, 12...question generation system, 20...question answering system, 30...user terminal, 101, 301...processor, 102, 302...ROM, 103, 303...RAM, 104, 304...auxiliary storage device, 105, 305...communication interface, 106, 308...signal line, 306...display, 307...input reception device

Claims

1. One or more processors; The one or more processors: Accepting at least one piece of information regarding the age, sex, race or ethnicity, and region of a user who uses the lifestyle improvement support service; generating a question sentence for instructing a machine learning model that has learned at least an expression according to the received information to change the expression to one suitable for a user group identified by the received information; Information processing system.

2. The processor, providing a first message addressed to the user and the question sentence to the machine learning model; sending a second message, the main text of which is the response text from the machine learning model, or a second message to which the response text is attached, to the user; The information processing system according to claim 1 .

3. The processor, providing a first message corresponding to information on the lifestyle habits of the user and the question sentence to the machine learning model; sending a second message, the main text of which is the response text from the machine learning model, or a second message to which the response text is attached, to the user; The information processing system according to claim 1 .

4. The processor, generating the first message in response to a diagnosis result of the lifestyle habit of the user; The information processing system according to claim 3 .

5. The processor, generating the first message in response to a behavior history of the user with respect to a diagnosis result of the lifestyle habit of the user; The information processing system according to claim 3 .

6. The computer A process of receiving at least one piece of information related to an age, a sex, a race or an ethnicity, and a region of a user who uses the lifestyle habit improvement support service; A process of generating a question sentence for instructing a machine learning model that has learned at least an expression according to the received information to change the expression to an expression suitable for a user group identified by the received information; An information processing method for performing the above.

7. On the computer, A function of receiving at least one piece of information regarding the age, sex, race or ethnicity, and region of a user who uses the lifestyle habit improvement support service; A function of generating a question sentence for instructing a machine learning model that has learned at least an expression according to the received information to change the expression to an expression suitable for a user group identified by the received information; and A program to achieve this.

Images