JP2023183921A - Determination apparatus, determination system, determination method, and program - Google Patents

Abstract

To determine whether to compensate for missing data.SOLUTION: A determination unit 10 includes: a query unit 101 which acquires data from a database using a query; a missing determination unit 102 which determines whether data acquired by the query unit 101 is missing or not; an analysis unit 103 which analyzes data determined to be missing by the missing determination unit 102; and a compensation determination unit 104 which determines whether to compensate for missing of data determined to be missing, on the basis of a result analyzed by the analysis unit 103.SELECTED DRAWING: Figure 1

Description

The present disclosure relates to a determination device, a determination system, a determination method, and a program.

Device-related data is collected from devices and the collected data is utilized. The collected data is stored, for example, in a database server, and the user can make use of the collected data by acquiring and analyzing the data stored in the database server using a query.

Here, the collected data may be missing due to temporary equipment failure, poor communication, or the like. For example, events may occur where information about temperature is lost only in a certain time period, or all data is lost in a different time period.

In order to deal with such problems, it is conceivable to interpolate missing data using some method. For example, Patent Document 1 discloses an information processing apparatus that, when data of a certain device is missing at a certain time, uses data of another device at the same time to interpolate the missing data.

JP 2021-114064 Publication

The information processing device described in Patent Document 1 uses data from other devices to interpolate missing data, regardless of data trends, meanings, etc., so even if interpolation is undesirable, There is a problem that interpolation may occur.

For example, regarding log data of air conditioners, midsummer daytime is the time period when the influence on the air conditioners is greatest, so if there is a loss in this time period, it is not very desirable to interpolate and utilize this data. This is because it is difficult to interpolate data with small errors because there is a possibility that there will be a particularly large change in temperature and room temperature during the above time period.

Therefore, there is a need for a technique for determining whether or not missing data should be interpolated.

In view of the above circumstances, an object of the present disclosure is to provide a determination device and the like that can determine whether data loss should be interpolated.

In order to achieve the above object, the determination device according to the present disclosure includes:
a query means for retrieving data from the database by query;
Deficiency determination means for determining whether or not there is a deficiency in the data acquired by the query means;
an analysis means for analyzing data determined to be defective by the defect determination means;
Interpolation possibility determining means for determining whether or not to interpolate the missing data determined to have the missing data based on the analysis result by the analyzing means;
Equipped with.

According to the present disclosure, it is possible to determine whether data loss should be interpolated.

A diagram showing the overall configuration of a determination system according to Embodiment 1 of the present disclosure A diagram showing an example of data acquired by the determination device according to Embodiment 1 of the present disclosure A diagram showing an example of data interpolation by the determination device according to Embodiment 1 of the present disclosure A diagram showing an example of data acquired by the determination device according to Embodiment 1 of the present disclosure A diagram showing an example of data interpolation by the determination device according to Embodiment 1 of the present disclosure A diagram showing an example of a condition table used in determining whether or not interpolation is possible by the determination device according to Embodiment 1 of the present disclosure. A diagram showing an example of the hardware configuration of a determination device according to Embodiment 1 of the present disclosure A flowchart illustrating an example of the operation of determining whether or not interpolation is possible by the determination device according to Embodiment 1 of the present disclosure. A diagram showing the overall configuration of a determination system according to Embodiment 2 of the present disclosure A diagram showing the overall configuration of a determination system according to Embodiment 3 of the present disclosure A diagram showing the overall configuration of a determination system according to a modification of the present disclosure

Hereinafter, a determination system according to an embodiment of the present disclosure will be described with reference to the drawings. In each drawing, the same or equivalent parts are given the same reference numerals.

(Embodiment 1)
A determination system 1 according to the first embodiment will be described with reference to FIG. The determination system 1 includes a determination device 10 and a database server 20. The determination device 10 and the database server 20 are communicably connected. Although details will be described later, the determination system 1 is a system in which the determination device 10 acquires data from the database server 20, and when the acquired data has a defect, determines whether or not the defect should be interpolated. The determination system 1 is an example of a determination system according to the present disclosure.

The database server 20 is a database server that stores, as a database, operational data collected from devices via the Internet, for example. The operating data stored in the database server 20 may be, for example, operating data of an air conditioner or an electric water heater.

The determination device 10 acquires data from the database server 20 based on a query, and determines whether or not the deletion should be interpolated when the acquired data has a defect. Details of this determination will be described later. The determination device 10 is, for example, a personal computer. The determination device 10 may be one that acquires data from the database server 20 at regular intervals through, for example, preset periodic batch processing, or one that acquires data from the database server 20 through user operations. You can. In either case, the determination device 10 determines whether or not the loss should be interpolated when there is a loss in the acquired data. The determination device 10 is an example of a determination device according to the present disclosure.

Further, although the details will be described later, when the determination device 10 determines that the missing data should be interpolated, it interpolates the missing data and then acquires the data again by query, and when it determines that the missing data should not be interpolated, Exclude all data from the period in which data is missing, and retrieve the data again by query.

Next, an example of data and an example of data interpolation when the data stored in the database server 20 is operation data of air conditioning equipment will be described with reference to FIGS. 2 and 3.

The data shown in FIG. 2 is room temperature data detected by the air conditioner during a certain period of time. However, as shown in FIG. 2, there are missing data.

If an attempt is made to interpolate the data shown in FIG. 2, the data shown in FIG. 3 will be obtained. Figure 3 shows that data at 01:20:00 and 01:40:00 can be interpolated from data before and after the loss, but data at 11:20:00 and 11:40:00 should not be interpolated. It is shown that. The reason why the former is interpolated is because the missing time period is late at night, a time period when the temperature is generally less likely to change. On the other hand, the reason why interpolation should not be performed in the latter case is because the time period of the loss is near noon, which is a time period when the temperature is generally likely to change. It is preferable not to perform interpolation on such data that change easily. For example, Lagrangian interpolation can be adopted as a data interpolation method. In FIG. 3, interpolation is performed based on data in the same column.

As another example, an example of data and an example of data interpolation when the data stored in the database server 20 is operating data of an electric water heater will be described with reference to FIGS. 4 and 5.

The data shown in Figure 4 includes the amount of hot water remaining in the electric water heater over a certain period of time, the cumulative amount of hot water, the status of whether hot water is being filled, the status of whether the shower is being used, and the status of whether hot water is being supplied to the kitchen. This is data regarding. FIG. 5 shows how each defect is interpolated or not interpolated. The difference between Figure 5 and Figure 3 is that in Figure 3, whether or not interpolation is possible is determined based on the data in the same column as the missing data, whereas in Figure 5, only the data in the same column as the missing data is determined. However, whether or not interpolation is possible is determined based on data that is different from the missing data.

The interpolation determination conditions in the case shown in FIG. 5 are more complex than those in the case shown in FIG. When the determination conditions are complicated, such as when data in a plurality of columns are involved in determining whether interpolation is possible, the determination whether interpolation is possible or not is determined based on a condition table as shown in FIG. 6, for example. In FIG. 6, for example, no "condition" is set for column A, so no interpolation is performed for column A. Also, for example, a "condition" is set for column B, and "single interpolation" is set to True, so when this "condition" is met, the missing data will be interpolated based on other data in column B. becomes. For example, for columns C and D, "condition" is set, "interpolation from other columns" is set to True, and "reference column name" is set, so when the "condition" is met, the "reference column name" is set. The missing data will be interpolated based on the data in "column name".

In this way, no matter what type of data is stored in the database server 20, by appropriately setting the judgment conditions as shown in FIG. You will be able to do it.

Referring again to FIG. 1, the functional configuration of the determination device 10 will be explained. The determination device 10 includes a communication section 100, a query section 101, a defect determination section 102, an analysis section 103, and an interpolation possibility determination section 104.

The communication unit 100 communicates with the database server 20. The communication unit 100 is realized by, for example, a network interface.

The query unit 101 communicates with the database server 20 via the communication unit 100, makes inquiries to the database server 20 using queries, and acquires data from the database server 20. The query unit 101 is an example of a query unit according to the present disclosure. In addition to the above, the query unit 101 has a function related to the interpolation determination unit 104 described later, which will be described later.

The defect determination unit 102 determines whether or not there is a defect in the data acquired by the query unit 101. For example, if the data acquired by the query unit 101 is as shown in FIG. 2, there is a loss, so the loss determination unit 102 determines that the data acquired by the query unit 101 has a loss. On the other hand, for example, if the data acquired by the query unit 101 is only data for the period from 02:00:00 to 02:40:00 shown in FIG. 2, there is no missing data in this range, so The loss determination unit 102 determines that there is no loss in the data acquired by the query unit 101. The defect determination unit 102 is an example of defect determination means according to the present disclosure.

The analysis unit 103 analyzes the data determined to be missing by the loss determination unit 102. The analysis unit 103 particularly analyzes the relationship between columns with defects and columns without defects. The analysis unit 103 analyzes the relationship between columns with defects and columns without defects, based on the condition table shown in FIG. 6, for example. This analysis result is used by the interpolation possibility determining unit 104, which will be described later, to determine whether or not interpolation is to be performed. The analysis unit 103 is an example of an analysis means according to the present disclosure.

The interpolation possibility determining unit 104 determines, based on the analysis result by the analyzing unit 103, whether or not the data determined to be missing by the loss determining unit 102 should be interpolated. The interpolation possibility determination unit 104 is an example of an interpolation possibility determination unit according to the present disclosure.

The functions of the query unit 101 related to the interpolation possibility determination unit 104 described above will be explained. When the interpolation determination unit 104 (described later) determines that interpolation should be performed, the query unit 101 interpolates the missing data and acquires data again by query. When the interpolation determination unit 104 (described later) determines that interpolation should not be performed, the query unit 101 removes all data from the period in which data is missing and acquires data again by query. For example, consider the case where the data is shown in FIG. The query unit 101 interpolates the data at 21:05:00 and 21:15:00, and then acquires the data again by query. For the data at 21:25:00, the query unit 101 removes the data at the date and time and all the data in the row corresponding to the date and time, and then acquires the data again by query.

Next, an example of the hardware configuration of the determination device 10 will be described with reference to FIG. 7. The determination device 10 shown in FIG. 7 is realized by, for example, a computer such as a personal computer or a microcontroller.

The determination device 10 includes a processor 1001, a memory 1002, an interface 1003, and a secondary storage device 1004, which are connected to each other via a bus 1000.

The processor 1001 is, for example, a CPU (Central Processing Unit). Each function of the determination device 10 is realized by the processor 1001 reading the operating program stored in the secondary storage device 1004 into the memory 1002 and executing it.

The memory 1002 is a main storage device composed of, for example, RAM (Random Access Memory). The memory 1002 stores an operating program read from the secondary storage device 1004 by the processor 1001. Furthermore, the memory 1002 functions as a work memory when the processor 1001 executes an operating program.

The interface 1003 is an I/O (Input/Output) interface such as a serial port, a USB (Universal Serial Bus) port, or a network interface. The functions of the communication unit 100 are realized by the interface 1003.

The secondary storage device 1004 is, for example, a flash memory, an HDD (Hard Disk Drive), or an SSD (Solid State Drive). Secondary storage device 1004 stores operating programs executed by processor 1001.

Next, an example of the operation of determining whether or not interpolation is possible by the determination device 10 will be described with reference to FIG. The operation shown in FIG. 8 is executed, for example, when the above-described batch processing is started. Alternatively, it is executed when the determination device 10 attempts to acquire data from the database server 20 due to a user's operation.

The query unit 101 of the determination device 10 acquires data from the database server 20 by a query (step S101).

The defect determination unit 102 of the determination device 10 determines whether there is a defect in the data acquired in step S101 (step S102).

When there is no loss in the data (step S102: No), since there is no loss to be interpolated in the first place, the determination device 10 ends the operation of determining whether or not interpolation is possible.

When the data is missing (step S102: Yes), the analysis unit 103 of the determination device 10 analyzes the data determined to be missing in step S102 (step S103).

The interpolation possibility determination unit 104 of the determination device 10 determines whether data loss should be interpolated based on the analysis result in step S103 (step S104).

When determining that data loss should be interpolated (step S104: Yes), the query unit 101 interpolates the loss and acquires data again by query (step S105). Then, the determination device 10 ends the operation of determining whether or not interpolation is possible.

When determining that data loss should not be interpolated (step S104: No), the query unit 101 excludes all data from the period in which the data loss occurs, and acquires data again by query (step S106). Then, the determination device 10 ends the operation of determining whether or not interpolation is possible.

Note that, after the determination device 10 completes the operation of determining whether or not interpolation is possible, the determination device 10 executes, for example, an operation for data utilization. Examples of actions for utilizing data include proposing an appropriate energy-saving plan and providing monitoring services.

The determination system 1 according to the first embodiment has been described above. According to the determination system 1 according to the first embodiment, it is possible to determine whether data loss should be interpolated. Therefore, the risk of undesirable interpolation can be reduced.

(Embodiment 2)
Determination system 1 according to Embodiment 2 will be described with reference to FIG. 9. The determination system 1 according to the second embodiment differs from the first embodiment in that the determination device 10 further includes a registration section 105, a metadata determination section 106, and a storage section 110. Furthermore, the function of the interpolation possibility determining unit 104 is also slightly different.

The storage unit 110 stores metadata that will be described later. Storage unit 110 is an example of storage means according to the present disclosure.

The registration unit 105 stores metadata regarding the data in the storage unit 110 when the interpolation possibility determining unit 104 determines that the data determined to be missing should not be interpolated. The metadata includes, for example, the data size of the data, a hash value that uniquely identifies the data, and the like. The registration unit 105 is an example of a registration unit according to the present disclosure.

The metadata determination unit 106 acquires the metadata of the data acquired by the query unit 101 and determines whether the acquired metadata matches the metadata stored in the storage unit 110. Note that there may be a case where there is a plurality of metadata stored in the storage unit 110 corresponding to the data that is the determination target of the interpolation possibility determination unit 104, but in this case, any one of the multiple metadata It is only necessary to determine whether or not they match. The metadata determining unit 106 is an example of a metadata determining unit according to the present disclosure.

The interpolation possibility determining unit 104 determines whether to interpolate the data obtained by the query unit 101 and which corresponds to the metadata determined by the metadata determining unit 106 to match the metadata stored in the storage unit. It is judged that it is not. This means that for data that was once determined not to be interpolated in the past, by utilizing metadata, it is possible to determine again that interpolation should not be performed without having the analysis unit 103 analyze the data again. That means it can be done.

The determination system 1 according to the second embodiment has been described above. According to the determination system 1 according to the second embodiment, by utilizing metadata, it is possible to determine again that data that has been determined not to be interpolated should not be interpolated without analyzing the data again. Therefore, the processing load on the determination device 10 can be reduced.

(Embodiment 3)
A determination system 1 according to the third embodiment will be described with reference to FIG. 10. The determination system 1 according to the third embodiment differs from the first embodiment in that the determination device 10 includes a learning section 107 and a storage section 110. Further, the function of the interpolation possibility determining unit 104 is also slightly different from that in the first embodiment.

The storage unit 110 stores a model generated by the learning unit 107, which will be described later.

The learning unit 107 uses machine learning to learn the relationship between missing columns and non-missing columns for the data determined to be missing by the missing determination unit 102, and generates a model for interpolating the missing data. and save it in the storage unit 110. Therefore, it becomes possible to determine whether or not interpolation is possible without previously setting a condition table as shown in FIG. 6, for example. The learning unit 107 is an example of a learning means according to the present disclosure.

The interpolation possibility determination unit 104 determines whether or not interpolation should be performed based on the model stored in the storage unit 110 and the analysis result of the analysis unit 103.

The determination system 1 according to the third embodiment has been described above. According to the determination system 1 according to the third embodiment, a model is generated by learning the relationship between a column with a defect and a column without a defect, and it is determined whether or not interpolation is possible based on the model. Therefore, the effort of setting the condition table in advance can be reduced. Furthermore, as learning progresses, the accuracy of judgments can be expected to improve.

In addition, although Embodiment 3 was described above as a modification of Embodiment 1, Embodiment 2 may be similarly modified.

(Modified example)
Determination system 1 according to a modified example of the present disclosure will be described with reference to FIG. 11. The determination system 1 according to the modification of the present disclosure differs from the first embodiment in that it further includes a display device 30 and that the determination device 10 includes a display control unit 108. The display control unit 108 is connected to the display device 30.

The display device 30 is, for example, a display. The display device 30 displays images under the control of the display control unit 108. The display device 30 is an example of a display means according to the present disclosure.

The display control unit 108 controls the display device 30 and displays images described below. When the interpolation determination unit 104 determines that interpolation should be performed, the display control unit 108 displays to the user the data with the missing data interpolated and the data excluding all the data from the period where the data is missing. An image for selection is displayed on the display device 30. The user can obtain desired data by, for example, checking this image and selecting desired data by operating an input device (not shown).

In each of the embodiments described above, the determination device 10 has been described as one device. However, each functional unit of the determination device 10 may be distributed and provided in a plurality of devices. For example, the query unit 101 is provided in a dedicated device for making queries, and the functions of the defect determination unit 102, analysis unit 103, and interpolation determination unit 104 are provided in separate devices, and these devices communicate via a network. It can also be used as a thing.

In the hardware configuration shown in FIG. 7, the determination device 10 includes a secondary storage device 1004. However, the present invention is not limited to this, and a configuration may be adopted in which the secondary storage device 1004 is provided outside the determination device 10 and the determination device 10 and the secondary storage device 1004 are connected via the interface 1003. In this embodiment, removable media such as a USB flash drive or a memory card can also be used as the secondary storage device 1004.

Further, instead of the hardware configuration shown in FIG. 7, the determination device 10 may be configured with a dedicated circuit using an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or the like. . Furthermore, in the hardware configuration shown in FIG. 7, part of the functions of the determination device 10 may be realized by a dedicated circuit connected to the interface 1003, for example.

Hereinafter, various aspects of the present disclosure will be collectively described as supplementary notes.

(Additional note 1)
a query means for retrieving data from the database by query;
Deficiency determination means for determining whether or not there is a deficiency in the data acquired by the query means;
an analysis means for analyzing data determined to be defective by the defect determination means;
Interpolation possibility determining means for determining whether or not to interpolate the missing data determined to have the missing data based on the analysis result by the analyzing means;
A determination device comprising:
(Additional note 2)
The analysis means analyzes the relationship between a column with a defect and a column without a defect with respect to the data determined to have a defect.
The determination device according to Supplementary Note 1.
(Additional note 3)
Further comprising a learning means for learning a relationship between a column with a defect and a column without a defect and generating a model for interpolating the defect with respect to the data determined to have a defect,
The interpolation possibility determining means determines whether or not the missing data should be interpolated based on the analysis result by the analyzing means and the model generated by the learning means.
The determination device according to appendix 2.
(Additional note 4)
Furthermore, when the interpolation possibility determining means determines that the missing data should be interpolated, the querying means interpolates the missing data and obtains the data again by query, and then interpolates the missing data. When the interpolation possibility determining means determines that the missing data should not be interpolated, all the data from the period in which the missing data is generated is excluded from the data, and the data is acquired again by query.
The determination device according to any one of Supplementary Notes 1 to 3.
(Appendix 5)
a registration means for registering metadata regarding the data in a storage means when the interpolation possibility determining means determines that the missing data should not be interpolated;
metadata determining means for determining whether metadata of data acquired by the query means matches metadata registered in the storage means;
Furthermore,
When the metadata determining means determines that the metadata of the data acquired by the querying means matches the metadata registered in the storage means, the interpolation possibility determining means should interpolate the missing data. It is judged that it is not,
The determination device according to any one of Supplementary Notes 1 to 4.
(Appendix 6)
further comprising display control means for displaying an image on the display means,
When the interpolation possibility determining means determines that the missing data should be interpolated, the display control means displays the data that has been interpolated for the missing data and the period in which the data is missing. displaying on the display means an image for allowing the user to select data from which all data have been excluded;
The determination device according to any one of Supplementary Notes 1 to 5.
(Appendix 7)
a storage means for storing a database to be queried;
query means for retrieving data from the database by query;
Deficiency determination means for determining whether or not there is a deficiency in the data acquired by the query means;
an analysis means for analyzing data determined to be defective by the defect determination means;
Interpolation possibility determining means for determining whether or not to interpolate the missing data determined to have the missing data based on the analysis result by the analyzing means;
A determination system comprising:
(Appendix 8)
The computer is
Retrieve data from the database by querying
Determine whether there is any missing data in the data obtained by the query,
Analyze data determined to be missing,
Based on the analysis result, determining whether or not to interpolate the missing data determined to have the missing data;
Judgment method.
(Appendix 9)
computer,
a query means for retrieving data from a database by query;
Deficiency determination means for determining whether or not there is a deficiency in the data acquired by the query means;
analysis means for analyzing the data determined to be missing by the defect determination means;
interpolation possibility determining means for determining whether or not to interpolate the missing data determined to have the missing data based on the analysis result by the analyzing means;
A program that functions as

Reference Signs List 1 determination system, 10 determination device, 20 database server, 30 display device, 100 communication unit, 101 query unit, 102 missing determination unit, 103 analysis unit, 104 interpolation availability determination unit, 105 registration unit, 106 metadata determination unit, 107 learning unit, 108 display control unit, 110 storage unit, 1000 bus, 1001 processor, 1002 memory, 1003 interface, 1004 secondary storage device.

Claims (9)

a query means for retrieving data from the database by query;
Deficiency determination means for determining whether or not there is a deficiency in the data acquired by the query means;
an analysis means for analyzing data determined to be defective by the defect determination means;
Interpolation possibility determining means for determining whether or not to interpolate the missing data determined to have the missing data based on the analysis result by the analyzing means;
A determination device comprising: The analysis means analyzes the relationship between a column with a defect and a column without a defect with respect to the data determined to have a defect.
The determination device according to claim 1. Further comprising a learning means for learning a relationship between a column with a defect and a column without a defect and generating a model for interpolating the defect with respect to the data determined to have a defect,
The interpolation possibility determining means determines whether or not the missing data should be interpolated based on the analysis result by the analyzing means and the model generated by the learning means.
The determination device according to claim 2. Furthermore, when the interpolation possibility determining means determines that the missing data should be interpolated, the querying means interpolates the missing data and obtains the data again by query, and then interpolates the missing data. When the interpolation possibility determining means determines that the missing data should not be interpolated, all the data from the period in which the missing data is generated is excluded from the data, and the data is acquired again by query.
The determination device according to claim 1. a registration means for registering metadata regarding the data in a storage means when the interpolation possibility determining means determines that the missing data should not be interpolated;
metadata determining means for determining whether metadata of data acquired by the query means matches metadata registered in the storage means;
Furthermore,
When the metadata determining means determines that the metadata of the data acquired by the querying means matches the metadata registered in the storage means, the interpolation possibility determining means should interpolate the missing data. It is judged that it is not,
The determination device according to claim 1. further comprising display control means for displaying an image on the display means,
When the interpolation possibility determining means determines that the missing data should be interpolated, the display control means displays the data that has been interpolated for the missing data and the period in which the data is missing. displaying on the display means an image for allowing the user to select data from which all data have been excluded;
The determination device according to claim 1. a storage means for storing a database to be queried;
query means for retrieving data from the database by query;
Deficiency determination means for determining whether or not there is a deficiency in the data acquired by the query means;
an analysis means for analyzing data determined to be missing by the defect determination means;
Interpolation possibility determining means for determining whether or not to interpolate the missing data determined to have the missing data based on the analysis result by the analyzing means;
A determination system comprising: The computer is
Retrieve data from the database by querying
Determine whether there is any missing data in the data obtained by the query,
Analyze data determined to be missing,
Based on the analysis result, determining whether or not to interpolate the missing data determined to have the missing data;
Judgment method. computer,
a query means for retrieving data from a database by query;
Deficiency determination means for determining whether or not there is a deficiency in the data acquired by the query means;
analysis means for analyzing the data determined to be missing by the defect determination means;
interpolation possibility determining means for determining whether or not to interpolate the missing data determined to have the missing data based on the analysis result by the analyzing means;
A program that functions as

Images