Information processing device, method for predicting wet bulb globe temperature, and program

Abstract

The purpose of the present invention is to predict the wet bulb globe temperature on an individualized basis. An information processing device according to the present invention operates a model generated using battery temperature data from a portable terminal as input data for generating a model and using the wet bulb globe temperature at the time the input data was acquired as ground truth data, and thereby acquires the wet bulb globe temperature predicted by the model on the basis of the battery temperature data, which is acquired by the portable terminal as data for acquiring the wet bulb globe temperature.

Description

Information Processing Apparatus, Heat Index Prediction Method, and Program

[0001] The present disclosure relates to an information processing apparatus, a heat index prediction method, and a program.

[0002] Generally, the causes of heat stroke can be broadly classified into environmental factors and human factors. Regarding environmental factors, as a risk index for prevention purposes, a heat index (WBGT: Wet Bulb Globe Temperature) is defined. To calculate the heat index, it is necessary to measure the wet bulb temperature, black globe temperature, and dry bulb temperature, and usually, dedicated measuring instruments are used.

[0003] Here, to obtain the heat index for each individual, a method of accessing the heat stroke prevention information site of the Ministry of the Environment or obtaining the measurement results necessary for calculating the heat index from weather information and calculating based on the above definition can be considered.

[0004] However, the heat index obtained by the above methods is only the heat index of a predetermined regional unit. On the other hand, the prevention of heat stroke is carried out by each individual, and it is desirable that the heat index be calculated on an individual basis (based on the measurement results at the current position of each individual).

[0005] Brady Tripp, Heather K Vincent, Michelle Buner, Michael Seth Smith, “Comparison of wet bulb globe temperature measured on-site vs estimated and the impact on activity modification in high school football,” Int J Biometeorol, 2020 Apr;64(4),593-600, 2019 Dec21. Tetsushi Murata, Shigeki Hosokawa, Yuan Xue, Satoshi Kawachi, Kaori Fujinami, “Environmental sensor module and base software for individual-participatory sensing considering the location of the accompanying sensor,” Proceedings of the Embedded Systems Symposium 2012, 2012 73-78, October 10, 2012. Azusa Onogi, Yuto Ogasawara, Yamato Matsuda, Je-Yeon Kim, Shota Matsuhashi, Tetsuya Manabe, “Study on a heatstroke prevention system based on the user's activity space,” SeMI2021-83, February 28, 2022.

[0006] In contrast, obtaining an individual heat index would require each person to carry a dedicated measuring device or an alternative measuring device. However, it is not always practical for each individual to constantly carry such a measuring device to obtain the heat index, due to the economic burden, the risk of loss and malfunction, and the inconvenience of having to carry more items.

[0007] This disclosure aims to provide a technology for predicting the heat index on an individual basis.

[0008] One aspect of the present disclosure is an information processing device that takes battery temperature data of a mobile terminal as input data for model generation, and operates a model generated using the heat index at the time the input data is acquired as ground truth data, thereby obtaining the heat index predicted by the model based on the battery temperature data acquired by the mobile terminal for obtaining the heat index.

[0009] According to this disclosure, it will be possible to predict the heat index on an individual basis.

[0010] Figure 1 shows the factors contributing to heatstroke. Figure 2 shows the method for calculating the heat index, a risk indicator for environmental factors. Figure 3 shows the heat index guideline labels. Figure 4 shows an example of the use of a heat index prediction device. Figure 5 shows an example of the hardware configuration of a mobile terminal. Figure 6 shows an example of the configuration of the learning data generation system and the functional configuration of each device. Figure 7 shows an example of learning data. Figure 8 shows the relationship between the data used to predict the heat index in a black globe heatstroke index meter and the battery temperature data of a mobile terminal. Figure 9 is a flowchart showing the flow of the learning data generation process by the learning data generation system. Figure 10 shows an example of the functional configuration of a learning device. Figure 11 is a flowchart showing the flow of the learning process by the learning device. Figure 12 shows an example of the functional configuration of a heat index prediction device. Figure 13 is a flowchart showing the flow of the prediction process by the heat index prediction device. Figure 14 shows an example of the screen of a heat index prediction device. Figure 15 shows an example of the prediction accuracy of the heat index predicted by the heat index prediction device. Figure 16 shows an example of the configuration of the heat index prediction system and the functional configuration of each device. Figure 17 shows an example of the hardware configuration of the server device. Figure 18 is a flowchart showing the processing flow by each device of the heat index prediction system.

[0011] Each embodiment will be described below with reference to the attached drawings. In this specification and the drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant descriptions will be omitted.

[0012] [First Embodiment] <Regarding the Factors Contributing to Heatstroke> First, the factors contributing to heatstroke will be explained. Generally, the factors contributing to heatstroke can be broadly divided into environmental factors and human factors. Figure 1 is a diagram illustrating the factors contributing to heatstroke.

[0013] As shown in Figure 1, environmental factors 110 include factors such as "high temperature," "high humidity," "weak wind," and "strong sunlight."

[0014] Furthermore, human factors include "physical factors" and "behavioral factors." "Physical factors" further include factors such as "elderly, infants, obesity," "people with disabilities," "pre-existing medical conditions," and "malnutrition." "Behavioral factors" include factors such as "strenuous exercise," "unfamiliar exercise," "prolonged outdoor work," and "difficulty in staying hydrated."

[0015] In this application, we will describe a prediction technique for the heat index, which is a risk indicator defined for the purpose of preventing heatstroke, as one of the environmental factors 110.

[0016] <Method for Calculating the Heat Index> The heat index, a risk indicator for environmental factors, has traditionally been calculated using the method shown in Figure 2. Figure 2 is a diagram showing the method for calculating the heat index, a risk indicator for environmental factors. As indicated by the symbol 200 in Figure 2, the formula for calculating the heat index used at the Japan Meteorological Agency's measurement sites is: Heat Index = 0.7 × Tw + 0.2 × Tg + 0.1 × Ta. However, in this formula, Tw is the wet-bulb temperature, which is the temperature affected by humidity and wind. Tg is the black-bulb temperature, which is the temperature affected by solar radiation. Ta is the dry-bulb temperature, which is the temperature affected by air temperature and wind.

[0017] <Heat Index Guideline Labels> Next, we will explain the heatstroke prevention measures for each label corresponding to the heat index calculated using the conventional calculation method. Figure 3 shows the heat index guideline labels.

[0018] As shown by reference numeral 300 in Figure 3, the heat index is labeled in four stages. Specifically, a heat index of 31 or higher is labeled "Dangerous," a heat index of 28 or higher but less than 31 is labeled "Severe Caution," a heat index of 25 or higher but less than 28 is labeled "Caution," and a heat index of less than 25 is labeled "Warning."

[0019] Furthermore, as shown by reference numeral 300 in Figure 3, each label is associated with information corresponding to that label, such as "guidelines for daily activities to be careful about" and "precautions." When an individual obtains a heat index, they can prevent heatstroke by acting based on the "guidelines for daily activities to be careful about" and "precautions" corresponding to the heat index label they obtained. However, conventionally, the heat index that each individual could obtain was only for a predetermined region, and in order to obtain the heat index for each individual's current location, each individual would need to constantly carry a dedicated measuring device or an alternative measuring device, which is not always practical.

[0020] In contrast, the heat index prediction device described below predicts the heat index on an individual basis (the heat index at each individual's current location) using a portable device that each individual can realistically carry with them.

[0021] <Examples of using the heat index prediction device> First, we will explain examples of using the heat index prediction device, which enables the prediction of the heat index on an individual basis. This heat index prediction device is realized by installing a heat index prediction program on a mobile terminal that each individual carries with them on a daily basis. The heat index prediction program predicts the heat index using data that is useful for predicting the heat index from the data that the mobile terminal can acquire.

[0022] As a result, with this heat index prediction device, each individual can obtain a predicted heat index on an individual basis without having to carry a dedicated measuring device or an alternative measuring device. The term "portable device" here includes any device that an individual can carry or wear, such as a smartphone, tablet, small laptop, or wearable device.

[0023] Figure 4 shows an example of the use of the heat index prediction device. As shown in Figure 4, users 410 and 420 are in the same area and are each carrying a heat index prediction device 400_1 and 400_2. However, user 410 is resting on the soil in the shade of a tree, while user 420 is walking on asphalt in the scorching sun.

[0024] If user 410 and user 420 obtain the heat index using the conventional method, they will obtain the same heat index because they are in the same area.

[0025] In contrast, the heat index prediction devices 400_1 and 400_2 can predict the heat index on an individual basis. For example, the heat index prediction device 400_1 can predict the heat index by taking into account the environment in which user 410 is currently located (an environment where the temperature and humidity are lower than when standing on asphalt in direct sunlight, as they are in the shade of a tree and on the ground). Similarly, the heat index prediction device 400_2 can predict the heat index by taking into account the environment in which user 420 is currently located (an environment where the temperature and humidity are higher than those indicated in the weather information, as they are standing on asphalt in direct sunlight).

[0026] The example in Figure 4 shows that user 410 obtained a heat index of "23.1°C" and user 420 obtained a heat index of "31.0°C".

[0027] In the example shown in Figure 4, we described a case where users 410 and 420 are outdoors and using the heat index prediction devices 400_1 and 400_2. However, users 410 and 420 may also use the heat index prediction devices 400_1 and 400_2 indoors. For example, even within the same building, the predicted heat index for each individual will differ significantly depending on whether the windows are closed and the air conditioner is not in use, or whether the air conditioner is in use.

[0028] <Explanation of Each Phase> Next, we will explain in detail each phase from the installation of the heat index prediction program on the mobile terminal to the realization of the heat index prediction devices 400_1 and 400_2. As mentioned above, the heat index prediction program predicts the heat index using data useful for predicting the heat index from the data that the mobile terminal can acquire. At this time, the heat index prediction program predicts the heat index by running a trained heat index prediction model, which is a machine learning model. Therefore, in order for the realization of the heat index prediction devices 400_1 and 400_2, the following will be executed: a learning phase including the generation of training data for training the heat index prediction model and the training of the heat index prediction model using the generated training data, and a prediction phase in which data useful for predicting the heat index is acquired and input into the trained heat index prediction model to predict the heat index. Below, we will explain in detail each phase.

[0029] <Details of the Learning Phase> First, we will explain the details of the learning phase.

[0030] (1) Hardware configuration of the mobile terminal used in the learning phase First, the hardware configuration of the mobile terminal used in the learning phase will be described. Figure 5 is a diagram showing an example of the hardware configuration of the mobile terminal. As shown in Figure 5, the mobile terminal 500 used in the learning phase has a processor 501, memory 502, auxiliary storage device 503, display device 504, operating device 505, and GPS device 506. The mobile terminal 500 used in the learning phase also has a battery device 507, communication device 508, connection device 509, and various sensors 510. In the mobile terminal 500, each piece of hardware is connected via a bus 520.

[0031] The processor 501 has various computing devices such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor 501 executes various programs (for example, training data collection programs, etc.) by reading them into the memory 502.

[0032] The memory 502 has main memory devices such as ROM (Read Only Memory) and RAM (Random Access Memory). The processor 501 and the memory 502 form a so-called computer (also called an information processing device), and the computer realizes various functions by the processor 501 executing various programs read into the memory 502.

[0033] The auxiliary storage device 503 stores various programs and various data used when these programs are executed by the processor 501.

[0034] The display device 504 is a device that outputs the results of internal processing in the mobile terminal 500, and the operating device 505 is a device for the operator to input various instructions and information to the mobile terminal 500. The GPS device 506 is a device that calculates the location information of the mobile terminal 500 based on signals from GPS (Global Positioning System) satellites.

[0035] The battery device 507 is a storage battery that supplies power to each piece of hardware in the mobile terminal 500. The communication device 508 is a device that communicates with a network (not shown), and the connection device 509 is a device for connecting external equipment with the mobile terminal 500.

[0036] The various sensors 510 are a group of sensors that measure data useful for predicting the heat index and data that can be acquired by the mobile terminal 500. The various sensors 510 include: a sensor that measures the battery temperature of the mobile terminal 500; a sensor that measures the operating status of the mobile terminal 500; a sensor that measures the atmospheric pressure around the mobile terminal 500; a sensor that measures the illuminance around the mobile terminal 500; a sensor that measures the temperature of hardware other than the battery of the mobile terminal 500, etc.

[0037] The various programs installed in the auxiliary storage device 503 are installed, for example, by downloading them from a network (not shown) via the communication device 508. Alternatively, the various programs may be pre-installed when the mobile terminal 500 is shipped or manufactured.

[0038] (2) Configuration of the training data generation system in the training phase Next, we will describe the configuration of the training data generation system for generating training data in the training phase, and the functional configuration of each device that makes up the training data generation system.

[0039] Figure 6 shows an example of the configuration of a learning data generation system and the functional configuration of each device. As shown in Figure 6, the learning data generation system 600 comprises a black globe heatstroke index meter 610, a portable terminal 500, and a learning data generation device 630. The black globe heatstroke index meter 610 and the portable terminal 500 are installed in the same location and are connected to the learning data generation device 630 in a communicative manner. However, the black globe heatstroke index meter 610 and the portable terminal 500 do not necessarily need to be connected to the learning data generation device 630 in a communicative manner. The black globe heatstroke index meter 610 and the portable terminal 500 should be configured to record the measured data, and the learning data generation device 630 should be able to acquire the measured data at any time after the measurement is completed. Furthermore, "installed in the same location" here is not limited to being installed adjacent to each other, but also includes being installed as close as possible so that the environment is the same or approximately the same. Therefore, in cases where it is not possible to install them adjacent to each other due to various circumstances, even if they are installed as close as possible, they shall be interpreted as being "installed in the same location."

[0040] The black globe heatstroke index meter 610 is a measuring device that acquires various data when the data used for training the heat index prediction model described later is measured by the portable terminal 500, and predicts the heat index. It is an alternative measuring device that can predict the heat index on an individual basis. In this context, an alternative measuring device refers to a measuring device that predicts the heat index in place of a dedicated measuring device that can measure wet bulb temperature, black globe temperature, and dry bulb temperature, which are necessary for calculating the heat index.

[0041] The black globe heatstroke index meter 610 is equipped with sensors that measure ambient temperature, humidity, and black globe temperature, and functions as a temperature data acquisition unit 611, a humidity data acquisition unit 612, a black globe temperature data acquisition unit 613, and a heat index prediction unit 614.

[0042] The ambient temperature data acquisition unit 611 acquires the ambient temperature data measured by the sensor, the humidity data acquisition unit 612 acquires the humidity data measured by the sensor, and the globe temperature data acquisition unit 613 acquires the globe temperature data measured by the sensor.

[0043] The heat index prediction unit 614 predicts the heat index using the acquired ambient temperature data, humidity data, and globe temperature data. The heat index predicted by the heat index prediction unit 614 is transmitted to the learning data generation device 630.

[0044] In this embodiment, as the correct data, the heat index predicted by the globe-type heat stroke index meter 610, which is an alternative measuring instrument, is used. However, as the correct data, the heat index calculated from the wet bulb temperature, globe temperature, and dry bulb temperature measured by a dedicated measuring instrument may also be used.

[0045] A learning data collection program is installed in the mobile terminal 500. When the learning data collection program is executed, the mobile terminal 500 functions as: ・ the battery temperature data acquisition unit 621, ・ the operating status data acquisition unit 622, ・ the atmospheric pressure data acquisition unit 623, ・ the illuminance data acquisition unit 624, ・ the other H / W temperature data acquisition unit 625, ・ the position information acquisition unit 626, ・ the weather information acquisition unit 627.

[0046] The battery temperature data acquisition unit 621 acquires the measured battery temperature data of the mobile terminal 500 (temperature data of the battery device 507). The operating status data acquisition unit 622 acquires the measured operating status data of the mobile terminal 500. The operating status data of the mobile terminal 500 may include, for example, any of the data of the CPU usage rate, the GPU usage rate, the memory usage rate, the usage time of the application, and the data indicating the communication load of the network communication. Alternatively, data other than these exemplified data may be included.

[0047] The atmospheric pressure data acquisition unit 623 acquires the measured atmospheric pressure data (atmospheric pressure data around the mobile terminal 500). The illuminance data acquisition unit 624 acquires the measured illuminance data (illuminance data around the mobile terminal 500). The other H / W temperature data acquisition unit 625 acquires other H / W temperature data measured by a sensor that measures the temperature of hardware other than the battery device 507 among the hardware of the mobile terminal 500.

[0048] The position information acquisition unit 626 acquires the position information (latitude, longitude, altitude) of the mobile terminal 500 calculated by the GPS device 506.

[0049] The weather information acquisition unit 627 acquires weather information (information such as maximum temperature, minimum temperature, precipitation, wind direction, wind speed, humidity, solar radiation amount, etc.) corresponding to the position information based on the position information of the mobile terminal 500 acquired by the position information acquisition unit 626 via a network not shown.

[0050] The battery temperature data, operation status data, atmospheric pressure data, illuminance data, other H / W temperature data, position information, and weather information acquired by each functional unit (battery temperature data acquisition unit 621 to weather information acquisition unit 627) of the mobile terminal 500 are transmitted to the learning data generation device 630.

[0051] The learning data generation device 630 is a general-purpose information processing terminal and functions as a learning data generation unit 631. The learning data generation unit 631 generates learning data with ・ the battery temperature data, operation status data, atmospheric pressure data, illuminance data, other H / W temperature data, position information, and weather information transmitted from the mobile terminal 500 as input data, ・ the heat index transmitted from the black globe type heat stroke index meter 610 as correct answer data, and stores it in the learning data storage unit 632.

[0052] FIG. 7 is an example of learning data. As shown in FIG. 7, the learning data 700 includes "input data" and "correct answer data" as information items.

[0053] The "input data" includes battery temperature data, operation status data, atmospheric pressure data, illuminance data, other H / W temperature data, position information, and weather information. The "correct answer data" includes the heat index.

[0054] (3) Explanation of Input Data for Training Next, we will explain the reason why the training data used when training the heat index prediction model is configured to include battery temperature data, operating status data, atmospheric pressure data, illuminance data, other hardware temperature data, location information, and weather information as input data.

[0055] In generating training data using the training data generation system 600, the applicant examined data that is useful for predicting the heat index from among the data that can be acquired by the mobile terminal 500.

[0056] Figure 8 shows the relationship between the data used to predict the heat index in a black globe thermometer and the battery temperature data of a mobile device. In Figure 8, graphs 810 and 820 show the data acquired by the black globe thermometer 610 when predicting the heat index (in the example in Figure 8, air temperature data and black globe temperature data) and the battery temperature data, which is among the data that the mobile device 500 can acquire. Note that graphs 810 and 820 are graphs for different dates and different time periods.

[0057] Table 830 shows the correlation between the various data used to predict the heat index in the black globe heatstroke index meter 610 (in the example in Figure 8, temperature data, black globe temperature data, and humidity data) and the battery temperature data, which is among the data that the mobile terminal 500 can acquire.

[0058] As can be seen from Graphs 810 and 820 and Table 830 in Figure 8, the battery temperature data acquired by the mobile terminal 500 shows a high correlation with the air temperature data and the black globe temperature data. Therefore, it is considered useful to use battery temperature data as a substitute for air temperature data and black globe temperature data when predicting the heat index.

[0059] On the other hand, it is expected that the battery temperature data of the mobile terminal 500 will be affected by the operating status data of the mobile terminal 500. Therefore, when using battery temperature data as a substitute for ambient temperature data and black globe temperature data, the system is configured to use operating status data as input data so that the battery temperature data is corrected by the operating status data.

[0060] Furthermore, when using battery temperature data as a substitute for air temperature data and black globe temperature data, it is desirable that it be corrected using atmospheric pressure information and solar radiation information. For this reason, the system is configured to use atmospheric pressure data and illuminance data as input data.

[0061] Furthermore, in order to predict the heat index, it is desirable to input wet-bulb temperature data, but the mobile terminal 500 cannot directly measure this data. On the other hand, wet-bulb temperature data can be interpreted as data that reflects wind speed and humidity information in addition to air temperature data. Therefore, the system was configured to use meteorological information (wind speed, humidity) as input data so that wind speed and humidity information are reflected.

[0062] Furthermore, to improve the accuracy of predicting the heat index, it is desirable that the above input data be corrected using data that causes errors, so that the error factors in the input data are eliminated. Therefore, the system was configured to use other hardware temperature data (temperature data of hardware other than the battery), which is an error factor in the battery temperature data, as input data. In addition, the system was configured to use altitude (information included in location information), which is an error factor in the atmospheric pressure data, as input data.

[0063] Furthermore, to improve the accuracy of predicting the heat index, it is desirable that sudden outliers in the input data be corrected so that the heat index is not affected by such outliers. Therefore, the system is configured to use meteorological information (maximum temperature, minimum temperature, precipitation, wind direction, wind speed, humidity, solar radiation, etc.) as input data so that such outliers are corrected.

[0064] (4) Flow of the learning data generation process by the learning data generation system Next, the flow of the learning data generation process by the learning data generation system 600 will be explained. Figure 9 is a flowchart showing the flow of the learning data generation process by the learning data generation system.

[0065] In step S901, the mobile terminal 500 acquires data to be used for generating training data.

[0066] In step S902, the mobile terminal 500 transmits the acquired data to the learning data generation device 630.

[0067] In step S903, the black globe thermometer 610 predicts the heat index.

[0068] In step S904, the black globe heatstroke index meter 610 transmits the predicted heat index to the learning data generation device 630.

[0069] In step S905, the training data generation device 630 determines whether a sufficient amount of data has been acquired to generate training data. If it is determined in step S905 that a sufficient amount of data has not been acquired (i.e., the answer is NO in step S905), the process returns to step S901. On the other hand, if it is determined in step S905 that a sufficient amount of data has been acquired (i.e., the answer is YES in step S905), the process proceeds to step S906.

[0070] In step S906, the learning data generation device 630 generates learning data based on the acquired data.

[0071] In step S907, the learning data generation device 630 stores the generated learning data in the learning data storage unit 632.

[0072] (5) Functional Configuration of the Learning Device Next, the functional configuration of the learning device that learns the heat index prediction model using the learning data generated by the learning data generation system 600 will be described. Figure 10 is a diagram showing an example of the functional configuration of the learning device. The learning device 1000 is realized by installing a learning program on a general-purpose information processing terminal. As shown in Figure 10, the learning device 1000 functions as a heat index prediction model 1010 and a comparison / modification unit 1020.

[0073] The heat index prediction model 1010 predicts the heat index when it receives input data from the training data storage unit 632, which is read from the training data storage unit 632.

[0074] The comparison / modification unit 1020 compares the heat index predicted by the heat index prediction model 1010 with the correct data from the training data 700 and calculates the error. The comparison / modification unit 1020 also trains the heat index prediction model by updating its model parameters based on the calculated error. This generates a trained heat index prediction model.

[0075] (6) Flow of learning process by the learning device Next, the flow of learning process by the learning device 1000 will be explained. Figure 11 is a flowchart showing the flow of learning process by the learning device.

[0076] In step S1101, the learning device 1000 reads the learning data 700 from the learning data storage unit 632.

[0077] In step S1102, the learning device 1000 uses the read-out training data 700 to perform a training process on the heat index prediction model.

[0078] In step S1103, the learning device 1000 determines whether the learning process has converged. If it determines in step S1103 that the learning process has not converged (i.e., the answer is NO in step S1103), the process returns to step S1101. On the other hand, if it determines in step S1103 that the learning process has converged (i.e., the answer is YES in step S1103), the learning process is terminated.

[0079] <Details of the Prediction Phase> Next, we will explain the details of the prediction phase.

[0080] (1) Hardware Configuration of the Heat Index Prediction Device First, we will describe the hardware configuration of the heat index prediction devices 400_1 and 400_2. As mentioned above, the heat index prediction devices 400_1 and 400_2 are realized by installing a heat index prediction program on a mobile terminal that users 410 and 420 carry with them on a daily basis. Here, the hardware configuration of the mobile terminal that users 410 and 420 carry with them on a daily basis is the same as the hardware configuration of the mobile terminal 500 explained using Figure 5, for example. Therefore, we will omit the explanation of the hardware configuration of the mobile terminal used as the heat index prediction devices 400_1 and 400_2 here. However, in the case of the mobile terminal used as the heat index prediction devices 400_1 and 400_2, the heat index prediction program is installed instead of the training data collection program.

[0081] (2) Functional Configuration of the Heat Index Prediction Device Next, the functional configuration of the heat index prediction devices 400_1 and 400_2 will be described. Figure 12 is a diagram showing an example of the functional configuration of the heat index prediction device. As shown in Figure 12, in the prediction phase, the heat index prediction program is executed, and the heat index prediction devices 400_1 and 400_2 function as a heat index prediction unit 1200.

[0082] The heat index prediction unit 1200 includes: a battery temperature data acquisition unit 621, an operating status data acquisition unit 622, a barometric pressure data acquisition unit 623, an illuminance data acquisition unit 624, a other H / W temperature data acquisition unit 625, a location information acquisition unit 626, a weather information acquisition unit 627, a learned heat index prediction model 1010', and a display unit 1210, and acquires the heat index.

[0083] Of these, the battery temperature data acquisition unit 621 to the weather information acquisition unit 627 are functional units similar to those of the mobile terminal 500 shown in Figure 6 (battery temperature data acquisition unit 621 to weather information acquisition unit 627), so their explanation is omitted here.

[0084] The trained heat index prediction model 1010' is a model generated by the learning device 1000 through a training process on the heat index prediction model 1010, and is stored in the prediction model storage unit 1220. The trained heat index prediction model 1010' is read from the prediction model storage unit 1220 when the heat index prediction program is executed, and the following are input to predict the heat index: - Battery temperature data acquired by the battery temperature data acquisition unit 621, - Operating status data acquired by the operating status data acquisition unit 622, - Atmospheric pressure data acquired by the atmospheric pressure data acquisition unit 623, - Illuminance data acquired by the illuminance data acquisition unit 624, - Other H / W temperature data acquired by the other H / W temperature data acquisition unit 625, - Location information acquired by the location information acquisition unit 626, and - Weather information acquired by the weather information acquisition unit 627. The heat index predicted by the trained heat index prediction model 1010' is notified to the display unit 1210.

[0085] The display unit 1210 is an example of an output unit, and displays the heat index notified by the trained heat index prediction model 1010' to users 410 and 420.

[0086] (3) Flow of prediction processing by the heat index prediction device Next, we will explain the flow of prediction processing by the heat index prediction devices 400_1 and 400_2. Figure 13 is a flowchart showing the flow of prediction processing by the heat index prediction device.

[0087] In step S1301, the heat index prediction devices 400_1 and 400_2 determine whether or not a command to start the heat index prediction program has been input from users 410 and 420. If it is determined in step S1301 that no command to start the heat index prediction program has been input (i.e., the answer in step S1301 is NO), the devices wait until a command to start the heat index prediction program is input.

[0088] On the other hand, if it is determined in step S1301 that an instruction to start the heat index prediction program has been entered (if the answer is YES in step S1301), the heat index prediction program is started and the process proceeds to step S1302.

[0089] In step S1302, the heat index prediction devices 400_1 and 400_2 acquire data used for predicting the heat index.

[0090] In step S1303, the heat index prediction devices 400_1 and 400_2 operate the trained heat index prediction model using the acquired data and obtain the heat index predicted by the trained heat index prediction model.

[0091] In step S1304, the heat index prediction devices 400_1 and 400_2 display the predicted heat index and information corresponding to the predicted heat index to the users 410 and 420.

[0092] In step S1305, the heat index prediction devices 400_1 and 400_2 determine whether or not to continue the prediction process. For example, if the users 410 and 420 have not given an instruction to terminate the heat index prediction program, the heat index prediction devices 400_1 and 400_2 determine to continue the prediction process. On the other hand, if the users 410 and 420 have given an instruction to terminate the heat index prediction program, the heat index prediction devices 400_1 and 400_2 determine not to continue the prediction process.

[0093] If it is determined in step S1305 to continue the prediction process (if the answer is YES in step S1305), the process returns to step S1302. On the other hand, if it is determined in step S1305 not to continue the prediction process (if the answer is NO in step S1305), the prediction process is terminated.

[0094] (4) Examples of screens for the heat index prediction device Next, examples of screens for the heat index prediction devices 400_1 and 400_2 will be described. Figure 14 shows an example of a screen for the heat index prediction device.

[0095] Screen 1401 in Figure 14(a) shows the screen in the heat index prediction devices 400_1 and 400_2 before the instruction to start the heat index prediction program is input. On screen 1401, the heat index prediction program is started when the “heat index prediction app” icon is tapped by users 410 and 420.

[0096] Screen 1402 in Figure 14(b) shows the screen displayed in the heat index prediction devices 400_1 and 400_2 after the heat index prediction program has been activated and the predicted heat index has been displayed.

[0097] The screen 1403 in Figure 14(c) shows the screens of the heat index prediction devices 400_1 and 400_2 that display information corresponding to the predicted heat index (labels, precautions corresponding to the labels, activity guidelines, etc.).

[0098] (5) Prediction Accuracy Next, we will explain the prediction accuracy of the heat index predicted by the heat index prediction devices 400_1 and 400_2. Figure 15 is a diagram showing an example of the prediction accuracy of the heat index predicted by the heat index prediction devices. In the graph of Figure 15, the horizontal axis represents time, and the vertical axis represents the heat index. Also, in the graph of Figure 15, the measured value is the heat index predicted by the black globe heatstroke index meter 610, which is located in the same position as the heat index prediction devices 400_1 and 400_2, and the predicted value is the heat index predicted by the heat index prediction devices 400_1 and 400_2.

[0099] A comparison of the measured values ​​and the predicted values ​​revealed that the RMSE (Root Mean Squared Error) was 0.1339 [°C] and the MAE (Mean Absolute Error) was 0.0643 [°C]. Thus, it was found that the heat index prediction program of this embodiment can predict the heat index with an accuracy comparable to that of the black globe heatstroke index meter 610 (alternative measuring device) by predicting the heat index based on data obtainable by the mobile terminal 500.

[0100] <Summary> As is clear from the above explanation, in the first embodiment, the information processing device (computer) of the mobile terminal used as the heat index prediction device 400_1, 400_2 takes as input data: battery temperature data, atmospheric pressure data, illuminance data acquired by the mobile terminal for generating a trained heat index prediction model, and other H / W temperature data, which is the temperature of other hardware owned by the mobile terminal, operating status data of the mobile terminal, location information of the mobile terminal, and weather information corresponding to the location information, acquired by the mobile terminal for generating a trained heat index prediction model, and operates a trained heat index prediction model which has been learned using the heat index measured by an alternative measuring device as the ground truth data when the input data is measured. - By running a pre-trained heat index prediction model, the heat index predicted by the pre-trained heat index prediction model is obtained based on battery temperature data, atmospheric pressure data, illuminance data acquired by the mobile device for obtaining the heat index, as well as other hardware temperature data acquired by the mobile device for obtaining the heat index, operating status data of the mobile device, location information of the mobile device, and weather information corresponding to the location information.

[0101] As a result, according to the first embodiment, it becomes possible to predict the heat index using data useful for predicting the heat index from among the data that can be acquired by a mobile device that the user carries on a daily basis. Consequently, according to the first embodiment, it becomes possible to predict the heat index with high accuracy on an individual basis.

[0102] [Second Embodiment] In the first embodiment described above, it was assumed that a heat index prediction program including a trained heat index prediction model 1010' is installed on the mobile terminal 500. However, the heat index prediction program including the trained heat index prediction model 1010' may be installed, for example, on a server device to which the mobile terminal 500 is communicatively connected. This allows the server device to be used as a heat index prediction device that provides a heat index prediction service. In this case, a heat index output program may be installed on the mobile terminal 500. This allows the mobile terminal 500 to be used as a heat index output device that outputs a heat index using the heat index prediction service.

[0103] The following describes the second embodiment, focusing on the differences from the first embodiment described above.

[0104] <Configuration of the Heat Index Prediction System> First, we will explain the configuration of the heat index prediction system and the functional configuration of each device that makes up the heat index prediction system.

[0105] Figure 16 shows an example of the configuration of the heat index prediction system and the functional configuration of each device. As shown in Figure 16, the heat index prediction system 1600 includes a heat index output device 1610 and a heat index prediction device 1630. The heat index output device 1610 and the heat index prediction device 1630 are communicated together via a network 1650.

[0106] The heat index output device 1610 is realized by installing a heat index output program on the mobile terminal 500. When this program is executed, the heat index output device 1610 functions as a heat index output unit 1620.

[0107] The heat index output unit 1620 includes: a battery temperature data acquisition unit 621, an operating status data acquisition unit 622, a barometric pressure data acquisition unit 623, an illuminance data acquisition unit 624, a other H / W temperature data acquisition unit 625, a location information acquisition unit 626, a weather information acquisition unit 627, a data transmission unit 1621, a prediction result receiving unit 1622, and a display unit 1210, and acquires the heat index.

[0108] Of these, the battery temperature data acquisition unit 621 to the weather information acquisition unit 627 have the same functions as the respective functional units (battery temperature data acquisition unit 621 to weather information acquisition unit 627) of the heat index prediction devices 400_1 and 400_2 shown in Figure 12, so their explanation is omitted here.

[0109] The data transmission unit 1621 accesses the heat index prediction device 1630 via the network 1650 and transmits a request to use the heat index prediction service provided by the heat index prediction device 1630, as well as transmitting the following to the heat index prediction device 1630: - Battery temperature data acquired by the battery temperature data acquisition unit 621, - Operating status data acquired by the operating status data acquisition unit 622, - Atmospheric pressure data acquired by the atmospheric pressure data acquisition unit 623, - Illuminance data acquired by the illuminance data acquisition unit 624, - Other H / W temperature data acquired by the other H / W temperature data acquisition unit 625, - Location information acquired by the location information acquisition unit 626, and - Weather information acquired by the weather information acquisition unit 627.

[0110] The prediction result receiving unit 1622 receives the heat index predicted by the heat index prediction device 1630 in response to the data transmission unit 1621 sending a request to use the heat index prediction service. The prediction result receiving unit 1622 notifies the display unit 1210 of the received heat index.

[0111] The display unit 1210 displays the heat index notified by the prediction result receiving unit 1622 to the users 410 and 420.

[0112] The heat index prediction device 1630 is realized by installing a heat index prediction program on a server device. When this program is executed, the heat index prediction device 1630 functions as a heat index prediction unit 1640.

[0113] The heat index prediction unit 1640 includes: a data input unit 1641, a learned heat index prediction model 1010', and a prediction result transmission unit 1642, and acquires the heat index. As a result, the heat index prediction unit 1640 can provide a heat index prediction service to the heat index output device 1610 via the network 1650.

[0114] The data input unit 1641 receives a request from the heat index output device 1610 to use the heat index prediction service, and also receives the following: battery temperature data acquired by the battery temperature data acquisition unit 621, operating status data acquired by the operating status data acquisition unit 622, atmospheric pressure data acquired by the atmospheric pressure data acquisition unit 623, illuminance data acquired by the illuminance data acquisition unit 624, other hardware temperature data acquired by the other hardware temperature data acquisition unit 625, location information acquired by the location information acquisition unit 626, and weather information acquired by the weather information acquisition unit 627, and inputs them into the trained heat index prediction model 1010'.

[0115] The trained heat index prediction model 1010' has already been explained using Figure 12 in the first embodiment described above, so its explanation will be omitted here.

[0116] The prediction result transmission unit 1642 transmits the heat index, which is the prediction result predicted by the trained heat index prediction model 1010', to the heat index output device 1610 that has requested its use, via the network 1650.

[0117] <Hardware Configuration of Each Device Constituting the Heat Index Prediction System> Next, the hardware configuration of each device constituting the heat index prediction system (heat index output device 1610, heat index prediction device 1630) will be described. Of the devices constituting the heat index prediction system, the heat index output device 1610 was realized by installing a heat index output program on the mobile terminal 500, as described above. The hardware configuration of the mobile terminal 500 has already been explained using, for example, Figure 5, so the explanation will be omitted here.

[0118] On the other hand, the heat index prediction device 1630 is implemented by installing a heat index prediction program on the server device, as described above. Therefore, the hardware configuration of the server device will be described here.

[0119] Figure 17 shows an example of the hardware configuration of a server device. As shown in Figure 17, the server device 1700 includes a processor 1701, memory 1702, auxiliary storage device 1703, connection device 1704, communication device 1705, and drive device 1706. The hardware components of the server device 1700 are interconnected via a bus 1707.

[0120] The processor 1701 has various computing devices such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor 1701 executes various programs (for example, a heat index prediction program) by reading them into the memory 1702.

[0121] The memory 1702 has main memory devices such as ROM (Read Only Memory) and RAM (Random Access Memory). The processor 1701 and the memory 1702 form a so-called computer (also called an information processing device), and the computer realizes various functions by executing various programs read into the memory 1702 by the processor 1701.

[0122] The auxiliary storage device 1703 stores various programs and various data used when these programs are executed by the processor 1701.

[0123] The connection device 1704 is a connection device for connecting the display device 1711 and the operating device 1712. The communication device 1705 is a communication device for communicating with a mobile terminal 500, etc., via the network 1650.

[0124] The drive device 1706 is a device for setting the recording medium 1713. The recording medium 1713 here includes media for recording information optically, electrically, or magnetically, such as CD-ROMs, flexible disks, and magneto-optical disks. The recording medium 1713 may also include semiconductor memory for recording information electrically, such as ROMs and flash memory.

[0125] The various programs to be installed on the auxiliary storage device 1703 are installed, for example, when the distributed recording medium 1713 is set in the drive device 1706 and the various programs recorded on the recording medium 1713 are read. Alternatively, the various programs to be installed on the auxiliary storage device 1703 may be installed by downloading them from the network 1650 via the communication device 1705.

[0126] <Processing flow by each device constituting the heat index prediction system> Next, the processing flow by each device constituting the heat index prediction system 1600 (heat index output device 1610, heat index prediction device 1630) will be explained using Figure 18. Figure 18 is a flowchart showing the processing flow by each device of the heat index prediction system.

[0127] (1) Flow of heat index output processing by heat index output device First, the flow of heat index output processing by heat index output device 1610 will be explained. Figure 18(a) is a flowchart showing the flow of heat index output processing by heat index output device 1610 in the heat index prediction system 1600.

[0128] In step S1801, the heat index output device 1610 determines whether or not a command to start the heat index output program has been input from users 410 and 420. If it is determined in step S1801 that no command to start the heat index output program has been input (i.e., the answer in step S1801 is NO), the device waits until a command to start the heat index output program is input.

[0129] On the other hand, if it is determined in step S1801 that an instruction to start the heat index output program has been received (if the answer is YES in step S1801), the heat index output program is started and the process proceeds to step S1802.

[0130] In step S1802, the heat index output device 1610 acquires data used to predict the heat index.

[0131] In step S1803, the heat index output device 1610 accesses the heat index forecasting device 1630, sends a request to use the heat index forecasting service, and also sends data used for forecasting the heat index.

[0132] In step S1804, the heat index output device 1610 receives the heat index predicted by the heat index prediction device 1630.

[0133] In step S1805, the heat index output device 1610 displays the received heat index and information corresponding to the received heat index to the users 410 and 420.

[0134] In step S1806, the heat index output device 1610 determines whether or not to continue the heat index output process. If it is determined in step S1806 to continue the heat index output process (if the answer is YES in step S1806), the process returns to step S1802. On the other hand, if it is determined in step S1806 not to continue the heat index output process (if the answer is NO in step S1806), the heat index output process is terminated.

[0135] (2) Flow of prediction processing by the heat index prediction device Next, the flow of prediction processing by the heat index prediction device 1630 will be explained. Figure 18(b) is a flowchart showing the flow of prediction processing by the heat index prediction device 1630 in the heat index prediction system 1600.

[0136] In step S1811, the heat index prediction device 1630 determines whether or not there is access from the heat index output device 1610. If it is determined in step S1811 that there is no access (if the answer is NO in step S1811), it waits until it is determined that there has been access.

[0137] On the other hand, if it is determined in step S1811 that an access has occurred (i.e., if the answer is YES in step S1811), the process proceeds to step S1812.

[0138] In step S1812, the heat index prediction device 1630 receives a request to use the heat index prediction service from the heat index output device 1610, and also receives data used for predicting the heat index.

[0139] In step S1813, the heat index prediction device 1630 operates a trained heat index prediction model using the received data and obtains the heat index predicted by the trained heat index prediction model.

[0140] In step S1814, the heat index prediction device 1630 transmits the predicted heat index to the heat index output device 1610 that accessed it.

[0141] In step S1815, the heat index forecasting device 1630 determines whether to continue the forecasting process. If it determines in step S1815 to continue the forecasting process (if the answer is YES in step S1815), it returns to step S1811. On the other hand, if it determines in step S1815 not to continue the forecasting process (if the answer is NO in step S1815), it terminates the forecasting process.

[0142] <Summary> As is clear from the above explanation, in the second embodiment, the information processing device (computer) of the server device used as the heat index prediction device 1630 takes as input data: battery temperature data, atmospheric pressure data, illuminance data acquired by a mobile terminal for generating a trained heat index prediction model, and other H / W temperature data, which is the temperature of other hardware owned by the mobile terminal, operating status data of the mobile terminal, location information of the mobile terminal, and weather information corresponding to the location information, all acquired by the mobile terminal for generating a trained heat index prediction model. When the input data is measured, the trained heat index prediction model is operated using the heat index measured by an alternative measuring device as the ground truth data. - By running a pre-trained heat index prediction model, the heat index predicted by the pre-trained heat index prediction model is obtained based on battery temperature data, atmospheric pressure data, illuminance data acquired by the mobile device for obtaining the heat index, as well as other hardware temperature data acquired by the mobile device for obtaining the heat index, operating status data of the mobile device, location information of the mobile device, and weather information corresponding to the location information.

[0143] As a result, according to the second embodiment, it becomes possible to predict the heat index using data useful for predicting the heat index from among the data that can be acquired by a mobile device that the user carries on a daily basis. Consequently, according to the second embodiment, it becomes possible to predict the heat index with high accuracy on an individual basis.

[0144] Furthermore, in the second embodiment, the information processing device (computer) of the mobile terminal used as the heat index output device 1610 takes as input data: battery temperature data, atmospheric pressure data, illuminance data acquired by the mobile terminal for generating a trained heat index prediction model, and other H / W temperature data, which is the temperature of other hardware owned by the mobile terminal, operating status data of the mobile terminal, location information of the mobile terminal, and weather information corresponding to the location information, all acquired by the mobile terminal for generating a trained heat index prediction model. When the input data is measured, the trained heat index prediction model is operated using the heat index measured by an alternative measuring device as the ground truth data. - By running a pre-trained heat index prediction model, the heat index predicted by the pre-trained heat index prediction model is obtained based on battery temperature data, atmospheric pressure data, illuminance data acquired by the mobile device for obtaining the heat index, as well as other hardware temperature data acquired by the mobile device for obtaining the heat index, operating status data of the mobile device, location information of the mobile device, and weather information corresponding to the location information.

[0145] As a result, according to the second embodiment, it becomes possible to predict the heat index using data useful for predicting the heat index from among the data that can be acquired by a mobile device that the user carries on a daily basis. Consequently, according to the second embodiment, it becomes possible to predict the heat index with high accuracy on an individual basis.

[0146] [Third Embodiment] In each of the above embodiments, the heat index was predicted by inputting battery temperature data, operating status data, atmospheric pressure data, illuminance data, other H / W temperature data, location information, and weather information into a trained heat index prediction model. However, the data to be input into the trained heat index prediction model when predicting the heat index is not limited to these.

[0147] For example, data used to correct battery temperature data, atmospheric pressure data, and illuminance data (operating status data, other hardware temperature data, location information, weather information) may be excluded from the data input to the trained heat index prediction model.

[0148] Alternatively, other data (operating status data, other hardware temperature data, location information, and data other than weather information) that corrects battery temperature data, atmospheric pressure data, and illuminance data may be input into the trained heat index prediction model.

[0149] Alternatively, the data input to the trained heat index prediction model may be battery temperature data, while atmospheric pressure data, illuminance data, etc., used to correct the battery temperature data may be excluded from the data input to the trained heat index prediction model. In other words, the trained heat index prediction model may predict the heat index from the battery temperature data.

[0150] Alternatively, the data input to the trained heat index prediction model may consist of battery temperature data and atmospheric pressure data, while illuminance data and other data used to correct the battery temperature data may be excluded from the input data to the trained heat index prediction model. In other words, the trained heat index prediction model may predict the heat index from battery temperature data and atmospheric pressure data.

[0151] Alternatively, the data input to the trained heat index prediction model may consist of battery temperature data and illuminance data, while atmospheric pressure data and other data used to correct the battery temperature data may be excluded from the input data to the trained heat index prediction model. In other words, the trained heat index prediction model may predict the heat index from battery temperature data and illuminance data.

[0152] Alternatively, the data input to the trained heat index prediction model may consist of battery temperature data and weather information, while atmospheric pressure data, illuminance data, etc., used to correct the battery temperature data may be excluded from the data input to the trained heat index prediction model. In other words, the trained heat index prediction model may predict the heat index from battery temperature data and weather information.

[0153] Alternatively, the data input to the trained heat index prediction model may consist of battery temperature data and other hardware temperature data, while atmospheric pressure data, illuminance data, etc., used to correct the battery temperature data may be excluded from the data input to the trained heat index prediction model. In other words, the trained heat index prediction model may predict the heat index from battery temperature data and other hardware temperature data.

[0154] Alternatively, the data input to the trained heat index prediction model may consist of battery temperature data and operating status data, while atmospheric pressure data, illuminance data, etc., used to correct the battery temperature data may be excluded from the data input to the trained heat index prediction model. In other words, the trained heat index prediction model may predict the heat index from battery temperature data and operating status data.

[0155] Furthermore, although the above embodiments have described the use of a machine learning model as the model for predicting the heat index, the model for predicting the heat index is not limited to a machine learning model. At a minimum, any model that can predict the heat index from battery temperature data, atmospheric pressure data, and illuminance data is applicable. Any model referred to here includes models that calculate the heat index using a function that represents the relationship between battery temperature data, atmospheric pressure data, illuminance data, and the heat index.

[0156] Furthermore, in each of the above embodiments, the case in which the predicted heat index and information corresponding to the heat index are displayed and output to users 410 and 420 has been described. However, the method of outputting the predicted heat index and information corresponding to the heat index to users 410 and 420 is not limited to this. For example, the predicted heat index and information corresponding to the heat index may be output to users 410 and 420 by voice or warning sound.

[0157] Furthermore, in the first embodiment described above, the heat index prediction program was launched by users 410 and 420 tapping the icon for the "heat index prediction app." However, the method of launching the heat index prediction program is not limited to this. For example, the heat index prediction program may run continuously in the background of the heat index prediction devices 400_1 and 400_2, and notify users 410 and 420 when their risk of heatstroke increases.

[0158] It should be noted that the present invention is not limited to the configurations shown in the above embodiments, including combinations with other elements. These aspects can be modified without departing from the spirit of the present invention and can be appropriately determined according to their application.

[0159] This application claims priority based on Japanese Patent Application No. 2024-193158, filed on November 1, 2024, which is incorporated herein by reference to the entire contents of the said Japanese Patent Application.

[0160] 400_1, 4002_2: Heat index prediction device 500: Mobile terminal 600: Learning data generation system 610: Black globe type heatstroke index meter 621: Battery temperature data acquisition unit 622: Operating status data acquisition unit 623: Barometric pressure data acquisition unit 624: Illuminance data acquisition unit 625: Other H / W temperature data acquisition unit 626: Location information acquisition unit 627: Weather information acquisition unit 630: Learning data generation device 631: Learning data generation unit 700: Learning data 1000: Learning device 1010: Heat index prediction model 1010': Learned heat index prediction model 1200: Heat index prediction unit 1210: Display unit 1220: Prediction model storage unit 1610: Heat index output device 1620 : Heat index output unit 1621 : Data transmission unit 1622 : Prediction result receiving unit 1630 : Heat index prediction device 1640 : Heat index prediction unit 1641 : Data input unit 1642 : Prediction result transmission unit

Claims

1. An information processing device that uses battery temperature data from a mobile device as input data for model generation, and operates a model generated using the heat index at the time the input data is acquired as ground truth data, thereby acquiring the heat index predicted by the model based on battery temperature data acquired by the mobile device.

2. The information processing apparatus according to claim 1, wherein the input data further includes atmospheric pressure data and / or illuminance data, and the heat index is further predicted by the model based on atmospheric pressure data and / or illuminance data acquired by a mobile terminal for obtaining the heat index.

3. The information processing apparatus according to claim 1, wherein the input data further includes weather information, and the heat index is further predicted by the model based on weather information acquired by a mobile terminal for obtaining the heat index.

4. The information processing apparatus according to claim 3, wherein the weather information includes any of the following: maximum temperature, minimum temperature, precipitation, wind direction, wind speed, humidity, and solar radiation.

5. The information processing apparatus according to claim 1, wherein the input data further includes data indicating the operating status of the mobile terminal, and the heat index is further predicted by the model based on data indicating the operating status of the mobile terminal, which is acquired by the mobile terminal for obtaining the heat index.

6. The information processing apparatus according to claim 5, wherein the data indicating the operating status includes any of the following: CPU usage data, GPU usage data, memory usage data, application usage time data, or data indicating the communication load of network communication.

7. The information processing apparatus according to claim 1, wherein the input data further includes temperature data of other hardware owned by the mobile terminal, and the heat index is further predicted by the model based on the temperature data of other hardware owned by the mobile terminal, which is acquired by the mobile terminal for the purpose of obtaining the heat index.

8. The information processing apparatus according to claim 1, wherein the model is a trained model generated by machine learning using training data including the input data and the correct answer data.

9. An information processing device according to any one of claims 1 to 8, which outputs an acquired heat index and information corresponding to the heat index.

10. A heat index prediction method in which an information processing device performs the step of obtaining a heat index predicted by the model based on battery temperature data acquired by a mobile device, by using battery temperature data of a mobile device as input data for model generation, and operating the generated model using the heat index at the time the input data is acquired as ground truth data.

11. A program to cause an information processing device to execute the process of obtaining the heat index predicted by the model based on the battery temperature data acquired by the mobile device, by using battery temperature data from a mobile device as input data for model generation, and running the generated model using the heat index at the time the input data was acquired as ground truth data.

Images