JP7590494B2 - Promotion device, promotion method and promotion program - Google Patents

Description

The present invention relates to a recommendation device, a recommendation method, and a recommendation program.

In recent years, there has been a growing expectation for technology that can learn and utilize the characteristics of time-series data generated by sensors and other sources in various locations in urban spaces, including plants.

JP 2019-185194 A

However, the distribution patterns of time-series data generated in plants and urban spaces often change depending on the situation. For example, the distribution pattern of time-series data may change due to differences in equipment operating rates or material concentrations. This may result in cases where a model previously generated by learning using time-series data cannot be applied.

The present invention has been made in consideration of the above, and aims to apply a trained model that corresponds to the distribution pattern of time-series data generated in plants and urban spaces.

In order to solve the above-mentioned problems and achieve the object, the recommendation device according to the present invention is characterized by having a storage unit that stores a model generated by learning using time series data for a predetermined period as teacher data in association with the distribution of the teacher data, a calculation unit that compares the distribution of the time series data for a predetermined period with the distribution of the teacher data to calculate a similarity, and an identification unit that identifies the model generated by learning using the teacher data when the calculated similarity is equal to or greater than a predetermined threshold.

The present invention makes it possible to apply a trained model that corresponds to the distribution pattern of time-series data generated in plants and urban spaces.

FIG. 1 is a schematic diagram illustrating a schematic configuration of a thrust device. FIG. 2 is a diagram illustrating an example of the data structure of a model list. FIG. 3 is a diagram for explaining the process of the presenting unit. FIG. 4 is a flow chart illustrating a recommendation process. FIG. 5 is a diagram illustrating a computer that executes the recommendation program.

One embodiment of the present invention will be described in detail below with reference to the drawings. Note that the present invention is not limited to this embodiment. In addition, in the drawings, the same parts are denoted by the same reference numerals.

[Configuration of the promotion device]
Fig. 1 is a schematic diagram illustrating a schematic configuration of a recommendation device. As illustrated in Fig. 1, the recommendation device 10 is realized by a general-purpose computer such as a personal computer, and includes an input unit 11, an output unit 12, a communication control unit 13, a storage unit 14, and a control unit 15.

The input unit 11 is realized using input devices such as a keyboard and a mouse, and inputs various instruction information, such as a command to start processing, to the control unit 15 in response to input operations by an operator. The output unit 12 is realized by a display device such as a liquid crystal display, a printing device such as a printer, etc.

The communication control unit 13 is realized by a NIC (Network Interface Card) or the like, and controls communication between the control unit 15 and an external device via a network. For example, the communication control unit 13 controls communication between the control unit 15 and a sensor that outputs time series data to be processed in the recommendation process described below, a management device that manages the time series data, and the like.

The storage unit 14 is realized by a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 14 stores in advance the processing program that operates the recommendation device 10 and data used during execution of the processing program, or stores the data temporarily each time processing is performed. The storage unit 14 may be configured to communicate with the control unit 15 via the communication control unit 13.

In this embodiment, the storage unit 14 prestores, for example, a model list 14a to be used in the recommendation process described below. The model list 14a is information that associates a model generated by learning using time-series data for a predetermined period as training data with the distribution of the training data.

Here, FIG. 2 is a diagram illustrating an example of the data configuration of a model list. Specifically, as illustrated in FIG. 2, model list 14a includes a trained model and a distribution of time series data over a predetermined time span in the past that was used as training data to generate the model. The distribution of time series data means, for example, a graph showing the relationship between time and data value. Model list 14a illustrated in FIG. 2 includes, for example, a combination of model A and training data distribution A corresponding to model A.

The model list 14a may associate models with the teacher data that corresponds to the models. For example, data values at each time may be stored as teacher data.

Returning to the explanation of FIG. 1, the control unit 15 is realized using a CPU (Central Processing Unit) or the like, and executes a processing program stored in memory. As a result, the control unit 15 functions as an acquisition unit 15a, a calculation unit 15b, a presentation unit 15c, an identification unit 15d, and a generation unit 15e, as exemplified in FIG. 1. Note that each or some of these functional units may be implemented in different hardware. For example, the generation unit 15e may be implemented in a device separate from the other functional units. The control unit 15 may also include other functional units.

The acquisition unit 15a acquires time series data for a predetermined period. For example, the acquisition unit 15a acquires time series data such as sensor values for a predetermined time width that are the subject of the recommendation process described below, via the input unit 11, or from a management device that manages the sensors or sensor values, via the communication control unit 13. Here, the period of the time series data acquired by the acquisition unit 15a is the same as or longer than the time width of the teacher data in the model list 14a.

Here, the distribution pattern of the time series data may change depending on the equipment operating rate, the concentration of the material, etc.

The calculation unit 15b compares the distribution of the acquired time series data with the distribution of the teacher data to calculate the similarity. Specifically, the calculation unit 15b first obtains the distribution of the acquired time series data. For example, the calculation unit 15b generates a graph that shows the relationship between the time and the data value of the acquired time series data.

The calculation unit 15b then compares the distribution of the acquired time series data with the distribution of teacher data corresponding to each model in the model list 14a to calculate the similarity. For example, the calculation unit 15b calculates the similarity by normalizing and comparing the values of the time series data and the teacher data of the same time width. The calculation unit 15b calculates the similarity between the distribution of the time series data to be processed and each of the multiple teacher data.

When the calculated similarity is equal to or greater than a predetermined threshold, the presentation unit 15c presents models corresponding to the distribution of the teacher data in descending order of similarity. For example, the presentation unit 15c outputs corresponding models in descending order of similarity of the distribution of multiple teacher data to a user terminal or the like via the output unit 12 or the communication control unit 13. This enables the presentation unit 15c to recommend multiple models in an appropriate order.

Here, FIG. 3 is a diagram for explaining the processing of the presentation unit. In FIG. 3, for example, 10% is set as a threshold, and models A to C with similarity of 10% or more are presented in order of decreasing similarity. As illustrated in FIG. 3, the presentation unit 15c may also display and output the distribution of teacher data corresponding to each model. As shown in FIG. 3, the distribution of the acquired time-series data to be processed may also be presented.

The presentation unit 15c may receive an operation input by the operator to select one of the presented models. In this case, the operator can select the optimal model while comparing the results. In particular, when the distribution of the training data and the time series data is displayed, the operator can visually check the distribution of the time series data to be processed and the training data, and then select the model trained with the training data with the highest similarity.

Returning to the explanation of FIG. 1, the identification unit 15d identifies a model generated by learning using teacher data when the calculated similarity is equal to or greater than a predetermined threshold. For example, the identification unit 15d identifies a model corresponding to teacher data having the highest similarity among multiple teacher data. In this case, the identification unit 15d can automatically apply the identified model to the time series data to be processed.

Alternatively, the identification unit 15d may identify a model selected by the operator from among the models presented to the operator by the presentation unit 15c described above as the model to be applied to the time series data to be processed.

If any of the similarities calculated by the calculation unit 15b described above are less than a predetermined threshold, the recommendation device 10 may issue an alarm. For example, the presentation unit 15c outputs a message to the operator that none of the trained models generated so far can be applied to the time series data to be processed. In this case, for example, the operator can instruct the generation unit 15e to execute the process described below.

Alternatively, if all of the calculated similarities are less than a predetermined threshold, the generation unit 15e may automatically execute the process. The generation unit 15e generates a model by learning using the acquired time series data as teacher data. For example, the generation unit 15e generates a model by extracting and learning the features of the acquired time series data.

The generation unit 15e also adds the generated model to the model list 14a in association with the distribution of the time series data to be processed. This automatically updates the model list 14a.

[Recommendation Processing Procedure]
Next, an example of a recommendation process by the recommendation device 10 according to the present embodiment will be described with reference to Fig. 4. Fig. 4 is a flowchart illustrating a procedure of the recommendation process. The flowchart in Fig. 4 is started, for example, when an input is made to instruct the start of the recommendation process.

First, the acquisition unit 15a acquires time series data for a predetermined period. For example, the acquisition unit 15a acquires time series data such as sensor values for a predetermined time span that is to be subjected to the recommendation process described later (step S1).

Next, the calculation unit 15b compares the distribution of the acquired time series data with the distribution of the teacher data corresponding to each model in the model list 14a to calculate the similarity (step S2).

Then, if the calculated similarity is equal to or greater than a predetermined threshold, the identification unit 15d identifies the model generated by learning using the teacher data as the model to be applied to the time series data to be processed (step S3).

For example, the identification unit 15d identifies, among the multiple pieces of teacher data, a model corresponding to the teacher data with the highest similarity as the model to be applied to the time series data to be processed. Alternatively, the identification unit 15d identifies a model selected by the operator as the model to be applied to the time series data to be processed. This completes the series of recommendation processes.

If the calculated similarity is equal to or greater than a predetermined threshold, the presenting unit 15c may present models corresponding to the distribution of the training data in descending order of similarity and accept a model selection operation by the operator.

If any of the calculated similarities is less than a predetermined threshold, the recommendation device 10 may issue an alarm. Alternatively, the generation unit 15e may automatically generate a model by learning using the acquired time-series data as teacher data, and update the model list 14a.

[effect]
As described above, in the recommendation device 10 of the above embodiment, the storage unit 14 stores a model generated by learning using time-series data for a predetermined period as teacher data in association with the distribution of the teacher data. The acquisition unit 15a acquires time-series data for a predetermined period. The calculation unit 15b compares the distribution of the acquired time-series data with the distribution of the teacher data to calculate a similarity. The identification unit 15d identifies the model generated by learning using the teacher data when the calculated similarity is equal to or greater than a predetermined threshold.

This makes it possible, for example, to automatically apply the optimal model to the time series data to be processed. In this way, it becomes possible to apply a trained model that corresponds to the distribution pattern of time series data generated in plants or urban spaces.

When the calculated similarity is equal to or greater than a predetermined threshold, the presenting unit 15c presents models corresponding to the distribution of the teaching data in descending order of similarity. This enables the recommendation device 10 to recommend multiple models in an appropriate order. It also enables the operator to compare the results and select the most suitable model.

In addition, if the calculated similarities are all less than a predetermined threshold, the generation unit 15e generates a model by learning using the acquired time-series data as training data. This automatically updates the model list 14a, making it possible to improve the accuracy of the recommendation process.

[System configuration, etc.]
Each component of each device shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or a part of it can be functionally or physically distributed and integrated in any unit depending on various loads and usage conditions. Furthermore, each processing function performed by each device can be realized in whole or in part by a CPU or GPU and a program analyzed and executed by the CPU or GPU, or can be realized as hardware using wired logic.

Furthermore, among the processes described in this embodiment, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using known methods. In addition, the information including the processing procedures, control procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified.

[program]
A program in which the processing performed by the lifting device described in the above embodiment is written in a language executable by a computer can also be created. For example, a program in which the processing performed by the lifting device 10 according to the embodiment is written in a language executable by a computer can also be created. In this case, the computer executes the program, thereby obtaining the same effect as in the above embodiment. Furthermore, such a program may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read and executed by a computer to realize the same processing as in the above embodiment.

Figure 5 is a diagram showing an example of a computer that executes a recommendation program. The computer 1000 has, for example, a memory 1010, a CPU 1020, a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. Each of these components is connected by a bus 1080.

The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM 1012. The ROM 1011 stores a boot program such as a BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to a hard disk drive 1031. The disk drive interface 1040 is connected to a disk drive 1041. A removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1041. The serial port interface 1050 is connected to a mouse 1051 and a keyboard 1052, for example. The video adapter 1060 is connected to a display 1061, for example.

Here, the hard disk drive 1031 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. Each piece of information described in the above embodiment is stored, for example, in the hard disk drive 1031 or memory 1010.

The recommendation program is stored in the hard disk drive 1031, for example, as a program module 1093 in which instructions to be executed by the computer 1000 are written. Specifically, the program module 1093 in which each process executed by the recommendation device 10 described in the above embodiment is written is stored in the hard disk drive 1031.

In addition, data used for information processing by the recommendation program is stored as program data 1094, for example, in the hard disk drive 1031. Then, the CPU 1020 reads the program module 1093 and program data 1094 stored in the hard disk drive 1031 into the RAM 1012 as necessary, and executes each of the procedures described above.

The program module 1093 and program data 1094 related to the recommendation program are not limited to being stored in the hard disk drive 1031, but may be stored in a removable storage medium, for example, and read by the CPU 1020 via the disk drive 1041 or the like. Alternatively, the program module 1093 and program data 1094 related to the recommendation program may be stored in another computer connected via a network such as a LAN (Local Area Network) or a WAN (Wide Area Network), and read by the CPU 1020 via the network interface 1070.

The above describes an embodiment of the invention made by the inventor, but the present invention is not limited to the description and drawings that form part of the disclosure of the present invention according to this embodiment. In other words, other embodiments, examples, operational techniques, etc. made by those skilled in the art based on this embodiment are all included in the scope of the present invention.

REFERENCE SIGNS LIST 10 recommendation device 11 input unit 12 output unit 13 communication control unit 14 storage unit 14a model list 15 control unit 15a acquisition unit 15b calculation unit 15c presentation unit 15d identification unit 15e generation unit

Claims (4)

A storage unit that stores a model generated by learning using time-series data for a predetermined period as training data and a distribution of the training data in association with each other;
a calculation unit that calculates a similarity by comparing a distribution of time-series data for a predetermined period with a distribution of the teacher data;
a presentation unit that presents the model corresponding to the distribution of the training data in descending order of the similarity;
an identification unit that identifies the model generated by learning using the teacher data when the calculated similarity is equal to or greater than a predetermined threshold;
A promotion device comprising: The recommendation device according to claim 1, further comprising a generation unit that generates a model by learning using the time series data as training data when the calculated similarities are all less than a predetermined threshold. A promotion method performed by a promotion device, comprising:
The recommendation device has a storage unit that stores a model generated by learning using time-series data for a predetermined period as teacher data and a distribution of the teacher data in association with each other,
a calculation step of calculating a similarity by comparing a distribution of time-series data for a predetermined period with a distribution of the teacher data;
an identification step of identifying the model generated by learning using the training data when the calculated similarity is equal to or greater than a predetermined threshold;
a presentation step of presenting the model corresponding to the distribution of the training data in descending order of the similarity;
A recommendation method comprising: Refer to a storage unit that stores a model generated by learning using time-series data for a predetermined period as training data and a distribution of the training data in association with each other;
a calculation step of calculating a similarity by comparing a distribution of time series data for a predetermined period with a distribution of the teacher data;
a step of identifying the model generated by learning using the training data when the calculated similarity is equal to or greater than a predetermined threshold;
a presentation step of presenting the model corresponding to the distribution of the training data in descending order of the similarity;
A recommended program for causing a computer to execute the above.

Images