Information processing system, information processing device, storage medium, trained model, and information processing method

Abstract

[Problem] To provide an information processing system, an information processing device, a storage medium, a trained model, and an information processing method capable of properly evaluating the degree of stress. [Solution] The system comprises: a biological information acquisition device that acquires information about a user's body from the user over time as biological information; and an information processing device that uses the biological information acquired by the biological information acquisition device as a basis for executing a stress level information generating process that generates stress level information in which the degree of stress experienced by the user is evaluated as a stress level.

Description

Information processing system, information processing device, storage medium, trained model, and information processing method

[0001] The present invention relates to an information processing system, an information processing device, a storage medium, a trained model, and an information processing method, and particularly to an information processing system, an information processing device, a storage medium, a trained model, and an information processing method for evaluating a user's stress level.

[0002] Due to various social factors, modern society is thought to be in an environment where people are more likely to experience stress and sleepiness due to stress-induced sleep deprivation.Since stress and sleepiness increase the likelihood of physical or mental disorders, it is preferable to properly conduct stress checks that can confirm the degree of stress.

[0003] Patent Document 1 discloses a technology that aims to quantitatively calculate a user's stress level, lifestyle habits, health status, etc. based on answers to questions posed to the user, and to accurately check the user's mental health based on these calculated evaluation values.

[0004] JP 2023-36409 A

[0005] However, the perception of stress is subjective and differs from person to person, so for example, even if the work environment is the same, it may be stressful for one person while not causing stress for another.

[0006] In this way, if it were possible to individually and objectively evaluate the degree of stress, which varies from person to person, it would be possible to prevent physical or mental disorders before they occur.

[0007] On the other hand, in the case of drowsiness caused by lack of sleep, if it is possible to assess whether a vehicle driver or an operator of heavy machinery or a facility is feeling drowsy, it is expected that serious accidents caused by drowsiness can be prevented.

[0008] The present invention has been made in consideration of the above circumstances, and aims to provide an information processing system, an information processing device, a storage medium, a trained model, and an information processing method that can appropriately evaluate the degree of stress and drowsiness.

[0009] In order to achieve the above-mentioned object, the information processing system of the present invention comprises a biometric information acquisition device that acquires information about the user's body as biometric information from the user over time, and an information processing device that executes a stress level information generation process that generates stress level information that evaluates the user's degree of stress as a stress level based on the biometric information acquired by the biometric information acquisition device.

[0010] This allows the biometric information acquisition device to acquire biometric information about the user's body over time, making it possible to properly evaluate the stress level of each individual user in accordance with changes in their condition over time.

[0011] The information processing device of this information processing system autonomously executes a stress level information generation process that generates stress level information that estimates the user's stress level using a trained model generated by machine learning using learning data generated by associating reference values ​​calculated from answers to questions about stress status with biometric information as training data.

[0012] The biometric information processed by this information processing system is composed of multiple different biometric data, and the information processing device autonomously executes a stress level information generation process that generates stress level information that estimates the user's stress level using a trained model generated by machine learning using learning data generated by associating reference values ​​calculated from answers to questions regarding stress status with the biometric information that combines multiple biometric data as training data.

[0013] The biological information processed by this information processing system is composed of multiple different biological data, and the biological data includes at least one of skin temperature, body temperature, heart rate, heart rate variability, respiratory rate, blood oxygen concentration, blood pressure, and sleep time.

[0014] In order to achieve the above-mentioned object, the information processing device of the present invention executes a stress level information generation process that generates stress level information that evaluates the degree of stress of a user as a stress level based on biometric information acquired by a biometric information acquisition device that acquires information about the user's body as biometric information from the user over time.

[0015] In order to achieve the above-mentioned object, the storage medium of the present invention stores a program that causes an information processing device implemented by a computer to execute a stress level information generation process that generates stress level information that evaluates the degree of stress of a user as a stress level based on biometric information acquired by a biometric information acquisition device that acquires information about the user's body as biometric information from the user over time.

[0016] To achieve the above-mentioned objective, the trained model of the present invention is generated by machine learning using training data generated by associating reference values ​​calculated from responses to questions regarding stress status with biometric information acquired by a biometric information acquisition device that acquires information about the user's body as biometric information over time from the user, and autonomously executes a stress level information generation process that generates stress level information that estimates the user's degree of stress as a stress level.

[0017] To achieve the above-mentioned object, the information processing method of the present invention uses an information processing device implemented by a computer to execute a stress level information generation process that generates stress level information that evaluates the user's degree of stress as a stress level based on biometric information acquired by a biometric information acquisition device that acquires information about the user's body as biometric information from the user over time.

[0018] In order to achieve the above-mentioned object, the information processing system of the present invention comprises a biometric information acquisition device that acquires information about the user's body as biometric information from the user over time, and an information processing device that executes a drowsiness score information generation process that generates drowsiness score information that evaluates the user's degree of drowsiness as a drowsiness score based on the biometric information acquired by the biometric information acquisition device.

[0019] According to this, the biometric information acquisition device can acquire biometric information about the user's body over time, and therefore the degree of drowsiness according to changes in condition over time can be appropriately evaluated as drowsiness score information for each individual user.

[0020] The information processing device of this information processing system autonomously executes a sleepiness score information generation process that estimates the user's level of sleepiness using a trained model generated by machine learning using learning data generated by associating reference values ​​calculated from answers to questions about sleepiness and previously acquired daily rhythm information about sleep and wakefulness with biological information as training data, and generates sleepiness score information based on the estimated level of sleepiness.

[0021] Furthermore, the trained model estimates the user's level of drowsiness after an arbitrary amount of time has passed since the time the biometric information was acquired, and generates drowsiness score information based on the estimated level of drowsiness.

[0022] The information processing device of this information processing system executes a notification generation process that generates a notification to give a warning to the user when drowsiness score information reaches a preset threshold value.

[0023] According to the present invention, it is possible to properly evaluate the stress level or the degree of sleepiness for each individual user.

[0024] FIG. 1 is a block diagram illustrating an outline of the configuration of an information processing system according to a first embodiment of the present invention. FIG. 1 is a block diagram illustrating an outline of the configuration of a wearable device of the information processing system according to the present embodiment. FIG. 1 is a block diagram illustrating an outline of the configuration of a user terminal of the information processing system according to the present embodiment. FIG. 1 is a block diagram illustrating an outline of the configuration of an information processing device of the information processing system according to the present embodiment. FIG. 1 is a block diagram illustrating an outline of the functions of the information processing device of the information processing system according to the present embodiment. FIG. 1 is a block diagram illustrating an example of training data processed by the information processing system according to the present embodiment. FIG. 1 is a diagram illustrating an example of biological information processed by the information processing system according to the present embodiment. FIG. 1 is a block diagram illustrating an outline of the processing of the information processing system according to the present embodiment. FIG. 2 is a flowchart illustrating an outline of the processing of the information processing system according to the present embodiment. FIG. 3 is a diagram illustrating an outline of the functions of the information processing device of the information processing system according to the second embodiment of the present invention. FIG. 4 is a block diagram illustrating an example of training data processed by the information processing system according to the present embodiment. FIG. 5 is a diagram illustrating an example of daily rhythm information processed by the information processing system according to the present embodiment. FIG. 6 is a block diagram illustrating an outline of the processing of the information processing system according to the present embodiment. 1A and 1B are block diagrams and a flowchart illustrating an outline of processing by the information processing system according to the present embodiment.

[0025] Next, an information processing system according to an embodiment of the present invention will be described with reference to FIGS.

[0026] First Embodiment First, an information processing system according to a first embodiment of the present invention will be described with reference to FIGS.

[0027] 1 is a block diagram illustrating an outline of the configuration of an information processing system according to the present embodiment. As shown in the figure, the information processing system 10 includes a wearable device 20 that is a biometric information acquisition device, a user terminal 30, and an information processing device 40.

[0028] In this embodiment, the wearable device 20 and the user terminal 30 are connected to each other so that they can access each other, for example, via a short-range communication network, and the user terminal 30 and the information processing device 40 are connected to each other so that they can access each other via a network N such as the Internet network.

[0029] The wearable device 20 is worn on the body of a user 1, the user terminal 30 is owned by the user 1, and the information processing device 40 is deployed at a business operator 2 that provides a service using the information processing system 10.

[0030] 2 is a block diagram illustrating an outline of the configuration of wearable device 20. Wearable device 20 is a wristwatch-type so-called smartwatch that is worn on any part of the body of user 1, in this embodiment on the wrist of user 1, and as shown in the figure, includes a control unit 21, a display 22, a biosensor 23, and an acceleration / direction sensor 24 as its main components.

[0031] In this embodiment, the control unit 21 controls each part of the wearable device 20, such as the display 22, the biosensor 23, the acceleration / orientation sensor 24, etc., and is composed of, for example, a processor, memory, storage, a transceiver unit, etc.

[0032] The control unit 21 stores various applications that run on the wearable device 20, and in this embodiment, stores an application that manages biometric information D1, which will be described later.

[0033] In this embodiment, the display 22 displays screen interfaces of various applications executed by the wearable device 20, such as an application that manages the biometric information D1.

[0034] The display 22 is a so-called touch panel that accepts information input by touching the display surface, and is implemented using various technologies such as a resistive film system or a capacitive system.

[0035] The biosensor 23 is a sensor that includes an LED that emits LED light toward the wrist of the user 1 and a light receiving unit implemented with a phototransistor that receives the LED light, and the light receiving unit receives the reflected LED light that the LED emits toward the wrist.

[0036] In this embodiment, the acceleration and direction sensor 24 is a sensor that detects the acceleration of the wearable device 20 when the wearable device 20 moves in accordance with the movement of the user 1, the direction in which the wearable device 20 is facing, and the like.

[0037] In this embodiment, these biometric sensors 23 and acceleration / orientation sensor 24 are used to acquire biometric information D1 consisting of multiple biometric data related to the body of user 1, such as skin temperature, body temperature, heart rate, heart rate variability, respiratory rate, blood oxygen concentration, blood pressure, and sleep time.

[0038] In this embodiment, wearable device 20 is constantly worn by user 1, and biometric information D1 is obtained that captures the physical condition of user 1 over time, such as changes in skin temperature over a 20-hour period.

[0039] 3 is a block diagram illustrating an outline of the configuration of the user terminal 30. In this embodiment, the user terminal 30 is a portable information terminal such as a smartphone or a tablet computer, and as shown in the figure, includes a control unit 31 and a display 32 as its main components.

[0040] In this embodiment, the control unit 31 controls each part of the user terminal 30 including the display 32, and is composed of, for example, a processor, a memory, a storage, a transmitting / receiving unit, and the like.

[0041] The control unit 31 stores various applications that are activated on the user terminal 30, and in this embodiment, for example, stores an application that processes information provided by the information processing device 40, etc.

[0042] In this embodiment, the display 32 displays a screen interface of an application executed by the user terminal 30, such as an application that processes information provided by the information processing device 40.

[0043] The display 32 is a so-called touch panel that accepts information input by touching the display surface, and is implemented using various technologies such as a resistive film system or a capacitive system.

[0044] In this embodiment, the information processing device 40 is implemented by, for example, a desktop or notebook computer.

[0045] 4 is a block diagram illustrating an outline of the configuration of the information processing device 40. As shown in the figure, the information processing device 40 mainly comprises a processor 41, a memory 42 which is a storage medium, a storage 43, a transmission / reception unit 44, and an input / output unit 45, which are electrically connected to each other via a bus 46.

[0046] The processor 41 is a computing device that controls the operation of the information processing device 40, controls the transmission and reception of data between each element, and performs processing necessary for executing application programs.

[0047] In this embodiment, the processor 41 is, for example, a CPU (Central Processing Unit), and executes application programs loaded in a memory 42 (to be described later) to perform various processes.

[0048] The memory 42 is implemented by a main storage device that is configured as a volatile storage device such as a DRAM (Dynamic Random Access Memory).

[0049] The memory 42 is used as a work area for the processor 41, and also stores the BIOS (Basic Input / Output System) that is executed when the computer is started, various setting information, and the like.

[0050] The storage 43 stores data and the like used for various processes by application programs and the like.

[0051] The transmitting / receiving unit 44 connects the information processing device 40 to a network, and in this embodiment, the information processing device 40 is connected to the user terminal 30 via the network via the transmitting / receiving unit 44 so that they can communicate with each other.

[0052] The transceiver 44 may be compatible with a wireless communication standard such as Wi-Fi, or may be equipped with a short-range communication interface such as Bluetooth (registered trademark) or BLE (Bluetooth Low Energy).

[0053] If necessary, information input devices such as a keyboard and a mouse and output devices such as a display are connected to the input / output unit 45. In this embodiment, a keyboard, a mouse, and a display are connected.

[0054] The bus 46 transmits, for example, address signals, data signals, and various control signals between the processor 41, memory 42, storage 43, transmission / reception unit 44, and input / output unit 45 that are connected to it.

[0055] 5 is a block diagram illustrating an outline of the functions of the information processing device 40. As shown in the figure, the information processing device 40 includes a training data generation unit 40a, a training data storage unit 40b, a training unit 40c, a trained model storage unit 40d, and a generation unit 40e.

[0056] The learning data generation unit 40a, the learning unit 40c, and the generation unit 40e are realized by executing a program stored in the memory 42 by the processor 41, and the learning data storage unit 40b and the trained model storage unit 40d are realized by partitioning the storage area of ​​the storage 43.

[0057] In this embodiment, the learning data generating unit 40a executes a process of generating learning data by associating reference values ​​as training data with physical characteristics of any subject indicated by the biometric information D1.

[0058] Here, in this embodiment, the subjects are assumed to be a cohort of subjects to be observed that share common factors, and this cohort may include user 1.

[0059] In this embodiment, the learning data storage unit 40b stores the learning data generated by the learning data generation unit 40a.

[0060] 6 is a block diagram illustrating an example of learning data. As shown in the figure, the learning data is data in which, for example, the characteristics indicated by waveforms F1, F2, F3, F4, etc. of skin temperature fluctuations over 20 hours, which are biological data constituting biological information D1, are associated with BDI scores, which are reference values, as training data, and each data is assigned a data ID.

[0061] Here, as shown in an example in Figure 7, waveform F1, which shows the variation in skin temperature over a 20-hour period, shows that although the skin temperature drops rapidly within 10 hours (approximately 7.5 hours) from the start of measurement, the drop does not last long and the skin temperature gradually rises, returning to the same level as when measurement began 20 hours later.

[0062] On the other hand, waveform F3, which shows the variation in skin temperature over a 20-hour period, shows that the skin temperature drops rapidly within 5 hours (approximately 3 hours) of the start of measurement, but returns to the same skin temperature as at the start of measurement 5 hours after the start of measurement.However, it then shows repeated rises and falls between 36.0°C and 36.8°C until 20 hours after the start of measurement.

[0063] In this embodiment, the BDI score shown in Figure 6 is a value called the "Beck Depression Scale" calculated from responses to questions about stress states designed for the purpose of self-evaluating the degree of depression, and the value is judged to be 1 to 10 (normal), 11 to 16 (mild mood swings), 17 to 20 (borderline between depression and normal), 21 to 30 (moderate depression), 31 to 40 (severe depression), and 41 or more (extreme depression).

[0064] For example, the BDI score of the subject in waveform F1 is 2.5 and therefore judged to be normal, and the BDI score of the subject in waveform F3 is 23.0 and therefore judged to be moderately depressed.

[0065] In this embodiment, the biometric information D1 used to generate the learning data is prepared in advance as a combination of multiple biometric data, such as skin temperature and body temperature, body temperature and heart rate, blood oxygen concentration and respiratory rate, etc.

[0066] For example, multiple combinations of biometric data are prepared that make it easy to grasp the subject's physical condition over time, and reference values ​​are associated with each piece of biometric data as training data, so highly accurate learning data can be generated.

[0067] 5 executes a process of generating a trained model by performing machine learning using the training data stored in the training data storage unit 40 b. As a machine learning method, various algorithms such as a neural network, a random forest, and an SVM (Support Vector Machine) are appropriately used.

[0068] In this embodiment, the trained model storage unit 40d stores the trained model generated by the training unit 40c.

[0069] 8 is a block diagram illustrating an outline of the processing of the generator 40e. As shown in the figure, the generator 40e executes a process of generating stress level information D2 when biological information D1 acquired by the wearable device 20 worn by the user 1 is input to the trained model M stored in the trained model storage unit 40d.

[0070] In this embodiment, this stress level information D2 is information that numerically evaluates the degree of stress of user 1 as a stress level, and evaluates the stress level by dividing it into multiple stages, such as a 0 to 10 scale or a 0 to 100 scale.

[0071] In this embodiment, the stress level information generation process is executed by the functions and processes of the learning data generation unit 40a, learning data storage unit 40b, learning unit 40c, learned model storage unit 40d, and generation unit 40e.

[0072] Next, an outline of the process of generating a trained model in the information processing system 10 will be described.

[0073] 9 is a flowchart outlining the process of generating a trained model. As shown in the figure, in step S1, training data is generated by associating reference values ​​as training data with physical characteristics indicated by biometric information D1 acquired from a given subject.

[0074] In this embodiment, the training data is prepared from a plurality of combinations of biological data that make it easy to grasp the subject's physical condition over time, and therefore highly accurate training data is generated.

[0075] The generated training data is stored in the training data storage unit 40b, and in the subsequent step S2, a trained model is generated by performing machine learning using the training data stored in the training data storage unit 40b, and in step S3, the generated trained model is stored in the trained model storage unit 40d.

[0076] Next, an outline of the processing of the information processing system 10 will be described.

[0077] 10 is a flowchart outlining the processing of the information processing system 10. As shown in the figure, in step S10, the wearable device 20 acquires biometric information D1 of the user 1.

[0078] In this embodiment, the wearable device 20 is always worn by the user 1, and therefore biometric information D1 that captures the physical condition of the user 1 over time for a certain period of time is acquired.

[0079] When the wearable device 20 acquires the biometric information D1 of the user 1, in the subsequent step S11, the biometric information D1 is transmitted to the information processing device 40 via the user terminal 30.

[0080] When the information processing device 40 receives the biometric information D1 transmitted via the user terminal 30, it inputs the received biometric information D1 into the trained model stored in the trained model storage unit 40d.

[0081] When the biometric information D1 is input into the trained model, the trained model estimates the stress level of user 1 in step S12, and generates stress level information D2 based on the estimated stress level in step S13.

[0082] In this embodiment, by constantly wearing the wearable device 20, it is possible to obtain biometric information D1 that indicates changes in condition over time, and therefore it is possible to estimate stress levels according to the circadian rhythm, which has a period of approximately 24 hours.

[0083] For example, if the biological information D1 is skin temperature, it generally has a cycle in which it tends to be lowest in the morning and highest from the afternoon to the evening. Therefore, if there is a significant deviation from this cycle, it can be estimated by estimating the stress level that there is a disturbance in the autonomic nervous system.

[0084] On the other hand, by performing estimations based on changes in condition over time, it is expected that more accurate estimations will be made compared to judgments that may change depending on mood swings at the time, such as when receiving a doctor's diagnosis or answering questions about stress levels.

[0085] Therefore, the stress level, which is the degree of stress, can be properly evaluated for each individual user 1 based on the biological information D1 acquired from the user 1.

[0086] Second Embodiment Next, an information processing system according to a second embodiment of the present invention will be described with reference to FIGS.

[0087] 11 to 16, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof will be omitted.

[0088] The information processing system of this embodiment evaluates the level of sleepiness of the user 1 as a measure of stress.

[0089] 11 is a block diagram illustrating an outline of the functions of the information processing device 40. As shown in the figure, the information processing device 40 includes a training data generation unit 40a1, a training data storage unit 40b1, a training unit 40c, a trained model storage unit 40d, a generation unit 40e1, and a notification unit 40f.

[0090] The learning data generation unit 40a1, learning unit 40c, generation unit 40e1, and notification unit 40f are realized by executing programs stored in memory 42 by processor 41, and the learning data storage unit 40b1 and trained model storage unit 40d are realized by partitioning the storage area of ​​storage 43.

[0091] In this embodiment, the learning data generation unit 40a1 performs a process of generating learning data by associating the daily rhythms and reference values ​​for sleep and wakefulness described below as teacher data with the physical characteristics of any subject indicated by the biometric information D1.

[0092] In this embodiment, the learning data storage unit 40b1 stores the learning data generated by the learning data generation unit 40a1.

[0093] 12 is a block diagram illustrating an example of the learning data stored in the learning data storage unit 40b1. As shown in the figure, the learning data is data in which, for example, the daily rhythm related to sleep and wakefulness and the sleepiness scale as reference values ​​are associated as teacher data with the characteristics shown by waveforms F1, F2, F3, F4, etc., of skin temperature, which is biological data constituting the biological information D1, over a 20-hour period, and each data is assigned a data ID.

[0094] As shown in Figure 13, the daily rhythm C1 of sleep and wakefulness over 24 hours is characterized in that sleepiness increases as the wakefulness time increases, while the level of wakefulness is highest a few hours before bedtime (around 9 pm).

[0095] In this embodiment, the circadian rhythms C1, C2, C3, C4, etc. related to sleep and wakefulness are information obtained based on an existing detection algorithm for any subject, and the subjects are assumed to be a cohort of subjects to be observed that share common factors.

[0096] In this embodiment, the sleepiness scale shown in FIG. 12 is an assessment of sleepiness obtained by the Epworth Sleepiness Scale, in which participants answer questions about the degree of sleepiness they have experienced in recent daily life in a variety of situations (e.g., "while sitting and reading"), and the points assigned to the answers are added up to show the degree of sleepiness on a scale.

[0097] In the Epworth Sleepiness Scale, when it comes to sleepiness in situations such as "sitting and reading" or "watching television," a score of 3 is assigned to an answer of "often feeling sleepy," a score of 2 to an answer of "sometimes feeling sleepy," a score of 1 to an answer of "rarely feeling sleepy," and a score of 0 to an answer of "never feeling sleepy."

[0098] If the total score assigned to the responses is less than 5, the subject is assessed as having little daytime sleepiness; if the total score is 5 or more but less than 10, the subject is assessed as having mild daytime sleepiness; and if the total score is 11 or more, the subject is assessed as having severe daytime sleepiness.

[0099] On the other hand, the evaluation of sleepiness using a sleepiness scale may be, for example, a "5-point sleepiness scale (Likert equal interval scale)" in which the degree of sleepiness is reported over time and the change in sleepiness over time is evaluated.

[0100] In this five-point sleepiness scale (Likert equal interval scale), for example, a response of "not sleepy" is rated as 1, a response of "slightly sleepy" is rated as 2, a response of "sleepy" is rated as 3, a response of "fairly sleepy" is rated as 4, and a response of "very sleepy" is rated as 5.

[0101] In this embodiment, the generation unit 40e1 shown in Figure 11 executes a process of estimating the degree of drowsiness and generating drowsiness score information when biometric information D1 acquired by the wearable device 20 worn by the user 1 is input into the learned model M stored in the learned model memory unit 40d.

[0102] In this embodiment, this drowsiness score information is information that numerically evaluates the drowsiness scale, which indicates the degree of drowsiness of user 1, as a drowsiness score, and the drowsiness score is evaluated by dividing it into multiple levels, such as a 0 to 10 level or a 0 to 100 level.

[0103] In this embodiment, the generation unit 40e1 generates drowsiness score information for user 1 at the time when the biometric information D1 was acquired, and also estimates the degree of drowsiness of user 1 after an arbitrary time has passed since the time when the biometric information D1 was acquired (for example, 40 minutes after the biometric information D1 was acquired) to generate drowsiness score information (prediction and estimation of the degree of drowsiness).

[0104] In this embodiment, this prediction and estimation of the degree of drowsiness is realized by using a time delay embedding method to determine the correlation between biometric information D1 at an arbitrary time and drowsiness after an arbitrary amount of time has passed from that time, and then using machine learning to perform a trained model M based on training data that analyzes this correlation.

[0105] 14 and 15 are diagrams outlining the process of predicting and estimating the level of drowsiness. As shown in Fig. 14, the trained model M estimates the level of drowsiness at time T, an arbitrary time period after time T-x, based on biological information D1 at that time.

[0106] 15 , for example, based on biological information D1 at 10:30, drowsiness at 11:10, 40 minutes later, is estimated. In this embodiment, the estimated drowsiness level has an arbitrary amplitude, so the drowsiness level can be estimated to be higher.

[0107] In this embodiment, the drowsiness score information generation process is executed by the functions and processes of the learning data generation unit 40a1, learning data storage unit 40b1, learning unit 40c, learned model storage unit 40d, and generation unit 40e1.

[0108] In this embodiment, the notification unit f shown in Figure 11 executes a notification generation process that generates a notification to warn user 1 when drowsiness score information reaches a preset threshold value.

[0109] The threshold for generating a notification is set to level 6, for example, if the drowsiness score information is divided into 10 levels, and when the drowsiness score information reaches level 6, the notification unit f generates a notification and sends it to the user terminal 30 or the wearable device 20.

[0110] The notification may be, for example, a sound such as an alarm, a vibration of the user terminal 30 or the wearable device 20, a text message, lighting up an LED on the user terminal 30 or the like, or a combination of these, but is not limited to these, and any method that can alert the user 1 through a stimulus may be used as appropriate.

[0111] Next, an outline of the processing of the information processing system of this embodiment will be described.

[0112] 16 is a flowchart outlining the processing of the information processing system. As shown in the figure, in step S20, the wearable device 20 acquires biometric information D1 of the user 1.

[0113] In this embodiment, the wearable device 20 is always worn by the user 1, and therefore biometric information D1 that captures the physical condition of the user 1 over time for a certain period of time is acquired.

[0114] When the wearable device 20 acquires the biometric information D1 of the user 1, in the subsequent step S21, the biometric information D1 is transmitted to the information processing device 40 via the user terminal 30.

[0115] When the information processing device 40 receives the biometric information D1 transmitted via the user terminal 30, it inputs the received biometric information D1 into the trained model stored in the trained model storage unit 40d.

[0116] When the biometric information D1 is input into the trained model, the trained model estimates the level of drowsiness of the user 1 in step S22 and generates drowsiness score information based on the estimated level of drowsiness.

[0117] In the subsequent step S23, when the generated drowsiness score information reaches a preset threshold, in step S24, a notification is generated to warn the user 1, and the generated notification is sent to the user terminal 30 or the wearable device 20.

[0118] In this way, the wearable device 20 can acquire biometric information D1 about the user 1's body over time, and therefore the degree of drowsiness according to changes in condition over time can be appropriately evaluated as drowsiness score information for each individual user 1.

[0119] In particular, in this embodiment, when the drowsiness score information reaches a preset threshold, a notification is generated to warn the user 1 and transmitted to the user terminal 30 or the wearable device 20 .

[0120] Therefore, for example, if a vehicle driver or an operator of heavy machinery or a facility feels drowsy, the driver can be woken up by receiving a notification, thereby preventing the driver from going from drowsy to drowsy and causing a serious accident.

[0121] The present invention is not limited to the above-described embodiments, and various modifications are possible without departing from the spirit of the invention.

[0122] In the first embodiment, the stress level information generation process is performed using the trained model M. However, for example, the stress level information may be generated by comparing biometric information D1 acquired in advance and stored in the information processing device 40 with biometric information D1 acquired by the wearable device 20 to evaluate the stress level of the user 1. The same applies to the case where drowsiness score information is generated in the second embodiment.

[0123] In each of the above embodiments, the biometric information D1 has been mainly described as skin temperature, but the biometric information D1 may also be body temperature, heart rate, heart rate variability, respiratory rate, blood oxygen concentration, blood pressure, sleep time, etc.

[0124] In the above embodiments, the case where the biometric information D1 is transmitted to the information processing device 40 via the user terminal 30 has been described, but the biometric information D1 may be transmitted directly from the wearable device 20.

[0125] In the above embodiments, the information processing device 40 is installed in the business operator 2, but the information processing device 40 may be implemented in a cloud environment.

[0126] REFERENCE SIGNS LIST 1 User 10 Information processing system 20 Wearable device (biometric information acquisition device) 30 User terminal 40 Information processing device 42 Memory (storage medium)

Claims

1. An information processing system comprising: a biometric information acquisition device that acquires information about a user's body over time as biometric information from the user; and an information processing device that executes a stress level information generation process that generates stress level information that evaluates the user's degree of stress as a stress level based on the biometric information acquired by the biometric information acquisition device.

2. The information processing system of claim 1, wherein the information processing device autonomously executes the stress level information generation process to generate the stress level information that estimates the stress level of the user using a trained model generated by machine learning using learning data generated by associating reference values ​​calculated from answers to questions regarding stress state with the biometric information as teacher data.

3. The information processing system of claim 1 or 2, wherein the biometric information is composed of multiple different biometric data, and the information processing device autonomously executes the stress level information generation process to generate the stress level information that estimates the stress level of the user using a trained model generated by machine learning with learning data generated by associating reference values ​​calculated from answers to questions regarding stress states as teacher data with the biometric information combined with the multiple biometric data.

4. An information processing system as described in claim 1 or 2, wherein the biometric information is composed of multiple different biometric data, and the biometric data includes at least one of skin temperature, body temperature, heart rate, heart rate variability, respiratory rate, blood oxygen concentration, blood pressure, and sleep time.

5. An information processing device that executes a stress level information generation process that generates stress level information that evaluates the degree of stress of the user as a stress level based on biometric information acquired by a biometric information acquisition device that acquires information about the user's body over time as biometric information from the user.

6. A storage medium having stored therein a program that causes an information processing device implemented by a computer to execute a stress level information generation process that generates stress level information that evaluates the degree of stress of the user as a stress level based on biometric information acquired by a biometric information acquisition device that acquires information about the user's body as biometric information from the user over time.

7. A trained model which is generated by performing machine learning using learning data generated by associating reference values ​​calculated from answers to questions regarding a stress state with biometric information acquired by a biometric information acquisition device which acquires information regarding a user's body over time as biometric information from the user, as teacher data, and which autonomously executes a stress level information generation process which generates stress level information which estimates the degree of stress of the user as a stress level.

8. An information processing method that uses an information processing device implemented by a computer to execute a stress level information generation process that generates stress level information that evaluates the degree of stress of the user as a stress level based on biometric information acquired by a biometric information acquisition device that acquires information about the user's body as biometric information from the user over time.

9. An information processing system comprising: a biometric information acquisition device that acquires information about a user's body over time as biometric information from the user; and an information processing device that executes a drowsiness score information generation process that generates drowsiness score information that evaluates the user's level of drowsiness as a drowsiness score based on the biometric information acquired by the biometric information acquisition device.

10. The information processing system described in claim 10, wherein the information processing device autonomously executes the drowsiness score information generation process to estimate the level of drowsiness of the user using a trained model generated by machine learning using learning data generated by associating a reference value calculated from answers to questions regarding drowsiness and previously acquired daily rhythm information regarding sleep and wakefulness with the biological information as teacher data, and to generate the drowsiness score information based on the estimated level of drowsiness.

11. The information processing system described in claim 10 or 11, wherein the information processing device autonomously executes the drowsiness score information generation process, which estimates the level of drowsiness of the user using a trained model generated by machine learning using learning data generated by associating a reference value calculated from answers to questions regarding drowsiness and the previously acquired daily rhythm information regarding sleep and wakefulness with the biometric information as teacher data, and generates the drowsiness score information based on the estimated level of drowsiness, and the trained model estimates the level of drowsiness of the user after an arbitrary time has elapsed since the time the biometric information was acquired, and generates the drowsiness score information based on the estimated level of drowsiness.

12. The information processing system according to claim 10 or 11, wherein the information processing device executes a notification generation process that generates a notification to give a warning to the user when the drowsiness score information reaches a preset threshold value.

Images