Information processing device and information processing method

Abstract

To provide an information processing device and information processing method, which enable accurate weather prediction.SOLUTION: An information processing device disclosed herein comprises a processor configured to acquire first measurements of odor sensors and output first weather information indicative of the weather corresponding to the first measurements in response to an output request transmitted from an information terminal of a user.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present embodiment relates to an information processing device and an information processing method. [Background technology]

[0002] Conventionally, there is a technology for forecasting weather using meteorological observation data. The meteorological observation data used here includes air pressure data, temperature data, humidity data, wind direction data, wind speed data, precipitation data, solar radiation data, etc. Also, a service is known that provides the probability of precipitation by hour based on information from a raincloud radar that visualizes the location and movement of rainclouds. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special Publication No. 2022-531862 Summary of the Invention [Problem to be solved by the invention]

[0004] However, there was a problem in that the accuracy of conventional weather forecasts and the accuracy of estimating the probability of precipitation using information from raincloud radar were insufficient.

[0005] An object of the present invention is to provide an information processing device and an information processing method capable of estimating weather with high accuracy. [Means for solving the problem]

[0006] The information processing device according to the present invention includes a processor that acquires a first measurement value by an odor sensor and outputs first weather information indicating the weather corresponding to the first measurement value in response to an output request transmitted from a user's information terminal. Effect of the Invention

[0007] According to the present invention, it is possible to provide an information processing device and an information processing method that can estimate the weather with high accuracy. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of a system to which an information processing device is applied. [Diagram 2] FIG. 2 is a diagram illustrating an example of a configuration of the server. [Diagram 3] FIG. 3 is a diagram illustrating an example of a functional configuration of the server. [Figure 4] FIG. 4 is a diagram showing an example of the data configuration of a teacher data set for a certain area stored in the data storage unit. [Diagram 5] FIG. 5 is a flowchart showing the operation of generating a weather estimation model in the server. [Figure 6] FIG. 6 is a flowchart showing the operation of estimating weather in the server. [Figure 7] FIG. 7 is a diagram for explaining information transferred between the components during the series of operations shown in FIG. [Figure 8] FIG. 8 is a diagram for explaining information transferred between the components during the series of operations shown in FIG. [Figure 9] FIG. 9 is a diagram illustrating another example of the functional configuration of the server. [Figure 10] FIG. 10 is a diagram illustrating yet another example of the functional configuration of the server. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an information processing apparatus and an information processing method according to an embodiment will be described in detail with reference to the accompanying drawings. Note that the present invention is not limited to the embodiment.

[0010] (Embodiment) It is known that when it starts to rain, an odor called petrichor is produced, and after the rain, the odor caused by an odorous component called geosmion becomes stronger. Because these odors are directly related to rainfall, if we can detect these odors, we can estimate the weather more accurately than methods that estimate the weather based on information from raincloud radar or meteorological observation data.

[0011] Therefore, in the embodiment, an odor closely related to weather, such as the odor component of geosmin or the odor of petrichor, is measured by an odor sensor. The information processing device estimates the weather based on the measurement value of the odor sensor. To estimate the weather, the information processing device uses a trained model configured to be able to estimate the current weather from the measurement value of the odor sensor.

[0012] Hereinafter, the measurement value by the odor sensor will be simply referred to as sensor data. Furthermore, the model used to estimate the weather will be referred to as the weather estimation model. Furthermore, the weather estimated by the weather estimation model will be referred to as the estimated weather.

[0013] FIG. 1 is a diagram illustrating an example of a system to which an information processing device is applied.

[0014] One or more odor sensors 2a are arranged in each of a plurality of districts. Each district may be a prefecture, or an area smaller than a prefecture (e.g., a city, ward, town, or village). Each district may be an area divided in some way different from a prefecture or a city, ward, town, or village.

[0015] Here, as an example, areas A1 to A3 are shown as a plurality of areas. An odor sensor 2a-1 is placed in area A1, odor sensors 2a-2 and 2a-3 are placed in area A2, and odor sensors 2a-4, 2a-5, and 2a-6 are placed in area A3.

[0016] Each odor sensor 2a is connected to the server 1 via a wireless or wired communication path and a network 3. The odor sensors 2a placed in each district measure odors at the locations where they are placed and transmit the sensor data obtained by the measurement to the server 1.

[0017] This system can also utilize an odor sensor owned by a user. In the example shown in Fig. 1, a user 5-1 located in area A1 owns an information terminal 4-1 and an odor sensor 2b. The odor sensor 2b measures the odor at the position of the odor sensor 2b and outputs sensor data obtained by the measurement. The sensor data output by the odor sensor 2b is transmitted to the server 1 via the information terminal 4-1.

[0018] The arranged odor sensors 2a and the user-owned odor sensor 2b are collectively referred to as odor sensors 2. Moreover, the information terminal 4-1 and information terminals 4-2 to 4-6 described later are collectively referred to as information terminals 4. Moreover, the user 5-1 and users 5-2 to 5-6 described later are collectively referred to as users 5.

[0019] The odor sensor 2 includes an odor sensor element. The odor sensor 2 takes in air around the odor sensor 2, and the odor sensor element is exposed to the taken-in air. The odor sensor element then measures the amount of a specific odor component contained in the taken-in air. An element that reacts with an odor component that constitutes an odor closely related to weather and outputs an electrical signal according to the amount of the odor component is selected as the odor sensor element. The odor sensor 2 may include one sensor element, or may include two or more sensor elements that react to different types of odor components. The odor sensor 2 outputs data obtained by the odor sensor element as sensor data.

[0020] As the odor sensor element, a QCM (Quartz Crystal Microbalance) type sensor element, a MEMS (Micro Electro Mechanical Systems) type sensor element, an FBAR (Film Bulk Acoustic Resonator) type sensor element, or any other sensor element may be used.

[0021] The server 1 is an example of an information processing device according to the embodiment. That is, the server 1 can estimate the weather using sensor data.

[0022] For example, when a user 5-6 designates a certain area and requests the server 1 to output the weather via the information terminal 4-6, the server 1 acquires sensor data from the odor sensor 2b located in the designated area in advance or in response to the request. The server 1 then outputs weather information indicating the weather corresponding to the acquired sensor data, i.e., the estimated weather, to the information terminal 4-6.

[0023] Furthermore, in the embodiment, the server 1 can generate a weather estimation model used for estimating the weather as well as estimate the weather. To this end, the server 1 collects teacher data for training the weather estimation model.

[0024] The teacher data includes at least a pair of sensor data and weather information as correct answer data. The correct answer data is, for example, information input by a user located in the area about the weather at that time. For example, "sunny" is input. Alternatively, "rain" or "cloudy" can be input instead of "sunny". The correct answer data is sometimes called a weather answer. The server 1 can acquire sensor data to be used as teacher data from each of the arranged odor sensors 2a and the odor sensor 2b owned by the user.

[0025] When acquiring the sensor data, the user 5 is requested to input the current weather via the information terminal 4. When the server 1 receives a response about the weather from the user 5, the server 1 stores the response about the weather in association with the sensor data.

[0026] For example, in FIG. 1, in area A1, there is a user 5-1 who carries an information terminal 4-1 and owns a sensor 2b. In area A2, there is a user 5-2 who carries an information terminal 4-2 and a user 5-3 who carries an information terminal 4-3. In area A3, there is a user 5-5 who carries an information terminal 4-5. The server 1 transmits a weather input request to the information terminals 4-1 to 4-5 to request input of the current weather. The server 1 then receives weather responses from the information terminals 4-1 to 4-5.

[0027] The server 1 executes the collection of pairs of weather responses and sensor data for each district multiple times, thus obtaining a large number of pairs of sensor data and responded weather to be used for training.

[0028] A relationship between the sensor data and the weather is learned by training based on pairs of weather and sensor data. Based on the relationship between the sensor data and the weather optimized by training, it becomes possible to obtain the weather corresponding to the sensor data from the sensor data as an estimated weather.

[0029] The information terminal 4 is a computer capable of inputting and outputting information. The information terminal 4 is a smartphone, a wearable device, a personal computer, or the like. Information is input via a keyboard or a touch panel. Information is output via a display device or a sound output device.

[0030] Hereinafter, the weather information obtained from the user will be simply referred to as "user response."

[0031] FIG. 2 is a diagram illustrating an example of the configuration of the server 1. As shown in FIG.

[0032] The server 1 has a configuration capable of executing a computer program. Specifically, the server 1 has a processor 10, a network interface 11, and a memory 12.

[0033] The network interface 11 is an interface device for communicating with external devices via the network 3. In this embodiment, the odor sensor 2 and the information terminal 4 correspond to the external devices.

[0034] The processor 10 is a circuit capable of executing various calculations based on a computer program, and is, for example, a CPU (Central Processing Unit).

[0035] The memory 12 stores a processing program 100 in advance.

[0036] By executing the processing program 100, the processor 10 realizes functions such as collecting teacher data, generating a weather estimation model based on the teacher data, and estimating weather using the generated weather estimation model.

[0037] FIG. 3 is a diagram illustrating an example of a functional configuration of the server 1. As shown in FIG.

[0038] The server 1 includes a collection unit 101, a data storage unit 102, a model generation unit 103, a model storage unit 104, a weather estimation unit 105, and an output unit 106. The collection unit 101, the model generation unit 103, and the output unit 106 are realized by the processor 10 executing the processing program 100. The data storage unit 102 and the model storage unit 104 are realized by the memory 12.

[0039] The collection unit 101 executes collection of teacher data. The collection unit 101 stores the collected teacher data in the data storage unit 102. The collection unit 101 classifies the teacher data by district and stores the data in the data storage unit 102. That is, the data storage unit 102 stores teacher data set 200-A1 which is a set of teacher data related to district A1, teacher data set 200-A2 which is a set of teacher data related to district A2, and teacher data set 200-A3 which is a set of teacher data related to district A3.

[0040] The teacher data sets 200-A1 to 200-A3 are collectively referred to as teacher data set 200.

[0041] The data storage unit 102 stores a teacher data set 200. The data storage unit 102 may be a simple storage area, or may be a database.

[0042] The model generation unit 103 generates a weather estimation model. Specifically, the model generation unit 103 generates a weather estimation model 300-A1 that is a weather estimation model capable of estimating the current weather in area A1 through training using the teacher data set 200-A1. The model generation unit 103 generates a weather estimation model 300-A2 that is a weather estimation model capable of estimating the current weather in area A2 through training using the teacher data set 200-A2. The model generation unit 103 generates a weather estimation model 300-A3 that is a weather estimation model capable of estimating the current weather in area A3 through training using the teacher data set 200-A3.

[0043] The weather estimation models 300-A1 to 300-A3 are collectively referred to as weather estimation model 300.

[0044] In one example, the weather estimation model 300 is a neural network model. When the weather estimation model 300 is a neural network model, the weather estimation model 300 includes at least an input layer and an output layer. The input layer of the weather estimation model 300 is configured to be able to receive sensor data. The output layer of the weather estimation model 300 outputs weather information indicating the estimated weather when the sensor data is input to the input layer.

[0045] The weather estimation model 300 does not necessarily have to be generated for each area. The weather estimation model 300 may be configured to be capable of outputting weather information indicating the estimated weather in a particular area when information specifying the area is input in addition to the sensor data.

[0046] Furthermore, as long as the weather estimation model 300 is configured to be capable of acquiring weather information indicating estimated weather based on at least sensor data, the configuration example of the weather estimation model 300 is not limited to the above configuration.

[0047] The model generation unit 103 stores the generated weather estimation model 300 in the model storage unit 104 .

[0048] The weather estimation unit 105 acquires weather information indicating the estimated weather in response to a request from the information terminal 4 of the user 5. The weather estimation unit 105 uses a weather estimation model 300 to acquire the weather information indicating the estimated weather.

[0049] The output unit 106 outputs weather information indicating the estimated weather acquired by the weather estimation unit 105 to the information terminal 4 of the user 5 who sent the request. In the information terminal 4, the received weather information may be output as image information on a display device of the information terminal 4, or may be output as audio information from an audio output device of the information terminal 4.

[0050] FIG. 4 is a diagram showing an example of the data structure of a teacher data set 200 for a certain area stored in the data storage unit 102. As shown in FIG.

[0051] The teacher data set 200 includes a large number of pairs of user weather and sensor data collected at different times.

[0052] In the example shown in FIG. 4, the odor sensor 2 is equipped with N sensor elements. Thus, one piece of sensor data includes measurement values ​​Dx1 to DxN by the N sensor elements, where "x" is any of "a" to "d." These measurement values ​​Dx1 to DxN may be values ​​of electrical signals output by the N sensor elements that measure odors, or may be values ​​obtained by correcting these output signals with a reference value. These output signals or corrected values ​​may be values ​​corrected with the temperature or humidity of the environment surrounding the odor sensor elements, or output values ​​of other environmental sensors.

[0053] In this way, the teacher data set 200 includes pairs of correct answer data and sensor data for various weather conditions, such as sensor data when it is sunny, sensor data when it is cloudy, sensor data when it is rainy, etc. Therefore, by using this teacher data set 200, it is possible to generate a weather estimation model 300 that can estimate weather from sensor data.

[0054] Next, the operation of the system using the server 1 will be described in detail.

[0055] 5 is a flowchart showing the operation of generating the weather estimation model 300 in the server 1. Here, a weather estimation model 300 is already stored in the model storage unit 104, and the operation of updating the weather estimation model 300 with a new weather estimation model 300 will be described.

[0056] The collection unit 101 transmits a weather input request to each information terminal 4 (S101). In addition, the collection unit 101 transmits a measurement request requesting measurement of an odor to each odor sensor 2 (S102).

[0057] The order of the process of S101 and the process of S102 is not limited to this. After the process of S101 to S102, the process of S103 to S104 and the process of S105 to S106 are executed in parallel.

[0058] In S103, the collection unit 101 acquires the user answer and the location information of the information terminal 4 from the information terminal 4. The information terminal 4 includes a location sensor. When a user answer is input by the user 5, the information terminal 4 acquires the location information by the location sensor, and attaches the location information to the user answer and transmits it to the server 1. This allows the server 1 to acquire the user answer and the location information of the information terminal 4.

[0059] In S104, the collection unit 101 determines whether acquisition of user responses from all information terminals 4 has been completed or whether a timeout has occurred (S104). A timeout in the process of S104 means that a predetermined time has elapsed since the process of S101.

[0060] If there remains any information terminal 4 that has not yet completed obtaining a user answer and if the timeout has not yet occurred (S104: No), the control transitions to S103.

[0061] In S105, the collection unit 101 acquires sensor data from the odor sensor 2. The odor sensor 2 measures the odor in response to the measurement request, and transmits the sensor data obtained by the measurement to the server 1. In this way, the server 1 can acquire the sensor data.

[0062] In S106, the collection unit 101 determines whether acquisition of sensor data from all odor sensors 2 has been completed or whether a timeout has occurred (S106). A timeout in the process of S106 means that a predetermined time has elapsed since the process of S102.

[0063] If there are still odor sensors 2 that have not yet completed acquiring sensor data and the timeout has not yet occurred (S106: No), control transitions to S105.

[0064] When it is determined that acquisition of user responses from all information terminals 4 has been completed or timed out (S104: Yes) and acquisition of sensor data from all odor sensors 2 has been completed or timed out (S106: Yes), the collection unit 101 selects user responses and sensor data for each district (S107). Then, the collection unit 101 associates the selected user responses with the sensor data and stores them in the data storage unit 102 for each district (S108).

[0065] For example, there are a plurality of users 5 and odor sensors 2a in the area A2 shown in Fig. 1. The collection unit 101 selects, for the area A2, one user answer from among the plurality of user answers acquired from the plurality of users 5, and one sensor data from among the plurality of sensor data acquired from the plurality of odor sensors 2a. Then, the collection unit 101 associates the selected user answer with the sensor data, and adds them to the teacher dataset 200-A2.

[0066] The method of selecting one user answer from among a plurality of user answers is not limited to a specific method.

[0067] For example, the collection unit 101 may select one user response by majority vote. If "sunny" is the most common response among the multiple user responses, the collection unit 101 selects "sunny" as the single user response.

[0068] The majority vote may be a weighted majority vote. For example, a user 5 with a deep knowledge of weather may be given a higher weight than a user 5 with an average level of knowledge about weather.

[0069] Furthermore, the collection unit 101 may execute a weighted majority vote according to the difference between the timing of acquiring the sensor data and the timing of acquiring the user response. When there is a large difference between the timing of acquiring the sensor data and the timing of acquiring the user response, the reliability of the user response as correct answer data is lower than when there is a small difference between the timing of acquiring the sensor data and the timing of acquiring the user response. Therefore, the collection unit 101 increases the weight as the difference between the timing of acquiring the sensor data and the timing of acquiring the user response decreases. This makes it possible to obtain highly reliable correct answer data, which in turn improves the estimation accuracy of the weather estimation model 300 obtained by training.

[0070] Furthermore, the method of selecting one piece of sensor data from the multiple pieces of sensor data acquired from the multiple odor sensors 2a is not limited to a specific method.

[0071] Furthermore, the method by which the server 1 knows the district in which the odor sensor 2a is placed is not limited to a specific method. For example, each odor sensor 2a is given a unique ID. Furthermore, when each odor sensor 2a is placed, location information indicating the placement position is associated with the unique ID and registered in the server 1. When each odor sensor 2 transmits sensor data to the server 1, it also transmits its own unique ID. Based on the unique ID transmitted together with the sensor data, the server 1 can identify the district in which the odor sensor 2 that is the source of the sensor data is placed.

[0072] Each of the user answers selected by the process of S107 and then stored in the data storage unit 102 by the process of S108 is an example of second weather information. Also, when multiple user answers are acquired from multiple users 5 located in one area, the multiple user answers are an example of multiple third weather information.

[0073] Following the process of S108, the model generation unit 103 generates a weather estimation model 300 for each district based on the teacher dataset 200 for each district stored in the data storage unit 102 (S109). The model generation unit 103 updates the weather estimation model 300 for each district stored in the model storage unit 104 with the weather estimation model 300 for each district generated by the process of S108 (S110). Then, the operation of generating the weather estimation model 300 ends.

[0074] The processes of S101 to S110 are executed at different times. The different times may be at a predetermined time interval. The weather estimation model 300 for each area is updated every time the processes of S101 to S110 are executed.

[0075] The processes of S109 and S110 may be executed asynchronously with the processes of S101 to S108. For example, the processes of S109 and S110 may be executed less frequently than the processes of S101 to S108.

[0076] Furthermore, the collection unit 101 may update the teacher dataset 200 when acquiring pairs of user answers and sensor data. Specifically, when the collection unit 101 adds a first number of pairs of user answers and sensor data to a certain teacher dataset 200, the collection unit 101 may delete the oldest first number of pairs of user answers and sensor data included in the teacher dataset 200 from the teacher dataset 200. The first number is 1 or more.

[0077] The above-mentioned update of the teacher dataset 200 may be executed by the model generation unit 103. That is, the model generation unit 103 may use a set of the latest second number of pairs of pairs of user responses and sensor data obtained from a certain district as the teacher dataset 200 for the district. The second number is 2 or more.

[0078] FIG. 6 is a flowchart showing the operation of the server 1 for estimating weather.

[0079] As described above, the user 5 can transmit a weather output request to the server 1 via the information terminal 4. The weather output request can include designation of one or more areas for which the user wishes to know the current weather.

[0080] In the following description, the weather output request includes the designation of one area. If the weather output request includes the designation of multiple areas, steps S202 to S204 are executed for each designation of an area.

[0081] When the server 1 receives a weather output request including a designation of a certain area from the information terminal 4 (S201), the weather estimation unit 105 acquires sensor data from the odor sensor 2a in the designated area (S202). In the explanation of Fig. 6, the area designated by the weather output request in S201 will be referred to as the target area.

[0082] Next, the weather estimation unit 105 inputs the acquired sensor data into the weather estimation model 300 for the target area, thereby acquiring weather information indicating the estimated weather in the target area (S203).

[0083] The output unit 106 outputs the weather information indicating the estimated weather in the target area acquired by the weather estimation unit 105 to the information terminal 4 that transmitted the weather output request (S204). Then, the operation of estimating the weather in the server 1 is completed.

[0084] The weather information output in S204 is an example of first weather information.

[0085] FIG. 7 is a diagram for explaining information transferred between the components during the series of operations shown in FIG.

[0086] First, the server 1 transmits a weather input request to each information terminal 4 (S301). The server 1 also transmits a measurement request to each odor sensor 2a (S302).

[0087] The information terminal 4-1 of the user 5-1 who owns the odor sensor 2b transmits a measurement request to the odor sensor 2b in response to the weather input request (S303), and acquires sensor data from the odor sensor 2b.

[0088] Each information terminal 4 transmits a user answer to the server 1 in response to the current weather input from the user 5 (S304). Note that the information terminal 4-1 transmits the sensor data acquired from the odor sensor 2b together with the user answer.

[0089] Furthermore, each odor sensor 2a transmits sensor data (S305).

[0090] FIG. 8 is a diagram for explaining information transferred between the components during the series of operations shown in FIG.

[0091] When the information terminal 4 of a certain user 5 transmits a weather output request to the server 1 (S401), the server 1 transmits a measurement request to the odor sensor 2a located in the area specified by the weather output request (S402).

[0092] When the server 1 acquires the sensor data from the odor sensor 2a (S403), it acquires weather information indicating the estimated weather, and then outputs the weather information to the information terminal 4 (S404).

[0093] In the above description, the server 1 acquires sensor data to be used for estimating the weather, triggered by the reception of a weather output request. The timing of acquiring sensor data to be used for estimating the weather does not have to correspond to the timing of receiving the weather output request.

[0094] For example, the weather estimation unit 105 periodically acquires sensor data from the odor sensor 2a in each district. Then, when a weather output request is received, the weather estimation unit 105 may use the latest sensor data acquired from the odor sensor 2a in the district specified by the weather output request for estimating the weather.

[0095] Also, it has been described that the server 1 not only estimates the weather but also generates the weather estimation model 300. The server 1 may execute only one of estimating the weather using the weather estimation model 300 and generating the weather estimation model 300.

[0096] FIG. 9 is a diagram illustrating another example of the functional configuration of the server 1 (referred to as server 1a).

[0097] The server 1a includes a collection unit 101, a data storage unit 102, a model generation unit 103, and a model storage unit 104. The collection unit 101 and the model generation unit 103 cooperate to execute the operations shown in FIGS.

[0098] FIG. 10 is a diagram showing yet another example of the functional configuration of the server 1 (referred to as server 1b).

[0099] The server 1b includes a model storage unit 104, a weather estimation unit 105, and an output unit 106. The weather estimation unit 105 and the output unit 106 cooperate to execute the operations shown in FIGS.

[0100] In this way, the function of generating the weather estimation model 300 and the function of estimating the weather using the weather estimation model 300 may be implemented in different information processing devices.

[0101] As described above, according to the embodiment, the server 1 as an information processing device is equipped with a processor 10 that acquires sensor data, which is a value from the odor sensor 2a, and outputs weather information indicating the weather corresponding to the sensor data in response to a weather output request from the information terminal 4 of the user 5.

[0102] Therefore, it becomes possible to estimate the weather based on odors that are directly related to the weather, making it possible to estimate the weather with greater accuracy than methods that identify the weather based on information from raincloud radar or meteorological observation data.

[0103] Furthermore, according to the embodiment, the memory 12 stores a weather estimation model 300 configured to be able to estimate the current weather based on at least the sensor data from the odor sensor 2a. The processor 10 acquires weather information indicating the weather corresponding to the sensor data from the odor sensor 2a based on the sensor data and the weather estimation model 300.

[0104] Therefore, it is possible to estimate the weather with greater accuracy than with methods that identify the weather based on raincloud radar information or meteorological observation data.

[0105] Furthermore, according to the embodiment, the weather estimation model 300 (more precisely, a group of weather estimation models 300-A1 to 300-A3) is configured to be able to estimate the current weather in each of a plurality of regions based on sensor data from the odor sensor 2a in each of the plurality of regions. The weather output request includes a designation of the region. The processor 10 acquires sensor data from the odor sensor 2a in the designated region, and estimates the current weather in the designated region based on the sensor data.

[0106] This allows the user 5 to know the current weather in a remote location.

[0107] Furthermore, according to the embodiment, the processor 10 acquires sensor data from the odor sensors 2a in each of the multiple districts, and acquires weather information indicating the current weather from the information terminal 4 of the user 5 located in each of the multiple districts. The processor 10 generates a weather estimation model 300 based on the sensor data acquired from the odor sensors 2a in each of the multiple districts and the weather information indicating the current weather from the information terminal 4 of the user 5 located in each of the multiple districts.

[0108] Therefore, it is possible to estimate the current weather for each area.

[0109] Furthermore, according to the embodiment, when the processor 10 acquires a plurality of user answers from the information terminals 4 of a plurality of users 5 located in one area, the processor 10 selects the weather as the correct answer data from the plurality of user answers by majority vote.

[0110] The processor 10 may select the weather as the correct answer data by a weighted majority vote in which a weight is applied to a user answer from the information terminal 4 of a particular user 5, for example, a user 5 who has a deep knowledge of weather.

[0111] Furthermore, the processor 10 may select the weather as the correct answer data by a weighted majority vote according to the difference between the timing of acquiring the user answer and the timing of acquiring the sensor data.

[0112] Furthermore, according to the embodiment, the server 1a as an information processing device includes a processor 10 that acquires sensor data from an odor sensor 2 in a district, acquires user responses indicating the current weather from an information terminal 4 of a user 5 located in the district, and generates a weather estimation model 300 configured to be able to estimate the current weather based on the acquired sensor data and user responses. The weather estimation model 300 estimates the current weather based on measurements by the odor sensor 2.

[0113] Therefore, it becomes possible to estimate the weather based on odors that are directly related to the weather, making it possible to estimate the weather with greater accuracy than methods that identify the weather based on information from raincloud radar or meteorological observation data.

[0114] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0115] 1, 1a, 1b Server, 2, 2a, 2a-1 to 2a-6, 2b Odor sensor, 3 Network, 4, 4-1 to 4-6 Information terminal, 5, 5-1 to 5-6 User, 10 Processor, 11 Network interface, 12 Memory, 100 Processing program, 101 Collection unit, 102 Data storage unit, 103 Model generation unit, 104 Model storage unit, 105 Weather estimation unit, 106 Output unit, 200, 200-A1 to 200-A3 Teacher dataset, 300, 300-A1 to 300-A3 Weather estimation model, A1 to A3 Area.

Claims

1. a processor that acquires a first measurement value by the odor sensor and outputs first weather information indicating the weather corresponding to the first measurement value in response to an output request transmitted from a user's information terminal; An information processing device comprising:

2. A memory in which a first model is stored, the memory storing the first model configured to be able to estimate weather based on a first measurement value; the processor outputs the first weather information based on the first measurement and the first model. The information processing device according to claim 1 .

3. the first model is configured to be capable of estimating weather in each of a plurality of districts based on a measurement value obtained by an odor sensor disposed in each of the plurality of districts; the output request includes designation of one or more districts from the plurality of districts; the first measurement value is a measurement value obtained by an odor sensor disposed in the specified area, The first weather information indicates the weather in the specified area. The information processing device according to claim 2 .

4. The processor, acquiring second measurement values ​​by the odor sensors arranged in the plurality of districts; acquiring second weather information indicating weather from information terminals of users located in each of the plurality of districts; generating the first model based on the second measurements and the second weather information; The information processing device according to claim 3 .

5. when the processor acquires a plurality of pieces of third weather information from information terminals of a plurality of users located in a first district that is one of the plurality of districts, the processor selects the second weather information from the plurality of pieces of third weather information by majority vote; The information processing device according to claim 4.

6. the processor selects the second weather information by a weighted majority vote in which a weight is assigned to third weather information from an information terminal of a specific user among the plurality of third weather information; The information processing device according to claim 5 .

7. the processor selects the second weather information by a weighted majority vote according to a difference between a timing at which the third weather information is acquired and a timing at which the second measurement value is acquired. The information processing device according to claim 5 .

8. Memory, a processor that acquires a first measurement value from an odor sensor disposed in a district, acquires weather information indicating the weather from an information terminal of a user located in the district, generates a model based on the first measurement value and the weather information, configured to be able to estimate the weather based on the first measurement value, and stores the model in the memory; An information processing device comprising:

9. acquiring a first measurement value from an odor sensor disposed in the area; acquiring weather information indicating weather from an information terminal of a user located in the area; generating a model based on the first measurement value and the weather information, the model being configured to be able to estimate the weather based on the first measurement value; An information processing method comprising:

Images