JP7024187B2 - Learning device and attribute estimation device - Google Patents

Description

Embodiments of the present invention relate to a learning device, an attribute estimation device, a learning method, an attribute estimation method, and a program.

Consider developing a business that provides new services and providing it to customers. When developing a business, it is necessary to consider the content of services according to the needs of consumers based on the attribute information of consumers such as the family structure, lifestyle, and lifestyle of consumers such as customers. Since attribute information such as lifestyles often differs for each consumer, it is difficult to correctly grasp them by data analysis or the like.
Therefore, it is performed to grasp the attribute information of the consumer by using the electric power consumption data of the consumer, which is relatively directly connected to the life of the consumer. Specifically, hypothesis verification is performed on power consumption data based on empirical rules each time, and consumers are categorized by numerical analysis methods such as clustering.

Regarding the technique for analyzing data, a technique for specifying a process item as a factor causing fluctuation in the distribution of characteristic values for a product having a spatial spread is known (see, for example, Patent Document 1). According to this technique, based on the characteristic distribution data of a plurality of products, a plurality of components representing the characteristics of the fluctuation of the characteristic distribution data are extracted by the pattern classification method, and for each of the characteristic distribution data, Each of the weighting factors that contributes to the representation of the characteristic distribution data of the plurality of components is calculated. Then, based on the calculated weighting coefficient, at least one component to be analyzed is selected from the plurality of components. Then, the process item as a fluctuation occurrence factor in the characteristic distribution data is extracted based on the weighting coefficient for the selected component and the manufacturing data acquired for each process item.

Japanese Unexamined Patent Publication No. 2012-63928

The above-mentioned technology has the following problems.
When considering the content of services according to the needs of consumers, it is necessary to repeat and redo a wide variety of analyzes according to the needs of consumers.
In addition, when conducting hypothesis verification based on empirical rules for power consumption data each time, it is necessary to formulate a hypothesis and verify the hypothesis each time. Hypothesis planning, etc. largely depends on the experience and skills of the analyst. In addition, the applications required to provide services basically need to be developed individually.
The present invention has been made in view of the above points, and an object of the present invention is to provide home attribute information based on information indicating a household power consumption.

One aspect of the present invention is an acquisition unit that acquires information indicating the power consumption of each of the plurality of households and attribute information of the household from which the power consumption is obtained, and the plurality of households acquired by the acquisition unit. A learning unit that machine-learns information indicating each electric power consumption and associates the characteristic amount of the electric power consumption obtained by the machine learning with the attribute information of the household from which the electric power consumption is obtained, and the above-mentioned A storage unit that stores the characteristic amount of the power consumption associated with the learning unit in association with the attribute information of the household from which the power consumption is obtained, and the power usage of each of the plurality of households acquired by the acquisition unit. Among the information indicating the amount, one or more power consumptions for each use indicating the usage amount for each use of electric power are acquired from each of the power consumptions of households having similar attribute information, and one or more of the above-mentioned uses obtained. It is equipped with a creation unit that newly creates information indicating the amount of power consumption based on another amount of power consumption, and the information indicating the amount of power consumption is a diagram of the magnitude of the amount of power consumption in chronological order. There, the learning unit machine-learns the information indicating the power consumption newly created by the creation unit, and machine-learns the information indicating the power consumption newly created. It is a learning device that associates the attribute information similar to the feature amount of .
In the learning device of one aspect of the present invention, the information indicating the power consumption amount acquired by the acquisition unit represents the magnitude of the power consumption amount in different colors.
In the learning device of one aspect of the present invention, the information indicating the power consumption acquired by the acquisition unit further expresses the concept of time by different colors.
In the learning device of one aspect of the present invention, the information indicating the power consumption acquired by the acquisition unit includes the power consumption and the information indicating the environmental condition measured in the home where the power consumption is obtained. It is expressed by the difference in color .
One aspect of the present invention is an acquisition unit that acquires information indicating the power consumption of a household whose attribute information is unknown, and the power usage by machine learning the information indicating the power consumption acquired by the acquisition unit. The characteristic amount of the electric power consumption and the electric power consumption amount obtained by machine learning the attribute information related to the characteristic amount and the information indicating the electric power consumption amount of each of a plurality of households are obtained. It is provided with an estimation unit that estimates from the information associated with the obtained home attribute information, and the information indicating the power consumption is a graphic representation of the magnitude of the power consumption, and the estimation unit. Acquires the attribute information related to the characteristic amount of the electric power consumption from each of the electric power consumptions of households having similar attribute information among the information indicating the electric power consumption of each of the plurality of households. The above, which is similar to the characteristic amount of the electric power consumption obtained by machine learning the information indicating the electric power consumption newly created based on the electric power consumption according to one or more applications which indicates the electric power consumption according to the use of electric power. It is an attribute estimation device that estimates from information associated with attribute information .
In the attribute estimation device of one aspect of the present invention, the information indicating the power consumption of a household whose attribute information is unknown acquired by the acquisition unit represents the magnitude of the power consumption by different colors.
In the attribute estimation device of one aspect of the present invention, the information indicating the power consumption of a household whose attribute information is unknown acquired by the acquisition unit further expresses the concept of time by different colors.
In the attribute estimation device of one aspect of the present invention, the information indicating the power consumption of a household whose attribute information is unknown acquired by the acquisition unit is measured in the power consumption amount and the household in which the power consumption amount is obtained. The information indicating the environmental conditions to be measured is represented by different colors .

According to the embodiment of the present invention, it is possible to provide home attribute information based on information indicating the amount of power used by the home.

It is a figure which shows an example (the 1) of the information which shows the electric power consumption. It is a figure which shows an example (the 2) of the information which shows the electric power consumption. It is a figure which shows an example of the hardware composition of the learning apparatus which concerns on embodiment. It is a figure which shows an example of the attribute information estimation table. It is a functional block diagram of the learning apparatus which concerns on embodiment. It is a figure which shows an example of the heat map of the information which shows the electric power consumption of each of a plurality of households. It is a flowchart which shows an example of the operation of a learning device. It is a functional block diagram of the learning apparatus which concerns on a modification. It is a figure which shows the example of making the information which shows the electric power consumption. It is a flowchart which shows an example of the operation of the learning apparatus which concerns on a modification. It is a figure which shows an example of the hardware composition of the attribute estimation apparatus which concerns on embodiment. It is a functional block diagram of the attribute estimation apparatus which concerns on embodiment. It is a flowchart which shows an example of the operation of the attribute estimation device. It is a figure which shows an example of the operation of the attribute estimation system which concerns on embodiment. It is a figure which shows an example (the 1) of the estimation accuracy of the attribute information estimated by the attribute estimation apparatus which concerns on embodiment. It is a figure which shows an example (the 2) of the estimation accuracy of the attribute information estimated by the attribute estimation apparatus which concerns on embodiment. It is a figure which shows an example (the 3) of the estimation accuracy of the attribute information estimated by the attribute estimation apparatus which concerns on embodiment. It is a figure which shows an example of the heat map of the information which shows the electric power consumption of each of a plurality of households. It is a figure which shows an example (the 4) of the estimation accuracy of the attribute information estimated by the attribute estimation apparatus which concerns on embodiment.

Next, a mode for carrying out the present invention will be described with reference to the drawings. The embodiments described below are merely examples, and the embodiments to which the present invention is applied are not limited to the following embodiments.
In all the drawings for explaining the embodiment, the same reference numerals are used for those having the same function, and the repeated description will be omitted.

(First Embodiment)
(Learning device)
The learning device according to the embodiment will be described. The learning device according to the embodiment acquires information indicating the power consumption of each of the plurality of households. Here, the information indicating the amount of electric power used represents the magnitude of the amount of electric power used such as a heat map by different colors.
Further, the learning device according to the embodiment acquires the attribute information of the household from which the information indicating the power consumption of each of the acquired plurality of households is obtained. The learning device according to the embodiment machine-learns information indicating the electric power consumption of each of a plurality of households, and the characteristic amount of the electric power consumption obtained by machine learning and the attribute information of the household from which the electric power consumption is obtained. To associate with. The learning device according to the embodiment stores the feature amount of the electric power consumption and the attribute information of the household from which the electric power consumption is obtained in association with each other.

FIG. 1 is a diagram showing an example (No. 1) of information indicating the amount of electric power used. In FIG. 1, the horizontal axis represents time and the vertical axis represents date and time. FIG. 1 shows a heat map of information showing the amount of power used acquired from HEMS (Home Energy Management System) installed in each of a plurality of buildings. In the example shown in FIG. 1, the magnitude of the power consumption is visualized in chronological order.
According to FIG. 1, it can be seen that the amount of electric power used increases between 6:00 and 18:00 and after 18:00. In addition, it can be seen that the amount of electricity used is low during the Golden Week period displayed as "GW". As shown in FIG. 1, the amount of electric power used can be visualized by plotting the magnitude of the amount of electric power used in chronological order. Therefore, it is possible to intuitively grasp the life dynamics.

FIG. 2 is a diagram showing an example (No. 2) of information indicating the amount of electric power used. In FIG. 2, the horizontal axis shows the outside air temperature, and the vertical axis shows the power consumption.
According to FIG. 2, it can be seen that the power consumption increases as the outside air temperature decreases, while the power consumption increases as the outside air temperature rises. By referring to the information indicating the amount of electric power used shown in FIG. 2, it is possible to estimate the performance of air-conditioning equipment such as heating equipment, the heat insulating property of a house, and the like.

(Configuration of learning device)
FIG. 3 is a diagram showing an example of the hardware configuration of the learning device according to the embodiment.
The learning device 100 displays a CPU (Central Processing Unit) 102, a ROM (Read Only Memory) 104, a RAM (Random Access Memory) 106, a non-volatile memory 108, a communication I / F 110, and an operation unit 112. A unit 114 and a bus 116 are provided.
The CPU 102 reads a program 1082 or the like that controls the operation of the learning device 100 from the non-volatile memory 108, expands it into the RAM 106, and executes it. The non-volatile memory 108 is composed of a flash memory, an HDD (Hard Disk Drive), an SSD (Solid State Drive), an SD (Secure Digital) card, and the like. The non-volatile memory 108 stores the program 1082 executed by the CPU 102, the power usage information 1084, the attribute information 1086, and the attribute information estimation table 1088. The electric power consumption information 1084 is information indicating the electric power consumption of each of a plurality of households, and represents the magnitude of the electric power consumption such as a heat map by different colors. Attribute information 1086 is information indicating household attributes such as wake-up time, number of family members, housing style, presence / absence of elderly people, presence / absence of infants, presence / absence of pets, absence time, and vacation acquisition information. The information indicating the electric power consumption of each of the plurality of households included in the electric power usage information 1084 and the information indicating the attributes of the household included in the attribute information 1086 are associated with each other. The attribute information estimation table 1088 will be described later.

The communication I / F 110 is composed of a communication module such as an NFC (Nearfield radio communication) module and a wireless LAN module. Further, the communication I / F 110 may be configured by USB (Universal General Bus). The communication I / F 110 communicates with an external device by NFC or wireless LAN. The communication I / F 110 is connected to a smart meter (not shown) and an attribute estimation device 300 described later.
The operation unit 112 is an input device that accepts user operations, and includes a pointing device such as a touch panel, buttons, dials, touch sensors, touch pads, and the like.
The display unit 114 is configured by, for example, a liquid crystal display or the like, and displays information indicating the amount of power used, attribute information of the household from which the amount of power used has been obtained, and the like.

(Attribute information estimation table)
FIG. 4 is a diagram showing an example of an attribute information estimation table. The attribute information estimation table 1088 is data in a table format in which the feature amount of the power consumption and the attribute information are associated with each other. The characteristic amount of electric power consumption is obtained by machine learning such as deep learning with information indicating the electric power consumption of each of a plurality of households. The attribute information is the attribute information of the household from which the power consumption is obtained. In the example shown in FIG. 4, the characteristic amount “AAA” of the electric power consumption and the attribute information “aaaa” of the household from which the electric power consumption is obtained are associated with each other.

(Functional configuration of learning device)
FIG. 5 is a functional block diagram of the learning device according to the embodiment.
The learning device 100 functions as an acquisition unit 152, a learning unit 154, and an output unit 156 by the CPU 102 executing the program 1082 expanded on the RAM 106 from the non-volatile memory 108.

(Each function configuration of the learning device)
Each functional configuration of the learning apparatus 100 will be described in detail with reference to FIGS. 3 and 5. In the following, in explaining each function of the learning device 100, the relationship with the main hardware for realizing each functional block of the learning device 100 shown in FIG. 5 will also be described.
The acquisition unit 152 is realized by a command from the CPU 102 and a communication I / F 110. The acquisition unit 152 acquires information indicating the amount of power used transmitted by smart meters installed in each of the plurality of households. Further, the acquisition unit 152 acquires the attribute information of the household from which the information indicating the amount of electric power used is obtained. When the acquisition unit 152 acquires the information indicating the power consumption of each of the plurality of households and the attribute information of the household from which the power consumption is obtained, the acquisition unit 152 obtains the information indicating the power consumption of each of the plurality of households in the non-volatile memory. It is output to the power consumption information 1084 of 108, and the attribute information of the home from which the power usage is obtained is output to the attribute information 1086 of the non-volatile memory 108.

The learning unit 154 is realized by an instruction from the CPU 102. The learning unit 154 acquires information indicating the power usage of each of the plurality of households from the power usage information 1084 of the non-volatile memory 108, and acquires the attribute information of the household from which the power usage is obtained from the attribute information 1086. .. When the learning unit 154 acquires the information indicating the electric power consumption of each of the plurality of households and the attribute information of the household from which the electric power consumption is obtained, the learning unit 154 obtains the information indicating the electric power consumption of each of the plurality of households by deep learning or the like. Do machine learning.
The learning unit 154 transfers the feature amount of the electric power consumption obtained by deep learning the information indicating the electric power consumption of each of the plurality of households and the attribute information of the household from which the electric power consumption is obtained to the output unit 156. Output.

The output unit 156 is realized by an instruction from the CPU 102. When the output unit 156 acquires the characteristic amount of the electric power consumption output by the learning unit 154 and the attribute information of the household from which the electric power consumption amount is obtained, the characteristic amount of the acquired electric power consumption amount and the electric power consumption amount are obtained. It is stored in the attribute information estimation table 1088 of the non-volatile memory 108 in association with the home attribute information.

(Example of information indicating power consumption)
FIG. 6 is a diagram showing an example of a heat map of information showing the power consumption of each of a plurality of households. Information indicating the power consumption of each of multiple households is represented by a heat map. The heat map includes a heat map 10 and a heat map 20.
The heat map 10 is information showing the power consumption of each of the plurality of households from January to June. The heat map 20 is information showing the power consumption of each of the plurality of households from July to December. In FIG. 6, the horizontal axis indicates time from 0:00 to 24:00, and the vertical axis indicates the month. That is, in an example of the heat map of the information showing the power consumption shown in FIG. 6, the magnitude of the power consumption is represented by the difference in color, and the concept of time is represented by the difference in color.

(Operation of learning device)
FIG. 7 is a flowchart showing an example of the operation of the learning device.
(Step S702) The acquisition unit 152 of the learning device 100 acquires information indicating the amount of power used transmitted by smart meters installed in each of the plurality of homes. When the acquisition unit 152 acquires the information indicating the power consumption of each of the plurality of households, the acquisition unit 152 outputs the information indicating the power consumption to the power usage information 1084 of the non-volatile memory 108. The power consumption information 1084 of the non-volatile memory 108 stores information indicating the power usage of each of the plurality of households output by the acquisition unit 152.
(Step S704) When the acquisition unit 152 of the learning device 100 acquires the information indicating the electric power consumption of each of the plurality of households, the acquisition unit 152 acquires the attribute information of the household from which the information indicating the electric power consumption is obtained. When the acquisition unit 152 acquires the attribute information from which the information indicating the power consumption is obtained, the acquisition unit 152 outputs the attribute information to the attribute information 1086 of the non-volatile memory 108. The attribute information 1086 of the non-volatile memory 108 stores the attribute information output by the acquisition unit 152.

(Step S706) The learning unit 154 of the learning device 100 acquires information indicating the power usage of each of the plurality of households from the power usage information 1084 of the non-volatile memory 108, and obtains the power usage from the attribute information 1086. Acquire the attribute information of the home. When the learning unit 154 acquires the information indicating the power consumption of each of the plurality of households and the attribute information obtained from the power consumption, the learning unit 154 performs deep learning for the information indicating the power consumption of each of the plurality of households. conduct.
(Step S708) When the learning unit 154 of the learning device 100 performs deep learning on the information indicating the power consumption of each of the plurality of households, the feature amount and the power of the power consumption obtained by performing the deep learning. Associate with the attribute information of the household from which the usage was obtained. When the learning unit 154 associates the characteristic amount of the electric power consumption with the attribute information of the household from which the electric power consumption is obtained, the learning unit 154 associates the characteristic amount of the electric power consumption with the attribute information of the household from which the electric power consumption is obtained. Output to the output unit 156.
(Step S710) When the output unit 156 of the learning device 100 acquires the characteristic amount of the power consumption output by the learning unit 154 and the information associated with the attribute information of the household from which the power consumption amount is obtained, the power consumption amount is obtained. The feature amount and the attribute information of the household from which the power consumption is obtained are associated with each other and output to the attribute information estimation table 1088 of the non-volatile memory 108. The attribute information estimation table 1088 of the non-volatile memory 108 stores the feature amount of the power consumption output by the output unit 156 in association with the attribute information of the household from which the power usage amount is obtained.
In the flowchart shown in FIG. 7, the order of steps S702 and S704 may be exchanged.

In the above-described embodiment, the case where the learning device 100 acquires the information indicating the power consumption transmitted by the smart meters installed in each of the plurality of homes has been described, but the present invention is not limited to this example. For example, when the information indicating the power consumption output from the smart meter is not a heat map, the information indicating the power consumption output from the smart meter may be output to a conversion device that converts the information to a heat map. When the conversion device acquires the information indicating the power consumption output by the smart meter, the conversion device converts the acquired information indicating the power consumption amount into a heat map. When the conversion device converts the information indicating the power consumption into the heat map, the conversion device may output the information indicating the heat map to the learning device 100.

In the above-described embodiment, the case where the learning device 100 includes the non-volatile memory 108 in which the attribute information estimation table 1088 is stored has been described, but the present invention is not limited to this example. For example, the attribute information estimation table 1088 may be stored in the cloud. In this case, the learning device 100 stores the feature amount of the electric power consumption and the attribute information of the household from which the electric power consumption is obtained in association with each other in the attribute information estimation table 1088 on the cloud.
In the above-described embodiment, the case where the learning device 100 acquires the information indicating the power consumption transmitted by the smart meters installed in each of the plurality of homes has been described, but the present invention is not limited to this example. For example, even if the learning device 100 acquires the information indicating the electric power consumption in which the electric power consumption and the information indicating the environmental condition measured in the home where the electric power consumption is obtained are expressed by different colors. good. Here, the information indicating the environmental condition measured at home includes information indicating the outside air temperature, information indicating the temperature of the living room, information indicating the temperature difference between the living room and the outside air, information indicating the humidity, and the like. Specifically, the amount of electric power used and the information indicating the environmental condition measured in the household from which the amount of electric power is used are represented by a heat map. Then, the learning device 100 may acquire the attribute information of the household from which the information indicating the power consumption of each of the acquired plurality of households has been obtained. The learning device machine-learns information indicating the power consumption of each of a plurality of households, and associates the feature amount of the power consumption obtained by machine learning with the attribute information of the household from which the power usage is obtained. You may do it. The learning device may store the feature amount of the electric power consumption and the attribute information of the household from which the electric power consumption is obtained in association with each other.

According to the learning device 100 according to the present embodiment, the feature amount of the power consumption amount obtained by deep learning the information indicating the power consumption amount in the heat map format is associated with the attribute information obtained from the power consumption amount. Remember. Since the heat map is "visualized", it is possible to intuitively grasp the life dynamics. In addition, by associating the characteristic amount of the electric power consumption obtained by deep learning the information indicating the electric power consumption in the heat map format with the attribute information of the household from which the electric power consumption is obtained, the characteristic amount of the electric power consumption Therefore, attribute information such as owned equipment, housing performance, and living dynamics can be estimated with high accuracy. Machine learning automatically extracts common features between information that indicates power usage, so at a deeper level than when an analyst analyzes information that indicates power usage. Analysis is possible. Further, by performing machine learning, it is possible to estimate any kind of attribute information by the same method.

(Modification example)
A learning device according to a modified example will be described. In the learning device 200 according to the modified example, among the information indicating the power consumption of each of the plurality of households stored in the power consumption information 1084 of the non-volatile memory 108, the attribute information from which the power consumption is obtained is the same. Or extract information indicating the amount of electricity used by similar households. The learning device 200 obtains the power consumption of the air conditioner, the power consumption of the water heater, the power consumption of the kitchen, and the power consumption of the kitchen from each of the information indicating the power consumption of the household whose attribute information is the same or similar. Acquire the power consumption of the air conditioner, the power consumption of the water heater, and the power consumption other than the power consumption of the kitchen (other power consumption). In the following, each of the power consumption of air conditioners, water heaters, kitchens, and other power consumption is referred to as power usage by use in order to show the usage of power by use. Call. That is, the learning device 200 acquires one or a plurality of usage-specific power consumptions from each of the information indicating the power consumptions of households whose extracted attribute information is the same or similar.

The learning device 200 newly creates information indicating the power consumption amount from the acquired power consumption amount for each use. The learning device 200 outputs the newly created information indicating the power usage amount to the power usage amount information 1084 of the non-volatile memory 108. Further, the learning device 200 stores similar household attribute information in the attribute information 1086 of the non-volatile memory 108 in association with the newly created information indicating the power consumption.
The learning device 200 deep-learns information indicating the newly created power consumption. When the learning device 200 deep-learns the newly created information indicating the power consumption, the non-volatile memory 108 associates the feature amount of the power consumption obtained by the deep learning with the attribute information of a similar household. It is stored in the attribute information estimation table 1088 of.

FIG. 2 can be applied to an example of the hardware configuration of the learning device 200 according to the modified example.
<Functional configuration of learning device>
FIG. 8 is a functional block diagram of the learning device according to the modified example.
The learning device 200 functions as an acquisition unit 252, a learning unit 254, an output unit 256, and a creation unit 258 by executing the program 1082 expanded on the RAM 106 from the non-volatile memory 108 by the CPU 102.

<Each functional configuration of the learning device>
Each functional configuration of the learning apparatus 200 will be described in detail with reference to FIGS. 2 and 8. In the following, in explaining each function of the learning device 200, the relationship with the main hardware for realizing each functional block of the learning device 200 shown in FIG. 8 will also be described.
The acquisition unit 252, the learning unit 254, and the output unit 256 can apply the acquisition unit 152, the learning unit 254, and the output unit 256 of the learning device 100, respectively.

The creation unit 268 is realized by a command from the CPU 102 and an operation unit 112. The creation unit 268 functions when an operation for increasing the information indicating the amount of power used is performed on the operation unit 112. When the creation unit 268 performs an operation to increase the information indicating the power consumption amount to the operation unit 112, the power consumption amount of each of the plurality of households stored in the power usage amount information 1084 of the non-volatile memory 108 is performed. From the information indicating the above, the information indicating the power consumption of households having the same or similar attribute information is extracted. Specifically, the preparation unit 268 extracts information indicating the power consumption of each of a plurality of households having the same or similar number of households, living style, floor area, age, and the like.
The creation unit 268 acquires one or a plurality of usage-specific power usages from each of the extracted information indicating the power usage. Specifically, the creation unit 268 uses each of the information indicating the electric power consumption, the pattern of the information indicating the electric power consumption of the air heater, the pattern of the information indicating the electric power consumption of the water heater, and the electric power of the kitchen. Acquires a pattern of information indicating usage and a pattern of information indicating other power usage.

The creation unit 268 newly creates information indicating the power consumption of households having similar attributes by synthesizing the acquired power usage for each use in a sequence combination.
The creation unit 268 stores the newly created information indicating the power consumption amount in the power consumption amount information 1084 of the non-volatile memory 108. Further, the creation unit 268 stores similar household attribute information in the attribute information 1086 of the non-volatile memory 108 in association with the newly created information indicating the power consumption.

(Example of heat map creation)
As described above, the creation unit 268 acquires one or a plurality of usage-specific power consumptions from each of the information indicating the power consumptions of households having the same or similar attribute information. Then, the creation unit 268 newly creates information indicating the power consumption of households having similar attributes by synthesizing the acquired power usage for each use in a sequence combination.
FIG. 9 is a diagram showing an example of creating information indicating the amount of electric power used. In the example shown in FIG. 9, it is assumed that there are two households (A and B) having the same or similar attribute information. The creation unit 268 acquires the power consumption for each use from the information A indicating the power consumption. In addition, the creation unit 268 acquires the power consumption for each use from the information B indicating the power consumption.
The creation unit 268 rounds up the power consumption for each use obtained from the information A indicating the power consumption and the power usage for each use obtained from the information B indicating the power usage. By combining and synthesizing, information ABA indicating power consumption, information AAB indicating power consumption, information ABB indicating power consumption, information BAB indicating power consumption, information BAA indicating power consumption, and power consumption Information BBA indicating the above is newly created. That is, 2 (other power consumption) x 2 (air conditioner power consumption) x 2 (water heater power consumption) x 2 (kitchen power consumption) = 8 pieces, and the creation unit 268 Information indicating the amount of power used for the six pieces can be newly created.

(Operation of learning device)
FIG. 10 is a flowchart showing an example of the operation of the learning device according to the modified example. FIG. 10 mainly shows the operation of the creating unit 268 of the learning device 200.
(Step S1002) The creating unit 268 of the learning device 200 obtained the power consumption among the information indicating the power usage of each of the plurality of households stored in the power usage information 1084 of the non-volatile memory 108. Extract information indicating the amount of electricity used that has the same or similar household attribute information.
(Step S1004) The creating unit 268 of the learning device 200 acquires information indicating the electric power consumption of the air conditioner from each of the information indicating the electric power consumption of the household having the same or similar attribute information.
(Step S1006) The creating unit 268 of the learning device 200 acquires information indicating the power consumption of the water heater from each of the information indicating the power consumption of the household having the same or similar attribute information.

(Step S1008) The creating unit 268 of the learning device 200 acquires information indicating the electric power consumption of the kitchen from each of the information indicating the electric power consumption of the household having the same or similar attribute information.
(Step S1010) The creating unit 268 of the learning device 200 acquires other information indicating the power consumption from each of the information indicating the power consumption of the household having the same or similar attribute information.
(Step S1012) The creating unit 268 of the learning device 200 has acquired information indicating the power consumption of the air heater, information indicating the power consumption of the water heater, information indicating the power consumption of the kitchen, and other power consumption. By synthesizing the information indicating the power consumption in a sequential combination, the information indicating the power consumption is newly created.
(Step S1014) The creating unit 268 of the learning device 200 outputs the newly created information indicating the power usage amount to the power usage amount information 1084 of the non-volatile memory 108. Further, the learning device 200 stores similar household attribute information in the attribute information 1086 of the non-volatile memory 108 in association with the newly created information indicating the power consumption.

In the above-mentioned modification, the power consumption of the air conditioner, the power consumption of the water heater, the power consumption of the kitchen, and the like are obtained from each of the information indicating the power consumption of the household having the same or similar attribute information. The case of acquiring the power consumption of the above has been described, but the present invention is not limited to this example. For example, from each of the information indicating the power consumption of households with the same or similar attribute information, the power consumption of the air conditioner other than the power consumption of the kitchen, the power consumption of the water heater, and other power consumption. It may be possible to acquire the power consumption of the living room and the power consumption of the entrance.
In the above-mentioned modification, the case where it is assumed that there are two households (A and B) having the same or similar attribute information has been described, but the present invention is not limited to this example. For example, it can be applied to the case where it is assumed that there are three households (A, B, C) having the same or similar attribute information. In this case, the creation unit 268 acquires three power usage amounts for each application (power consumption amount of the air conditioner, power consumption amount of the water heater, and other power consumption amounts) from the information A indicating the power consumption amount. In addition, the creation unit 268 acquires three power usage amounts for each application (power consumption amount of the air conditioner, power consumption amount of the water heater, and other power consumption amounts) from the information B indicating the power consumption amount. In addition, the creation unit 268 acquires three power usage amounts for each application (power consumption amount of the air conditioner, power consumption amount of the water heater, and other power consumption amounts) from the information C indicating the power consumption amount.
The creation unit 268 shows the three usage-specific power usages acquired from the information A indicating the power usage, the three usage-specific power usages acquired from the information B indicating the power usage, and the power usage. By combining and synthesizing the three power consumption amounts for each purpose acquired from the information C in a round-robin manner, information indicating the power consumption amount of 27 pieces is newly created. That is, 3 (power consumption of the air conditioner) x 3 (power usage of the water heater) x 3 (other power usage) = 27 pieces, and the creation unit 268 provides information indicating the power consumption of 24 pieces. It can be newly created.

According to the learning device 200 according to the modified example, when an operation for increasing the information indicating the power consumption is performed, new power is newly generated from each of the information indicating the power consumption of the household having the same or similar attribute information. Information indicating usage can be created. That is, even when the number of information indicating the amount of power used obtained from the smart meter is small, the information indicating the amount of power used can be increased. Since the information indicating the power consumption can be increased, the number of features of the power consumption can also be increased. Therefore, it is possible to increase the number of attribute information that can be estimated from the feature amount of the power consumption.

(Second embodiment)
(Attribute estimation device)
The attribute estimation device according to the embodiment will be described. The attribute estimation device 300 according to the embodiment acquires information indicating the power consumption of a household whose attribute information is unknown. Here, the information indicating the power consumption of a household whose attribute information is unknown represents the magnitude of the power consumption such as a heat map by different colors.
Further, the attribute estimation device 300 according to the embodiment machine-learns the information indicating the power consumption of the household whose attribute information is unknown, thereby determining the feature amount of the power consumption of the household whose attribute information is unknown. Ask. The attribute estimation device 300 according to the embodiment acquires information in which the feature amount of the electric power consumption and the attribute information of the household from which the electric power consumption is obtained are associated with each other from the learning device 100 described above. The attribute estimation device 300 according to the embodiment is based on the information associating the acquired feature amount of the power consumption with the attribute information of the household from which the power consumption is obtained, and the power consumption of the household whose attribute information is unknown. Estimate the attribute information related to the feature quantity of.

(Configuration of attribute estimation device)
FIG. 11 is a diagram showing an example of the hardware configuration of the attribute estimation device according to the embodiment.
The attribute estimation device 300 includes a CPU 302, a ROM 304, a RAM 306, a non-volatile memory 308, a communication I / F 310, an operation unit 312, a display unit 314, and a bus 316.
The CPU 302 reads a program 3082 or the like that controls the operation of the attribute estimation device 300 from the non-volatile memory 308, expands the program into the RAM 306, and executes the program. The non-volatile memory 308 is composed of a flash memory, an HDD, an SSD, an SD card and the like. The non-volatile memory 308 stores the program 3082 executed by the CPU 302 and the attribute information estimation table 3088. The attribute information estimation table 1088 described above can be applied to the attribute information estimation table 3088.
The communication I / F 310 is composed of a communication module such as an NFC module and a wireless LAN module. Further, the communication I / F 310 may be configured by USB. The communication I / F 310 communicates with an external device by NFC or wireless LAN. The communication I / F 310 is connected to a smart meter (not shown) and an attribute estimation device 300 described later.
The operation unit 312 is an input device that accepts user operations, and includes a pointing device such as a touch panel, buttons, dials, touch sensors, touch pads, and the like.
The display unit 314 is configured by, for example, a liquid crystal display or the like, and displays the attribute information of the home from which the power consumption has been obtained.

(Functional configuration of attribute estimation device)
FIG. 12 is a functional block diagram of the attribute estimation device according to the embodiment.
The attribute estimation device 300 functions as an acquisition unit 352, a learning unit 354, an estimation unit 356, and an output unit 358 by executing the program 3082 expanded on the RAM 306 from the non-volatile memory 308 by the CPU 302.

(Each functional configuration of the attribute estimation device)
Each functional configuration of the attribute estimation device 300 will be described in detail with reference to FIGS. 11 and 12. In the following, in explaining each function of the attribute estimation device 300, the relationship with the main hardware for realizing each functional block of the attribute estimation device 300 shown in FIG. 12 will also be described.
The acquisition unit 352 is realized by a command from the CPU 302 and a communication I / F 310. The acquisition unit 352 acquires information indicating the power consumption of a household whose attribute information is unknown. Here, the information indicating the power consumption of a household whose attribute information is unknown represents the magnitude of the power consumption such as a heat map by different colors. When the acquisition unit 352 acquires the information indicating the power consumption of the household whose attribute information is unknown, the acquisition unit 352 outputs the information indicating the power consumption to the learning unit 354.
The learning unit 354 is realized by an instruction from the CPU 302. When the learning unit 354 acquires the information indicating the power consumption of the household whose attribute information is unknown, which is output by the acquisition unit 352, deep learning is performed on the information indicating the power consumption of the household whose acquired attribute information is unknown. I do. When the learning unit 354 performs deep learning on the information indicating the power consumption of the household whose attribute information is unknown, the learning unit 354 outputs the feature amount of the power consumption obtained by performing the deep learning to the estimation unit 356. ..

The estimation unit 356 acquires the attribute information estimation table 1088 in which the feature amount of the power consumption and the attribute information of the power usage are associated with each other from the learning device 100 via the communication I / F 310. When the estimation unit 356 acquires the attribute information estimation table 1088, it stores it in the attribute information estimation table 3088 of the non-volatile memory 308.
When the estimation unit 356 acquires the feature amount of the power consumption of the household whose attribute information output by the learning unit 354 is unknown, the estimation unit 356 refers to the attribute information estimation table 3088 and the attribute related to the feature amount of the acquired power usage amount. Estimate the information. When the estimation unit 356 estimates the attribute information, the estimation unit 356 outputs the estimation result of the attribute information to the output unit 358.
The output unit 358 is realized by an instruction from the CPU 302. When the output unit 358 acquires the estimation result of the attribute information output by the estimation unit 356, the output unit 358 outputs the estimation result of the acquired attribute information to the display unit 314.

(Operation of learning device)
FIG. 13 is a flowchart showing an example of the operation of the attribute estimation device. In FIG. 13, the attribute estimation device 300 acquires the attribute information estimation table 1088 in which the feature amount of the power consumption and the attribute information of the power usage are associated with each other from the learning device 100, and the attribute information estimation table 1088 is used as the attribute information. The operation after being stored in the estimation table 3088 is shown.
(Step S1302) The acquisition unit 352 of the attribute estimation device 300 acquires information indicating the power consumption of a household whose attribute information is unknown. When the acquisition unit 352 acquires the information indicating the power consumption of the household whose attribute information is unknown, the acquisition unit 352 outputs the information indicating the power consumption of the household whose attribute information is unknown to the learning unit 354.
(Step S1304) When the learning unit 354 of the attribute estimation device 300 acquires the information indicating the power consumption of the household whose attribute information is unknown, the learning unit 354 of the attribute estimation device 300 obtains the power consumption of the household whose attribute information is unknown. Perform deep learning on the information shown. When the learning unit 354 performs deep learning on the information indicating the power consumption of the household whose attribute information is unknown, the learning unit 354 outputs the feature amount of the power consumption obtained by performing the deep learning to the estimation unit 356. ..

(Step S1306) When the estimation unit 356 of the attribute estimation device 300 acquires the feature amount of the power consumption of the household whose attribute information output by the learning unit 354 is unknown, the estimation unit 356 refers to the attribute information estimation table 3088 and acquires the acquired attribute. Estimate the attribute information associated with the feature quantity of the power consumption of the household whose information is unknown. When the estimation unit 356 estimates the attribute information related to the feature amount of the power consumption of the household whose attribute information is unknown, the estimation unit 356 outputs the estimation result of the attribute information to the output unit 358.
(Step S1308) When the output unit 358 of the attribute estimation device 300 acquires the estimation result of the attribute information output by the estimation unit 356, the output unit 358 outputs the estimation result of the acquired attribute information to the display unit 314.

In the above-described embodiment, the attribute estimation device 300 acquires the attribute information estimation table 1088 that associates the feature amount of the power consumption and the attribute information of the power consumption from the learning device 100, and the acquired attribute information estimation table 1088. Has been described in the case of storing in the attribute information estimation table 3088, but the present invention is not limited to this example. For example, the attribute estimation device 300 may inquire the learning device 100 of the attribute information related to the feature amount of the power consumption of the household whose attribute information is unknown. In response to an inquiry from the learning device 100, the attribute estimation device 300 notifies the learning device 100 of the attribute information associated with the feature amount of the power consumption of the household whose attribute information is unknown.

According to the attribute estimation device 300 according to the embodiment, the attribute information related to the characteristic amount of the power consumption of the household whose attribute information is unknown is estimated based on the attribute information estimation table 1088 acquired from the learning device 100. Can be done. Since it is possible to acquire attribute information of households for which unknown power consumption has been obtained, it is possible to help develop a business that provides new services based on the acquired attribute information.

(Third embodiment)
The attribute estimation system according to the embodiment will be described. The attribute estimation system according to the embodiment includes a learning device 100 according to the first embodiment and an attribute estimation device 300 according to the second embodiment.
As described above, the learning device 100 according to the embodiment acquires the information indicating the electric power consumption of each of the plurality of households and the attribute information of the household from which the information indicating the electric power consumption is obtained. The learning device 100 machine-learns information indicating the amount of electric power used, and obtains a feature amount of the amount of electric power used. The learning device 100 associates the feature amount of the electric power consumption with the attribute information of the household from which the information indicating the electric power consumption is obtained. The learning device 100 stores the feature amount of the electric power consumption and the attribute information in association with each other.

Further, as described above, the attribute estimation device 300 according to the embodiment acquires information indicating the power consumption of the household whose attribute information is unknown. The attribute estimation device 300 obtains the feature amount of the electric power consumption of the household whose attribute information is unknown by machine learning the information indicating the electric power consumption of the household whose attribute information is unknown. The attribute estimation device 300 according to the embodiment acquires information in which the feature amount of the power consumption and the attribute information are associated with each other from the learning device 100 described above. The attribute estimation device 300 according to the embodiment obtains attribute information related to the characteristic amount of the electric power consumption of a household whose attribute information is unknown, based on the information in which the acquired characteristic amount of the electric power consumption and the attribute information are associated with each other. presume.

(Operation of attribute estimation system)
FIG. 14 is a diagram showing the operation of the attribute estimation system according to the embodiment. In FIG. 14, the attribute estimation device 300 acquires the attribute information estimation table 1088 that associates the feature amount of the power consumption and the attribute information of the power usage from the learning device 100, and estimates the attribute information table 1088. The operation after being stored in the table 3088 is shown.
(Step S1402) The acquisition unit 152 of the learning device 100 acquires information indicating the power consumption of each of the plurality of households from the smart meter.
(Step S1404) When the acquisition unit 152 of the learning device 100 acquires the information indicating the electric power consumption of each of the plurality of households, the acquisition unit 152 acquires the attribute information of the household from which the information indicating the acquired electric power consumption is obtained.
(Step S1406) The learning unit 154 of the learning device 100 learns the information indicating the power consumption of each of the plurality of households acquired by the acquisition unit 152 by performing deep learning. When the learning unit 154 learned the information indicating the power consumption of each of the plurality of households by performing deep learning, the feature amount and the power consumption of the power consumption obtained by performing the deep learning were obtained. It is stored in association with home attribute information.

(Step S1408) The acquisition unit 352 of the attribute estimation device 300 acquires power usage information whose attribute information is unknown.
(Step S1410) The learning unit 354 of the attribute estimation device 300 learns the information indicating the power usage information whose attribute information is unknown by the acquisition unit 352 by performing deep learning.
(Step S1412) The estimation unit 356 of the attribute estimation device 300 acquires the attribute information associated with the feature amount of the power consumption obtained by performing deep learning from the attribute information estimation table 3088. The estimation unit 356 of the attribute estimation device 300 outputs the acquired attribute information to the output unit 358. When the output unit 358 acquires the attribute information output by the estimation unit 356, the output unit 358 displays the attribute information on the display unit 314. In the example shown in FIG. 14, as attribute information, wake-up time, family structure, presence / absence of elderly people, absence time, and vacation acquisition time are displayed.

The attribute estimation device 300 obtained the correct answer rate of the estimated attribute information by taking a questionnaire to the households from which the power consumption was obtained.
(Estimation accuracy (1))
FIG. 15 is a diagram showing an example (No. 1) of estimation accuracy of attribute information estimated by the attribute estimation device 300 according to the embodiment. FIG. 15 shows an estimation example of the presence or absence of the elderly. In FIG. 15, the horizontal axis is Epoch, the left vertical axis is Loss, and the right vertical axis is Accuracy. According to FIG. 15, it can be seen that as the Epoch increases, the loss decreases and the accuracy approaches a constant value. According to FIG. 15, it can be seen that an accuracy of about 70% is obtained.

(Estimation accuracy (2))
FIG. 16 is a diagram showing an example (No. 2) of estimation accuracy of attribute information estimated by the attribute estimation device 300 according to the embodiment. FIG. 16 shows an estimation example of whether or not the person is at home in the daytime. In FIG. 16, the horizontal axis is Epoch, the left vertical axis is Loss, and the right vertical axis is Accuracy. According to FIG. 16, it can be seen that as the Epoch increases, the loss decreases and the accuracy approaches a constant value. According to FIG. 16, it can be seen that an accuracy of about 80% is obtained.

(Estimation accuracy (3))
FIG. 17 is a diagram showing an example (No. 3) of estimation accuracy of attribute information estimated by the attribute estimation device 300 according to the embodiment. FIG. 17 shows an estimation example of whether or not a pet is kept. In FIG. 17, the horizontal axis is Epoch, the left vertical axis is Loss, and the right vertical axis is Accuracy. According to FIG. 17, it can be seen that as the Epoch increases, the loss decreases and the accuracy approaches a constant value. According to FIG. 17, it can be seen that an accuracy of about 60% is obtained.

In the above-described embodiment, the case where the attribute estimation device 300 acquires the attribute information estimation table 1088 from the learning device 100 has been described, but the present invention is not limited to this example. For example, the attribute estimation device 300 may inquire of the learning device 200 for the attribute information associated with the feature amount of the power usage information whose attribute information is unknown. Then, the learning device 100 may transmit the attribute information associated with the feature amount of the power usage information whose attribute information is unknown to the attribute estimation device 300 in response to the inquiry from the attribute estimation device 300. ..

In the above-described embodiment, the case where the learning device 100 and the attribute estimation device 300 are different devices has been described, but the present invention is not limited to this example. For example, the learning device 100 and the attribute estimation device 300 may be realized by one device.
In the above-described embodiment, the case where the attribute estimation system includes the learning device 100 and the attribute estimation device 300 has been described, but the present invention is not limited to this example. For example, the attribute estimation system may include the learning device 200 and the attribute estimation device 300. In this case, the learning device 200 and the attribute estimation device 300 may be realized by one device.

According to the attribute estimation system according to the embodiment, the attribute information of the household from which the information indicating the electric power consumption of each of the plurality of households and the information indicating the electric power consumption is obtained is acquired. The attribute estimation system machine-learns information indicating the amount of power used and obtains a feature amount of the amount of power used. The attribute estimation system stores the feature amount of the power consumption and the attribute information for which the information indicating the power consumption is obtained in association with each other.
Furthermore, according to the attribute estimation system, information indicating the power consumption of a household whose attribute information is unknown is acquired, and information indicating the power consumption of a household whose attribute information is unknown is deep-learned. Find the characteristic amount of power consumption in a household whose attribute information is unknown. The attribute estimation system uses information that associates the characteristic amount of power consumption with the attribute information for which information indicating the power consumption amount is obtained, and the attribute information related to the characteristic amount of power consumption of a household whose attribute information is unknown. To estimate. By configuring in this way, the attribute estimation system according to the embodiment can acquire the attribute information of the household from which the unknown power consumption is obtained, so that the business of providing a new service according to the acquired attribute information can be established. It can help you develop.

<Modification example>
The third embodiment described above can be applied to the attribute estimation system according to the modified example. The attribute estimation system according to the modified example applies a color heat map to the information indicating the amount of power used. In the attribute system according to the modified example, the concept of time is included in the heat map in addition to the amount of power used. With this configuration, the magnitude of power consumption and the concept of time can be represented by different colors.
FIG. 18 is an example of information showing the amount of power used by the attribute estimation system according to the modified example. In the example shown in FIG. 18, the information indicating the power consumption is represented by 256 gradations of RGB for each pixel constituting the information indicating the power consumption. Specifically, the electric energy is shown in red, the time (1 o'clock to 24 o'clock) is shown in green, and the date (1 / 1-12 / 31) is shown in blue. In FIG. 18, the magnitude of power consumption and the concept of time are visualized.

(Estimation accuracy (4))
FIG. 19 is a diagram showing an example (No. 4) of estimation accuracy of attribute information estimated by the attribute estimation device 300 according to the embodiment.
FIG. 19 shows an estimation example of the presence or absence of the elderly. In FIG. 19, the horizontal axis is Epoch, the left vertical axis is Loss, and the right vertical axis is Accuracy. According to FIG. 19, it can be seen that as the Epoch increases, the loss decreases and the accuracy approaches a constant value. According to FIG. 15, it can be seen that an accuracy of about 80% is obtained.
It can be seen that the estimation accuracy is improved by about 10% as compared with the estimation accuracy shown in (Estimation accuracy (No. 1)).

According to the above-mentioned modification, the case where the color heat map is applied to the attribute estimation system has been described, but the present invention is not limited to this example. For example, the color heat map may be applied to the learning device 100 according to the first embodiment, the learning device 200 according to the modified example, or the attribute estimation device 300 according to the second embodiment. May be applied to.
According to the above-mentioned modification, the case where the color heat map includes the concept of time in addition to the amount of electric power used has been described, but the present invention is not limited to this example. For example, outside air temperature data, gas usage data, date and time information may be included.
According to the attribute estimation system according to the modified example, by applying the color heat map to the information indicating the power consumption, the estimation accuracy of the attribute information of the information indicating the power consumption of the household whose attribute information is unknown is improved. be able to.

Although the embodiments and variations thereof have been described above, these embodiments and variations thereof are presented as examples, and the scope of the invention is not intended to be limited. These embodiments and variations thereof can be implemented in various other embodiments, and various omissions, replacements, changes, and combinations can be made without departing from the gist of the invention. These embodiments and variations thereof are included in the scope and gist of the invention, and at the same time, are included in the scope of the invention described in the claims and the equivalent scope thereof.

The learning device and the attribute estimation device described above may be realized by a computer. In that case, a program for realizing the function of each functional block is recorded on a computer-readable recording medium. It may be realized by having a computer system read a program recorded on this recording medium and executing it by a CPU. The term "computer system" as used herein includes hardware such as an OS (Operating System) and peripheral devices.
Further, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, or a CD-ROM. Further, the "computer-readable recording medium" includes a storage device such as a hard disk built in a computer system.

Further, the "computer-readable recording medium" may include a medium that dynamically holds the program for a short period of time. What dynamically holds the program for a short period of time is, for example, a communication line when the program is transmitted via a network such as the Internet or a communication line such as a telephone line.
Further, the "computer-readable recording medium" may include a medium that holds a program for a certain period of time, such as a volatile memory inside a computer system serving as a server or a client. Further, the above program may be for realizing a part of the above-mentioned functions. Further, the above-mentioned program may be realized by combining the above-mentioned functions with a program already recorded in the computer system. Further, the above program may be realized by using a programmable logic device. The programmable logic device is, for example, an FPGA (Field Programmable Gate Array).

The above-mentioned learning device and the attribute estimation device have a computer inside. The process of each process of the learning device and the attribute estimation device described above is stored in a computer-readable recording medium in the form of a program, and the computer reads and executes this program to perform the process. Will be.
Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Further, the computer program may be distributed to the computer via a communication line, and the computer receiving the distribution may execute the program.
Further, the above program may be for realizing a part of the above-mentioned functions.
Further, a so-called difference file (difference program) may be used, which can realize the above-mentioned function in combination with a program already recorded in the computer system.

100, 200 ... Learning device, 102 ... CPU, 104 ... ROM, 106 ... RAM, 108 ... Non-volatile memory, 110 ... Communication I / F, 112 ... Operation unit, 114 ... Display unit, 116 ... Bus, 152 ... Acquisition unit , 154 ... Learning unit, 156 ... Output unit, 252 ... Acquisition unit, 254 ... Learning unit, 256 ... Output unit, 258 ... Creation unit, 300 ... Attribute estimation device, 302 ... CPU, 304 ... ROM, 306 ... RAM, 308 ... non-volatile memory, 310 ... communication I / F, 312 ... operation unit, 314 ... display unit, 316 ... bus, 352 ... acquisition unit, 354 ... learning unit, 356 ... estimation unit, 358 ... output unit, 1082 ... program, 1084 ... Power usage information, 1086 ... Attribute information, 1088 ... Attribute information estimation table, 3082 ... Program, 3088 ... Attribute information estimation table

Claims (8)

An acquisition unit that acquires information indicating the amount of power used by each of a plurality of households and attribute information of the household from which the amount of power used has been obtained.
The information indicating the power consumption of each of the plurality of households acquired by the acquisition unit is machine-learned, and the feature amount of the power consumption obtained by the machine learning and the household from which the power consumption is obtained. The learning department that associates with the attribute information of
A storage unit that stores the feature amount of the power consumption associated with the learning unit and the attribute information of the household from which the power consumption is obtained in association with each other .
Of the information indicating the power consumption of each of the plurality of households acquired by the acquisition unit, the power consumption by use indicating the power usage by usage from each of the power usage of households having similar attribute information. With a creation unit that acquires one or more power sources and newly creates information indicating the power consumption amount based on the acquired power consumption amount for each application.
Equipped with
The information indicating the amount of electric power used is a graphic representation of the amount of electric power used in chronological order .
The learning unit is a feature amount of the power consumption obtained by machine learning the information indicating the power consumption newly created by the preparation unit and machine learning the information indicating the power usage newly created. A learning device that associates with the attribute information similar to . The learning device according to claim 1, wherein the information indicating the power consumption amount acquired by the acquisition unit represents the magnitude of the power consumption amount in different colors. The learning device according to claim 2, wherein the information indicating the amount of electric power used acquired by the acquisition unit further expresses the concept of time in different colors. The information indicating the power consumption acquired by the acquisition unit represents the power consumption and the information indicating the environmental condition measured at the home from which the power consumption was obtained, in different colors. The learning device according to claim 1. An acquisition unit that acquires information indicating the amount of electricity used in a household whose attribute information is unknown,
The feature amount of the power consumption amount is obtained by machine learning the information indicating the power consumption amount acquired by the acquisition unit, and the attribute information related to the feature amount is shown for the power consumption amount of each of the plurality of households. It is equipped with an estimation unit that estimates from information that associates the feature amount of the power consumption obtained by machine learning the information with the attribute information of the household from which the power usage is obtained.
The information indicating the amount of electric power used is a graphic representation of the amount of electric power used .
The estimation unit obtains the attribute information related to the feature amount of the power consumption from each of the power consumptions of households having similar attribute information among the information indicating the power consumptions of each of the plurality of households. The characteristic amount of the power consumption obtained by machine learning the information indicating the power usage newly created based on the power usage of one or more uses showing the usage amount of the acquired power for each use. An attribute estimation device that estimates from information associated with similar attribute information . The attribute estimation device according to claim 5 , wherein the information indicating the electric power consumption of a household whose attribute information is unknown acquired by the acquisition unit represents the magnitude of the electric power consumption by a difference in color. The attribute estimation device according to claim 6 , wherein the information indicating the power consumption of a household whose attribute information is unknown acquired by the acquisition unit further expresses the concept of time by different colors. The information indicating the power consumption of the household whose attribute information is unknown acquired by the acquisition unit is the color of the power consumption and the information indicating the environmental condition measured in the household from which the power consumption is obtained. The attribute estimation device according to claim 5 , which is represented by a difference.

Images