Data analysis device, data analysis method and program

Abstract

A data analysis device (1) comprises: a difference value calculation unit that, in a data segment (i, i+1) where the start point is i-th data at a predetermined date / time among a plurality of items of data, and the end point is (i+1)-th data at a date / time later than the predetermined date / time, calculates an explanatory variable difference value ΔX that is the difference between an initial value X(i) of an explanatory variable X included in the i-th data and a final value X(i+1) of the explanatory variable X included in the (i+1)-th data, and calculates an objective variable difference value ΔY that is the difference between an initial value Y(i) of an objective variable Y included in the i-th data and a final value Y(i+1) of the objective variable Y included in the (i+1)-th data; and a difference model derivation unit for deriving a difference model M indicating a relationship between the explanatory variable difference value ΔX and the objective variable difference value ΔY on the basis of a plurality of explanatory variable difference values ΔX and objective variable difference values ΔY.

Description

[Technical Field]

[0001] The present disclosure relates to a data analysis device for analyzing a plurality of data, a data analysis method, and a program for executing the data analysis method. [Background technology]

[0002] Conventionally, data analysis devices that analyze multiple data sets have been known. As an example of such devices, Patent Document 1 describes a data analysis device that performs multiple regression analysis on multiple time-series data sets and predicts future values ​​using the analysis results. Specifically, the data analysis device of Patent Document 1 calculates a multiple regression model of a response variable of performance data that has been assigned order information such as a time series, by adding terms obtained by first- and second-order differentiation of data fluctuation characteristics due to order information such as a time series as new explanatory variables to the explanatory variables of performance data that has been assigned order information such as a time series, and predicts the response variable at an arbitrary date and time or order. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-031714 Summary of the Invention

[0004] A data analysis apparatus according to one aspect of the present disclosure is a data analysis apparatus that analyzes a plurality of data constituted by one or more explanatory variables and one objective variable, and includes: a data acquisition unit that acquires the plurality of data; a difference value calculation unit that calculates, in a data interval having an ith data (i where i is an integer greater than or equal to 1) at a predetermined date and time among the plurality of data as a start value and an i+1th data (i+1) at a date and time after the predetermined date and time as an end value, an explanatory variable difference value that is the difference between the start value of the explanatory variable included in the ith data and the end value of the explanatory variable included in the i+1st data, and an objective variable difference value that is the difference between the start value of the objective variable included in the ith data and the end value of the objective variable included in the i+1st data; and a difference model derivation unit that derives a difference model that indicates a relationship between the explanatory variable difference values ​​and the objective variable difference values, based on the plurality of explanatory variable difference values ​​and the objective variable difference values.

[0005] A data analysis method according to one aspect of the present disclosure includes: The computer A data analysis method for analyzing a plurality of data items each consisting of one or more explanatory variables and one response variable, comprising: The computer acquiring the plurality of data; The computer calculating an explanatory variable difference value, which is the difference between the start value of the explanatory variable included in the i-th data and the end value of the explanatory variable included in the i+1-th data, and an objective variable difference value, which is the difference between the start value of the objective variable included in the i-th data and the end value of the objective variable included in the i+1-th data, in a data section having the i-th (i is an integer equal to or greater than 1) data at a predetermined date and time among the plurality of data as a start value and the i+1-th data at a date and time after the predetermined date and time as an end value; The computer deriving a difference model indicating a relationship between the explanatory variable difference values ​​and the dependent variable difference values ​​based on a plurality of the explanatory variable difference values ​​and the dependent variable difference values; The computer and predicting at least one of the end-point values ​​of the explanatory variables and the end-point values ​​of the response variables in the future data using the differential model.

[0006] These comprehensive or specific aspects may be realized as a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM, or may be realized as any combination of a system, a method, an integrated circuit, a computer program, and a recording medium. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a diagram illustrating an example of a data analysis system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating a configuration of a data analysis apparatus according to an embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of a data set according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of explanatory variables and response variables selected from a data set according to the embodiment. [Figure 5] FIG. 5 is a block diagram illustrating a functional configuration of the data analysis apparatus according to the first embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of a plurality of pieces of data acquired by the data analysis apparatus according to the first embodiment. [Figure 7] FIG. 7 is a diagram showing an example of a plurality of data sections set for a plurality of data. [Figure 8] FIG. 8 is a diagram showing the explanatory variable difference values ​​and the response variable difference values ​​in each data interval. [Figure 9] FIG. 9 is a diagram illustrating the relationship between the explanatory variable difference value and the response variable difference value in the first embodiment. [Figure 10] FIG. 10 is a diagram schematically illustrating the effect of the data analysis device according to the first embodiment. [Figure 11] FIG. 11 is a flowchart showing an example of a data analysis method according to the first embodiment. [Figure 12] FIG. 12 is a flowchart showing another example of the data analysis method according to the first embodiment. [Figure 13]FIG. 13 is a block diagram illustrating a functional configuration of a data analysis apparatus according to the second embodiment. [Figure 14] FIG. 14 is a diagram illustrating an example of a plurality of pieces of data acquired by the data analysis apparatus according to the second embodiment. [Figure 15] FIG. 15 is a diagram showing an example of a data section set for a plurality of data. [Figure 16] FIG. 16 is a diagram showing the explanatory variable difference values ​​and the response variable difference values ​​in each data interval. [Figure 17] FIG. 17 is a diagram illustrating the relationship between the explanatory variable difference value and the response variable difference value in the second embodiment. [Figure 18] FIG. 18 is a flowchart showing an example of a data analysis method according to the second embodiment. [Figure 19] FIG. 19 is a flowchart showing another example of the data analysis method according to the second embodiment. [Figure 20] FIG. 20 shows an example of the explanatory variable difference values ​​and the response variable difference values ​​calculated for the data set of FIG. [Figure 21] FIG. 21 shows an example in which the explanatory variable difference value and the response variable difference value are calculated using the standard values ​​for the data set in FIG. [Figure 22] FIG. 22 is a diagram showing another example of a plurality of data sections set for a plurality of data. DETAILED DESCRIPTION OF THE INVENTION

[0008] With the analysis device described in Patent Document 1, for example, when there are uncertain factors in the performance data that affect the objective variable, it becomes difficult to accurately analyze multiple data sets, making it difficult to accurately predict future values.

[0009] The present disclosure is intended to solve the above-mentioned problems, and aims to provide a data analysis device and the like that can analyze multiple data sets with high accuracy.

[0010] Hereinafter, embodiments will be described with reference to the drawings. The embodiments described below are all comprehensive or specific examples. The numerical values, shapes, materials, components, component placement and connection configurations, steps, and step order shown in the following embodiments are merely examples and are not intended to limit the present disclosure. Furthermore, among the components in the following embodiments, components not described in independent claims will be described as optional components.

[0011] Furthermore, each figure is a schematic diagram and is not necessarily an exact illustration. Furthermore, in each figure, substantially the same configuration is given the same reference numeral, and duplicate explanations may be omitted or simplified. Furthermore, even when the same object is shown in each figure, the scale may be changed for convenience.

[0012] (Embodiment 1) [Hardware configuration] FIG. 1 is a diagram showing an example of a data analysis system according to the present embodiment.

[0013] The data analysis system 900 in this embodiment includes a data analysis device 1 and a manufacturing management device 500.

[0014] The manufacturing management device 500 is installed in, for example, a manufacturing factory and manages a manufacturing system that manufactures products. The manufacturing management device 500 transmits a data set Ds obtained by the manufacturing system to the data analysis device 1 via a network such as the Internet. Details of the data set Ds will be described later with reference to FIGS. 3 and 4.

[0015] The data analysis device 1 is configured by a personal computer or the like, and receives a data set Ds from the above-described manufacturing control device 500. Then, the data analysis device 1 in this embodiment generates a plurality of models indicating the relationship between the data of the explanatory variables and the data of the response variables based on the data set Ds.

[0016] FIG. 2 is a diagram showing the configuration of data analysis device 1 according to this embodiment.

[0017] The data analysis device 1 includes an input unit 101 , an arithmetic circuit 102 , a memory 103 , an output unit 104 , a storage unit 105 , a database 106 , and a communication unit 107 .

[0018] The communication unit 107 communicates with devices external to the data analysis apparatus 1. The communication may be wireless or wired. The wireless communication method may be Wi-Fi (registered trademark), Bluetooth (registered trademark), ZigBee, or other methods. For example, the communication unit 107 communicates with the manufacturing management apparatus 500 and receives a data set Ds from the manufacturing management apparatus 500.

[0019] The input unit 101 has a function as an HMI (Human Machine Interface) that accepts input operations by a user, and includes, for example, a keyboard, a mouse, a touch sensor, a touch pad, and the like.

[0020] The output unit 104 has a display for displaying images, characters, etc., and the display is, for example, a liquid crystal display, a plasma display, an organic EL (Electro-Luminescence) display, etc. The output unit 104 may have a printer for printing images, characters, etc., and may have a function for storing data output from the arithmetic circuit 102 in the storage unit 105 in a file format.

[0021] The storage unit 105 stores a program (i.e., a computer program) 105a in which instructions to the arithmetic circuit 102 are written. The storage unit 105 may also store temporary data 105b temporarily generated by the processing of the arithmetic circuit 102. The storage unit 105 is a non-volatile recording medium, such as a magnetic storage device such as a hard disk, an optical disk, or a semiconductor memory. The program 105a is provided to the data analysis apparatus 1 via, for example, removable media or a network, and stored in the storage unit 105. The removable media is, for example, a CD-ROM (Compact Disc Read Only Memory) or a flash memory. For this reason, the communication unit 107 may be provided with an interface for reading the program 105a from removable media.

[0022] The memory 103 temporarily stores a program 105a that is read and expanded by the arithmetic circuit 102. Such memory 103 is, for example, a volatile RAM (Random Access Memory).

[0023] The arithmetic circuit 102 is a circuit that executes the program 105a loaded in the memory 103, and is, for example, a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). The arithmetic circuit 102 may use each piece of temporary data 105b stored in the storage unit 105 when executing the program 105a.

[0024] The database 106 is a non-volatile recording medium, similar to the storage unit 105, and is, for example, a magnetic storage device such as a hard disk, an optical disk, a semiconductor memory, etc. For example, the arithmetic circuit 102 acquires a data set Ds from the manufacturing control apparatus 500 via the network and the communication unit 107, and stores the data set Ds in the database 106.

[0025] In this embodiment, the storage unit 105 and the database 106 are separate recording media, but the storage unit 105 and the database 106 may be configured as a single recording medium that includes them.

[0026] [Dataset] FIG. 3 is a diagram showing an example of the data set Ds according to this embodiment.

[0027] The data set Ds is a raw data set transmitted from the manufacturing management device 500, and is a structured data set consisting of multiple manufacturing data indicating, for example, the physical properties, process conditions, and quality of the products manufactured by the manufacturing process of the above-mentioned manufacturing system. As shown in FIG. 3, such a data set Ds indicates the names of multiple variables and the data of those variables. Note that the data may be any data as long as it indicates at least one of letters and numbers. The first row of the data set Ds contains the names of the multiple variables, and the second and subsequent rows of the data set Ds contain the data of the multiple variables.

[0028] The leftmost column of the data set Ds shows the production date. Here, we will explain an example in which a manufacturing process is set for each production date, that is, an example in which once a manufacturing process is set, the same manufacturing process is used for production throughout the day.

[0029] As shown in FIG. 3, the first row of dataset Ds contains the variable names: physical property 1, physical property 2, physical property 3, process condition 1, inspection 1, and inspection 2. Physical property 1, physical property 2, and physical property 3 are appropriately selected from, for example, viscosity, particle size, solid content ratio, etc. Process condition 1 is appropriately selected from, for example, flow rate, pressure, etc. Inspection 1 and inspection 2 are inspection items for the product or semi-finished product manufactured using physical property 1, physical property 2, physical property 3, and process condition 1. Inspection 1 and inspection 2 are appropriately selected from, for example, coating weight, film thickness, coating area, etc. The second and subsequent rows of dataset Ds contain data for the variables identified by these variable names.

[0030] In this embodiment, physical property 1, physical property 2, physical property 3, and process condition 1 shown in Fig. 3 are explanatory variables, and inspection 1 and inspection 2 are response variables. In this example, four types of explanatory variables and two types of response variables are shown.

[0031] Fig. 4 is a diagram showing an example of explanatory variables and response variables selected from dataset Ds. Fig. 4 shows that physical property 1 and test 1 have been selected from dataset Ds shown in Fig. 3 and sorted by production date. In Fig. 4, data numbers are assigned in chronological order according to the production dates. In the figure, physical property 1 is selected as the explanatory variable, and test 1 is selected as the response variable.

[0032] Note that the method of selecting the explanatory variables and the dependent variable is not limited to this. For example, from the data set Ds, physical property 2 may be selected as the explanatory variable, and test 2 may be selected as the dependent variable. Alternatively, physical property 1 and physical property 2 may be selected as the explanatory variables, and test 1 may be selected as the dependent variable. Alternatively, physical property 1, physical property 2, and physical property 3 may be selected as the explanatory variables, and test 1 may be selected as the dependent variable. In other words, two or more types of explanatory variables and one type of dependent variable may be selected.

[0033] 4, the production date is selected every day, but the method of selecting the production date is not limited to this. For example, 5 / 13 and 5 / 15 may be selected every other day from 5 / 13 to 5 / 16 in dataset Ds, 5 / 20 and 5 / 22 may be selected every other day from 5 / 20 to 5 / 23, 5 / 27 and 5 / 29 may be selected every other day from 5 / 27 to 5 / 29, and 6 / 5 and 6 / 7 may be selected every other day from 6 / 5 to 6 / 7.

[0034] The data analysis device of this embodiment performs data analysis on the data set Ds exemplified above. In the following, to facilitate understanding of the invention, the explanatory variables and objective variables described above will be explained in a more simplified manner.

[0035] [Data analysis equipment configuration] The configuration of the data analysis device according to the first embodiment will be described with reference to FIGS.

[0036] FIG. 5 is a block diagram showing the functional configuration of the data analysis apparatus 1 according to the first embodiment.

[0037] 5, the data analysis apparatus 1 includes a data acquisition unit 10, a data interval setting unit 20, a differential value calculation unit 30, and a differential model derivation unit 40. The data analysis apparatus 1 also includes an end point value prediction unit 50 and an output unit 104. The functional configuration of the data analysis apparatus 1 is realized by executing a program stored in a storage unit 105.

[0038] The data acquiring unit 10 acquires a plurality of pieces of data from the outside. For example, the data acquiring unit 10 acquires a plurality of pieces of data through operation input by a user who uses the data analysis apparatus 1, or data input by an external device.

[0039] Each of the multiple data items is composed of one or more explanatory variables X, which are data that represent causes, and one response variable Y, which is data that represent results. Each of the explanatory variables X and the response variable Y is expressed as a physical quantity in SI base units, such as length, mass, current, temperature, or time. Note that the explanatory variables X may include things that cannot be expressed by the above physical quantities, such as people, tools, or locations. Furthermore, the multiple data items in this embodiment are represented in a time series, such as hours, minutes, days, weeks, or months. Time-series data represented in a time series is data that indicates changes over time in physical quantities, etc., and the physical quantities and time are associated with each other. The multiple data items represented in a time series may be data arranged at equal time intervals or may be data arranged at different time intervals.

[0040] Fig. 6 is a diagram showing an example of a plurality of pieces of data acquired by the data analysis apparatus 1. In (a) of Fig. 6, a plurality of pieces of data is shown in a table, and in (b) of Fig. 6, a plurality of pieces of data is shown in a graph.

[0041] FIG. 6 shows explanatory variables X and response variables Y included in each data, organized in chronological order. While FIG. 6 shows simplified data, the explanatory variables X in FIG. 6 may be input data (e.g., manufacturing condition data) input in the manufacturing process, and the response variable Y may be output data (e.g., inspection data) obtained based on intermediate products or finished products manufactured in the manufacturing process. Each data contains uncertain elements that affect the response variable Y, that is, elements that cannot be separated despite affecting the response variable Y. Uncertain elements that affect the response variable Y include, for example, noise and disturbance.

[0042] In the following, an example will be described in which each data item is composed of one explanatory variable X and one objective variable Y. The plurality of data items acquired by the data acquisition unit 10 are stored in the memory 103 of the data analysis device 1 and are also output to the data interval setting unit 20.

[0043] The data interval setting unit 20 sets a plurality of data intervals for the plurality of data output from the data acquiring unit 10. A data interval is an interval between two pieces of data with different dates and times among a plurality of pieces of data in which physical quantities and time are associated with each other, and the unit of the data interval is, for example, seconds, minutes, hours, days, weeks, etc. Each data interval has a start value, which is the data at the beginning of the data interval, and an end value, which is the data at the end of the data interval.

[0044] 7 is a diagram showing an example of multiple data intervals (i, i+1) set for multiple data. Note that i is a number (order) corresponding to each data when multiple data are arranged in chronological order. i is an integer equal to or greater than 1.

[0045] As shown in Figure 7, in a data interval (i, i+1), the i-th data at a specific date and time is the starting value, and the i+1-th data at a date and time after the specific date and time is the ending value. The i-th data and the i+1-th data each contain an explanatory variable X and a response variable Y. For example, in the first data interval (1, 2) in Figure 7, the explanatory variable X(1) = 6 and the response variable Y(1) = 17 are the starting values, and the explanatory variable X(2) = 8 and the response variable Y(2) = 22 are the ending values.

[0046] In the example shown in FIG. 7, the end value in one data interval (i, i+1) becomes the start value in the next data interval (i, i+1). Specifically, in data interval (1, 2), explanatory variable X(2)=8 becomes the end value, and in data interval (2, 3), explanatory variable X(2)=8 becomes the start value. In this way, the data interval setting unit 20 sets each data interval so that two adjacent data intervals in the time series have common data. The data interval setting unit 20 may set each data interval so that multiple data intervals are connected as a whole.

[0047] It should be noted that the data interval setting unit 20 does not necessarily have to set data intervals for data arranged in chronological order. For example, the data interval setting unit 20 may set data intervals for data arranged in chronological order, skipping intervals. It is desirable that the data intervals be set to a constant width, taking into consideration the period during which the manufacturing system is in operation, etc. How the data intervals are set may be determined by default, or may be changeable by manual input. The data interval (i, i+1) set by the data interval setting unit 20 is stored in the memory 103, and is also output to the difference value calculation unit 30 together with multiple data.

[0048] The difference value calculation unit 30 calculates a difference value for the explanatory variable X and a difference value for the response variable Y for each data interval (i, i+1) set by the data interval setting unit 20. Specifically, the difference value calculation unit 30 calculates an explanatory variable difference value ΔX, which is the difference between the start value X(i) of the explanatory variable X included in the i-th data and the end value X(i+1) of the explanatory variable X included in the i+1-th data (ΔX=X(i+1)-X(i)). The difference value calculation unit 30 also calculates a response variable difference value ΔY, which is the difference between the start value Y(i) of the response variable Y included in the i-th data and the end value Y(i+1) of the response variable Y included in the i+1-th data (ΔY=Y(i+1)-Y(i)). The difference here refers to the value obtained by subtracting the start value from the end value.

[0049] 8 is a diagram showing the explanatory variable difference value ΔX and the dependent variable difference value ΔY in each data interval (i, i+1). For example, FIG. 8 shows that the explanatory variable difference value ΔX in the data interval (1, 2) is 2 and the dependent variable difference value ΔY is 5. The multiple explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY calculated by the difference value calculation unit 30 are output to the differential model derivation unit 40.

[0050] The differential model derivation unit 40 derives a differential model M that indicates the relationship between the explanatory variable difference values ​​ΔX and the dependent variable difference values ​​ΔY, based on a plurality of explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY.

[0051] 9 is a diagram showing the relationship between the explanatory variable difference value ΔX and the dependent variable difference value ΔY. In FIG. 9, a plurality of explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY calculated by the difference value calculation unit 30 are plotted. In addition, in FIG. 9, a differential model M showing the relationship between the plurality of explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY is represented by a thick dashed line. The differential model M shown in FIG. 9 is defined, for example, by the following (Equation 1). Note that k is the number of explanatory variables X.

[0052]

number

[0053] In (Equation 1), the multiple regression coefficient β 10 , β 1k By estimating using the response variable difference value ΔY and the explanatory variable difference value ΔX, one regression model formula (formula for the difference model M) for the response variable difference value ΔY and the explanatory variable difference value ΔX can be obtained.

[0054] In the above, the objective variable difference value ΔY is defined as a linear expression of the explanatory variable difference value ΔX as the differential model M. However, the objective variable difference value ΔY can also be defined as a product-sum term of the explanatory variable difference value ΔX, and the differential model M can be defined by the following (Equation 2).

[0055]

number

[0056] Furthermore, the response variable difference value ΔY can be defined by an arbitrary polynomial of the explanatory variable difference value ΔX, and the difference model M can be defined by the following (Equation 3). Note that the relationship between the degrees r, p, and q of the polynomials is r>p,q,...>1. Polynomials represent a general-purpose expression that includes logarithms, exponentials, trigonometric functions, etc.

[0057]

number

[0058] The differential model M derived by the differential model derivation unit 40 is stored in the memory 103 and is also output to the end point value prediction unit 50.

[0059] The end point value prediction unit 50 uses the differential model M to predict at least one of the end point value X(i+1) of the explanatory variable X and the end point value Y(i+1) of the response variable Y in future data.

[0060] First, an example will be described in which the endpoint value prediction unit 50 predicts the endpoint value X(i+1) of the explanatory variable X in future data. Predicting the endpoint value X(i+1) of the input explanatory variable X is useful for bringing output data obtained based on intermediate products or finished products manufactured in a manufacturing process closer to a target value T (not shown), which is the output data that is originally desired. This prediction makes it possible to derive manufacturing conditions that will cause the inspection data of the intermediate products or finished products to reach the target value T (desired value).

[0061] For example, the end point value prediction unit 50 inputs the end point values ​​X(i+1) and Y(i+1) of the explanatory variable X and the response variable Y included in the past data into the differential model M as the start point values ​​X(i) and Y(i) of the explanatory variable X and the response variable Y in the future data. Then, the end point value prediction unit 50 assigns the target value T of the response variable Y in the future data to the differential model M to determine the end point value X(i+1) of the explanatory variable X in the future data. More specifically, the end point value prediction unit 50 determines the end point value X(i+1) of the explanatory variable X in the future data in the differential model M when the end point value Y(i+1) of the response variable Y in the future data is closest to the target value T.

[0062] In this case, after determining the end point value Y(i+1) of the response variable Y that is closest to the target value T, the differential model M may be reverse-calculated to determine the end point value X(i+1) of the explanatory variable X. The end point value Y(i+1) of the response variable Y that is closest to the target value T can be determined by calculating the distance between the target value T and the end point value Y(i+1) of the response variable Y. Alternatively, the differential model M may be varied to determine the end point value X(i+1) of the explanatory variable X when the end point value Y(i+1) of the response variable Y is closest to the target value T. The end point value Y(i+1) of the response variable Y that is closest to the target value T can be determined by calculating the distance between the target value T and the end point value Y(i+1) of the response variable Y. The target value T is a value set by the user and stored in the storage unit 105.

[0063] Next, an example will be described in which the end point value prediction unit 50 predicts the end point value Y(i+1) of the response variable Y in future data. Predicting the end point value Y(i+1) of the response variable Y, which is the output, is useful for understanding the output relative to the input.

[0064] For example, the end point value prediction unit 50 inputs the end point values ​​X(i+1) and Y(i+1) of the explanatory variable X and the response variable Y included in the past data into the differential model M as the start point values ​​X(i) and Y(i) of the explanatory variable X and the response variable Y in the future data. Then, the end point value prediction unit 50 inputs the end point value X(i+1) of the explanatory variable X in the future data into the differential model M, thereby obtaining the end point value Y(i+1) of the response variable Y in the future data.

[0065] The end point value X(i+1) of the explanatory variable X or the end point value Y(i+1) of the response variable Y predicted by the end point value prediction unit 50 is stored in the memory 103. In addition, the end point value X(i+1) of the explanatory variable X or the end point value Y(i+1) of the response variable Y may be output to the output unit 104 and displayed on the output unit 104.

[0066] The output unit 104 is a display device such as a liquid crystal panel, and displays the end point value X(i+1) of the explanatory variable X or the end point value Y(i+1) of the objective variable Y output from the end point value prediction unit 50. The output unit 104 may also display a plurality of data, data intervals, differential model M, and target value T.

[0067] FIG. 10 is a diagram schematically showing the effect of the data analysis device 1. In FIG.

[0068] Fig. 10(a) shows uncertainty factors included in multiple data. Fig. 10(b) shows the difference between the target value T and the end point value of the response variable Y predicted by a general multiple regression model. Fig. 10(c) shows the difference between the target value T and the end point value of the response variable Y predicted by the method disclosed in Patent Document 1. Fig. 10(d) shows the difference between the target value T and the end point value Y(i+1) of the response variable Y predicted by this embodiment. The horizontal axes of Figs. 10(b) to (d) show numbers assigned in chronological order corresponding to the time-series data, and the vertical axes show the difference between the end point value and the target value T.

[0069] As shown in this figure, the data analysis device 1 according to this embodiment produces a smaller difference between the end value Y(i+1) of the dependent variable Y and the target value T than a typical multiple regression model and the method described in Patent Document 1. Unlike a typical multiple regression model and the method described in Patent Document 1, the data analysis device 1 creates a differential model M based on the difference between the start value X(i) and the end value X(i+1) of the dependent variable X and the difference between the start value Y(i) and the end value Y(i+1) of the dependent variable Y. By generating a model based on the difference between the start value and the end value in this way, it is possible to cancel at least some of the uncertain factors contained in the data. Therefore, it is possible to derive a differential model M in which the influence of the uncertain factors is suppressed. This enables accurate analysis of multiple data sets.

[0070] Furthermore, by using the differential model M, it is possible to accurately predict the end point value X(i+1) of the explanatory variable X that is most suitable for bringing the end point value Y(i+1) of the dependent variable Y closer to the target value T. Furthermore, by using the differential model M, it is possible to accurately predict the end point value Y(i+1) of the dependent variable Y, which is the output, when the end point value X(i+1) of the explanatory variable X is input.

[0071] [Example of data analysis method] An example of the data analysis method according to the first embodiment will be described with reference to Fig. 11. In this example, an example will be described in which an end point value X(i+1) of an explanatory variable X that is most suitable for bringing an end point value Y(i+1) of a response variable Y closer to a target value T will be predicted.

[0072] FIG. 11 is a flowchart showing an example of a data analysis method according to the first embodiment.

[0073] First, the data acquisition unit 10 of the data analysis device 1 acquires a plurality of data as shown in FIG. 6 (step S11).

[0074] Next, the data interval setting unit 20 organizes the plurality of data in chronological order (step S12). Specifically, the data interval setting unit 20 arranges the plurality of data in ascending order of time. For example, the plurality of data organized in chronological order is assigned data numbers in order instead of dates and times. When extracting some data from the plurality of data, it is desirable to extract the data so that the date and time intervals are equal. The data interval setting unit 20 sets a plurality of data intervals (i, i+1) as shown in FIG. 7 for the plurality of data organized in chronological order (step S13).

[0075] Note that step S12 may be omitted if the data organized in chronological order is input in advance to the data acquisition unit 10 or the data interval setting unit 20. Also, if the difference value calculation unit 30 described below has the functions of the data interval setting unit 20, steps S12 and S13 may be executed by the difference value calculation unit 30.

[0076] The difference value calculation unit 30 sets an explanatory variable X and a response variable Y for the plurality of data based on the setting conditions determined by the user (step S14). Note that the explanatory variable X and the response variable Y may be set in advance for the plurality of data input to the data acquisition unit 10, or may be set by the data interval setting unit 20.

[0077] Next, the difference value calculation unit 30 calculates, in each data interval (i, i+1), an explanatory variable difference value ΔX, which is the difference between the start value X(i) of the explanatory variable X included in the ith data and the end value X(i+1) of the explanatory variable X included in the (i+1)th data. Furthermore, the difference value calculation unit 30 calculates, in each data interval (i, i+1), an objective variable difference value ΔY, which is the difference between the start value Y(i) of the objective variable Y included in the ith data and the end value Y(i+1) of the objective variable Y included in the (i+1)th data (step S15; see FIG. 8).

[0078] Next, the differential model derivation unit 40 derives a differential model M that indicates the relationship between the explanatory variable difference values ​​ΔX and the dependent variable difference values ​​ΔY based on the multiple explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY (step S16). The definition of the differential model M is as described with reference to FIG.

[0079] Next, the end point value prediction unit 50 inputs the end point values ​​X(i+1) and Y(i+1) of the explanatory variable X and the response variable Y included in the past data into the differential model M as the start point values ​​X(i) and Y(i) of the explanatory variable X and the response variable Y in the future data (step S17). Then, the end point value prediction unit 50 assigns the target value T of the response variable Y in the future data to the differential model M to obtain the end point value X(i+1) of the explanatory variable X in the future data (step S18). More specifically, the end point value prediction unit 50 obtains the end point value X(i+1) of the explanatory variable X in the future data in the differential model M when the end point value Y(i+1) of the response variable Y in the future data is closest to the target value T.

[0080] The output unit 104 displays the end point value X(i+1) of the explanatory variable X predicted by the end point value prediction unit 50 (step S19). By executing these steps S11 to S19, a plurality of data can be analyzed with high accuracy.

[0081] [Another example of data analysis method] Another example of the data analysis method according to the first embodiment will be described with reference to Fig. 12. In this other example, when an end point value X(i+1) of a predetermined explanatory variable X is input, an end point value Y(i+1) of a response variable Y, which is an output, is predicted.

[0082] Fig. 12 is a flowchart showing another example of the data analysis method according to Embodiment 1. Steps S11 to S16 are the same as those in the data analysis method of Fig. 11, and therefore description thereof will be omitted.

[0083] In this example, the end point value prediction unit 50 inputs the end point values ​​X(i+1) and Y(i+1) of the explanatory variable X and the response variable Y included in the past data into the differential model M as the start point values ​​X(i) and Y(i) of the explanatory variable X and the response variable Y in the future data (step S17). Then, the end point value prediction unit 50 inputs the end point value X(i+1) of the explanatory variable X in the future data into the differential model M, thereby obtaining the end point value Y(i+1) of the response variable Y in the future data (step S18a).

[0084] The output unit 104 displays the end point value Y(i+1) of the response variable Y predicted by the end point value prediction unit 50 (step S19a). By executing these steps S11 to S19, a plurality of data can be analyzed with high accuracy.

[0085] [Effects, etc.] A data analysis apparatus 1 according to this embodiment is an apparatus for analyzing a plurality of data sets each composed of one or more explanatory variables X and one response variable Y, and includes a data acquisition unit 10, a differential value calculation unit 30, and a differential model derivation unit 40. The data acquisition unit 10 acquires a plurality of data sets. The differential value calculation unit 30 calculates, in a data interval (i, i+1) in which the ith data set at a predetermined date and time among the plurality of data sets is a start value and the (i+1)th data set at a date and time after the predetermined date and time is an end value, an explanatory variable differential value ΔX, which is the difference between the start value X(i) of the explanatory variable X included in the ith data set and the end value X(i+1) of the explanatory variable X included in the (i+1)th data set, and an objective variable differential value ΔY, which is the difference between the start value Y(i) of the objective variable Y included in the ith data set and the end value Y(i+1) of the objective variable Y included in the (i+1)th data set. The differential model derivation unit 40 derives a differential model M that indicates the relationship between the explanatory variable difference values ​​ΔX and the dependent variable difference values ​​ΔY, based on a plurality of explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY.

[0086] In this way, by generating the differential model M based on the explanatory variable difference value ΔX and the response variable difference value ΔY in each data interval (i, i+1), it is possible to cancel at least some of the uncertain factors contained in the data. Therefore, it is possible to derive a differential model M in which the uncertain factors are suppressed. This allows for accurate analysis of multiple data sets.

[0087] In addition, the data analysis device 1 may further include a data interval setting unit 20 that sets a data interval (i, i+1) for the multiple data acquired by the data acquisition unit 10, and the difference value calculation unit 30 may calculate the explanatory variable difference value ΔX and the objective variable difference value ΔY for each data interval (i, i+1) set by the data interval setting unit 20.

[0088] According to this configuration, it is possible to appropriately set the data interval (i, i+1) and derive an appropriate differential model M based on the explanatory variable difference value ΔX and the objective variable difference value ΔY for each set data interval (i, i+1). This allows for accurate analysis of multiple data.

[0089] The data analysis device 1 may further include an end point value prediction unit 50 that uses a differential model M to predict at least one of the end point value X(i+1) of the explanatory variable X and the end point value Y(i+1) of the objective variable Y in future data.

[0090] According to this configuration, the differential model M in which uncertain factors are suppressed can be used to accurately predict the end value X(i+1) of the explanatory variable X or the end value Y(i+1) of the response variable Y.

[0091] Furthermore, the end point value prediction unit 50 may input the end point values ​​X(i+1) and Y(i+1) of the explanatory variable X and the response variable Y contained in the past data into the differential model M as the start point values ​​X(i) and Y(i) of the explanatory variable X and the response variable Y in the future data, respectively, and assign the target value T of the response variable Y in the future data to the differential model M, thereby obtaining the end point value X(i+1) of the explanatory variable X in the future data.

[0092] This makes it possible to accurately predict the end point value X(i+1) of the explanatory variable X that is suitable for bringing the end point value Y(i+1) of the response variable Y closer to the target value T.

[0093] In addition, the end point value prediction unit 50 may calculate the end point value X(i+1) of the explanatory variable X in future data in the differential model M when the end point value Y(i+1) of the objective variable Y in future data is the value closest to the target value T.

[0094] This makes it possible to easily predict the end point value X(i+1) of the explanatory variable X with high accuracy.

[0095] Furthermore, the end point value prediction unit 50 may input the end point values ​​X(i+1) and Y(i+1) of the explanatory variable X and the dependent variable Y contained in the past data into the differential model M as the start point values ​​X(i) and Y(i) of the explanatory variable X and the dependent variable Y in the future data, and may also input the end point value X(i+1) of the explanatory variable X in the future data into the differential model M, thereby obtaining the end point value Y(i+1) of the dependent variable Y in the future data.

[0096] This makes it possible to accurately predict the end point value Y(i+1) of the response variable Y corresponding to the end point value X(i+1) of the explanatory variable X.

[0097] The data analysis apparatus 1 may further include an output unit 104 that displays at least one of the end point value X(i+1) of the explanatory variable X and the end point value Y(i+1) of the response variable Y in future data.

[0098] This allows the output unit 104 to notify the user of information regarding the end point value X(i+1) of the explanatory variable X or the end point value Y(i+1) of the response variable Y.

[0099] The data analysis method according to this embodiment is a method for analyzing a plurality of data items each consisting of one or more explanatory variables X and one response variable Y. This data analysis method includes the steps of: acquiring a plurality of data; calculating an explanatory variable difference value ΔX, which is the difference between the start value X(i) of the explanatory variable X included in the i-th data and the end value X(i+1) of the explanatory variable X included in the i+1-th data, and an objective variable difference value ΔY, which is the difference between the start value Y(i) of the objective variable Y included in the i-th data and the end value Y(i+1) of the objective variable Y included in the i+1-th data, in a data interval (i, i+1) having the ith data from the plurality of data at a predetermined date and time as a start value and the i+1th data at a date and time after the predetermined date and time as an end value; deriving a differential model M that indicates the relationship between the explanatory variable difference values ​​ΔX and the objective variable difference values ​​ΔY based on the plurality of explanatory variable difference values ​​ΔX and the objective variable difference values ​​ΔY; and predicting at least one of the end value X(i+1) of the explanatory variable X and the end value Y(i+1) of the objective variable Y in future data using the differential model M.

[0100] In this way, by generating a differential model M based on the explanatory variable difference value ΔX and the response variable difference value ΔY in each data interval (i, i+1), it is possible to cancel at least some of the uncertain factors contained in the data. Therefore, it is possible to derive a differential model M in which the uncertain factors are suppressed. This allows multiple data to be analyzed with high accuracy.

[0101] The program according to this embodiment is a program for causing a computer to execute the above-described data analysis method.

[0102] By running this program, multiple data can be analyzed with high accuracy.

[0103] (Embodiment 2) [Data analysis equipment configuration] The configuration of a data analysis apparatus 1A according to a second embodiment will be described with reference to FIGS. 13 to 17. In the second embodiment, an example will be described in which the start value of a data section is replaced with a predetermined standard value to find the difference. Furthermore, in the second embodiment, an example will be described in which an end value found using the predetermined standard value is compared with the end value found in the first embodiment to select a desirable end value. Note that the description of the same configuration as in the first embodiment will be omitted or simplified.

[0104] FIG. 13 is a block diagram showing a functional configuration of a data analysis apparatus 1A according to the second embodiment.

[0105] 13, the data analysis apparatus 1A includes a standardized differential value calculation unit 30A, a standardized differential model derivation unit 40A, a standardized endpoint value prediction unit 50A, and a selection unit 60A. The data analysis apparatus 1A also includes the data acquisition unit 10, data interval setting unit 20, differential value calculation unit 30, differential model derivation unit 40, endpoint value prediction unit 50, and output unit 104 shown in the first embodiment.

[0106] The data acquiring unit 10 acquires a plurality of data items through, for example, operation input by a user who uses the data analysis apparatus 1A, or data input by an external device.

[0107] 14 is a diagram showing an example of a plurality of data sets acquired by the data analysis apparatus 1A. In FIG. 14, explanatory variables X and response variables Y included in each data set are shown organized in chronological order. Each data set contains uncertain elements that affect the response variable Y, that is, elements that cannot be measured despite having an effect on the response variable Y.

[0108] The data interval setting unit 20 sets a plurality of data intervals for the plurality of data output from the data acquisition unit 10.

[0109] 15 is a diagram showing an example of multiple data intervals (i, i+1) set for multiple pieces of data. As shown in FIG. 15, in the data interval (i, i+1), the i-th piece of data at a predetermined date and time is the starting value, and the i+1-th piece of data at a date and time after the predetermined date and time is the ending value.

[0110] The standardized difference value calculation unit 30A calculates a difference value for the explanatory variable X and a difference value for the response variable Y for each data interval (i, i+1) set by the data interval setting unit 20. In this second embodiment, the start value in the data interval (i, i+1) is set as a predetermined standardized value Sx or Sy, and the difference value is calculated.

[0111] Specifically, the standardized difference value calculation unit 30A calculates an explanatory variable difference value ΔX when the standardized value Sx is used, which is the difference between the standardized value Sx of the explanatory variable X included in the i-th data and the end point value X(i+1) of the explanatory variable X included in the i+1-th data (ΔX=X(i+1)-Sx). The standardized difference value calculation unit 30A also calculates an explanatory variable difference value ΔY when the standardized value Sy is used, which is the difference between the standardized value Sy of the objective variable Y included in the i-th data and the end point value Y(i+1) of the objective variable Y included in the i+1-th data (ΔY=Y(i+1)-Sy). The standardized value Sx of the explanatory variable X is the same for each data interval (i, i+1), and is set to, for example, 7.5. The standardized value Sy of the objective variable Y is also the same for each data interval (i, i+1), and is set to, for example, 27.

[0112] 16 is a diagram showing the explanatory variable difference value ΔX and the dependent variable difference value ΔY in each data interval (i, i+1). For example, FIG. 15 shows that in the data interval (1, 2), the explanatory variable difference value ΔX is 0.5 when the standard value Sx is used, and the dependent variable difference value ΔY is −5 when the standard value Sy is used. The multiple explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY when the standard values ​​are used, calculated by the standard difference value calculation unit 30A, are output to the standard difference model derivation unit 40A.

[0113] The standard differential model derivation unit 40A derives a standard differential model MA that indicates the relationship between the explanatory variable difference value ΔX and the dependent variable difference value ΔY, based on a plurality of explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY when the standard values ​​Sx and Sy are used.

[0114] 17 is a diagram showing the relationship between the explanatory variable difference value ΔX and the dependent variable difference value ΔY in the second embodiment. FIG. 17 plots a plurality of explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY when the standard values ​​Sx and Sy are used. FIG. 17 also shows, with a thick dashed line, a standard differential model MA showing the relationship between the plurality of explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY when the standard values ​​Sx and Sy are used. The definition of the standard differential model MA is the same as the definition of the differential model M in the first embodiment. The standard differential model MA derived by the standard differential model derivation unit 40A is stored in the memory 103 and is also output to the standard endpoint value prediction unit 50A.

[0115] The standard end value prediction unit 50A uses the standard difference model MA to predict at least one of the end value X(i+1) of the explanatory variable X and the end value Y(i+1) of the response variable Y in future data. The method of calculating the end value X(i+1) and the end value Y(i+1) of the response variable Y is the same as in the first embodiment.

[0116] The selection unit 60A compares the difference d (not shown) between the end point value Y(i+1) of the objective variable Y calculated by the end point value prediction unit 50 and the target value T of the objective variable Y, with the difference dA (not shown) between the end point value Y(i+1) of the objective variable Y calculated by the standard end point value prediction unit 50A and the target value T of the objective variable Y. Then, the selection unit 60A selects the end point value Y(i+1) of the objective variable Y having the smaller difference from the differences d and dA. Furthermore, the selection unit 60A selects the end point value prediction unit 50 or the standard end point value prediction unit 50A that calculated the end point value Y(i+1) of the objective variable Y having the smaller difference.

[0117] The end point value prediction unit 50 or standard end point value prediction unit 50A selected by the selection unit 60A predicts the end point value X(i+1) of the explanatory variable X based on the end point value Y(i+1) of the response variable Y selected by the selection unit 60A. The end point value X(i+1) of the explanatory variable X predicted by the end point value prediction unit 50 or standard end point value prediction unit 50A is stored in the memory 103 and output to the output unit 104.

[0118] The output unit 104 displays the end point value X(i+1) of the explanatory variable X predicted by the end point value prediction unit 50 or the standard end point value prediction unit 50A. The output unit 104 also displays the end point value Y(i+1) of the objective variable Y predicted by the end point value prediction unit 50 or the standard end point value prediction unit 50A. The output unit 104 may also display a plurality of data, data intervals, standard values ​​Sx, Sy, difference model M, standard difference model MA, and target value T.

[0119] In the data analysis device 1A of the second embodiment, a standard difference model MA is created based on the difference between the standard value Sx and the end point value X(i+1) of the explanatory variable X and the difference between the standard value Sy and the end point value Y(i+1) of the dependent variable Y. By generating a model based on the difference between the standard value and the end point value in this way, it is possible to cancel at least some of the uncertain factors contained in the data. Therefore, it is possible to derive a standard difference model MA in which the uncertain factors are suppressed. This makes it possible to accurately analyze multiple data and predict at least one of the end point value X(i+1) of the explanatory variable X and the end point value Y(i+1) of the dependent variable Y in future data.

[0120] [Example of data analysis method] An example of a data analysis method according to the second embodiment will be described with reference to FIG.

[0121] FIG. 18 is a flowchart showing an example of a data analysis method according to the second embodiment.

[0122] First, the data acquisition unit 10 of the data analysis device 1 included in the data analysis device 1A acquires a plurality of data as shown in FIG. 6 (step S11).

[0123] Next, the data interval setting unit 20 organizes the plurality of data in chronological order (step S12).Then, the data interval setting unit 20 sets a plurality of data intervals (i, i+1) as shown in FIG.

[0124] The following steps S14 to S16 are the same as those in embodiment 1. Note that steps S14 to S16 may be executed after steps S24 to S26 described below, or may be executed simultaneously with steps S24 to S26.

[0125] In steps S24 to S26, first, the standard difference value calculation unit 30A sets an explanatory variable X and a response variable Y for the multiple data organized in step S12 (step S24). Next, the standard difference value calculation unit 30A calculates, for each data interval (i, i+1), a response variable difference value ΔX when the standard value Sx is used, which is the difference between the standard value Sx of the explanatory variable X and the end point value X(i+1) of the response variable X included in the (i+1)-th data. In addition, the standard difference value calculation unit 30A calculates, for each data interval (i, i+1), a response variable difference value ΔY when the standard value Sy is used, which is the difference between the standard value Sy of the response variable Y and the end point value Y(i+1) of the response variable Y included in the (i+1)-th data (step S25; see FIG. 16).

[0126] Next, the standard differential model derivation unit 40A derives a standard differential model MA that indicates the relationship between the explanatory variable difference values ​​ΔX and the response variable difference values ​​ΔY based on the multiple explanatory variable difference values ​​ΔX and response variable difference values ​​ΔY using the standard values ​​Sx and Sy (step S26). The definition of the standard differential model MA is as described with reference to FIG.

[0127] These steps S11 to S26 generate the differential model M and the standard differential model MA. Hereinafter, an example will be described in which the differential model M and the standard differential model MA are used to predict the end value X(i+1) of the explanatory variable X in future data.

[0128] First, the end point value prediction unit 50 inputs the end point values ​​X(i+1) and Y(i+1) of the explanatory variable X and the response variable Y contained in the past data into the differential model M as the start point values ​​X(i) and Y(i) of the explanatory variable X and the response variable Y in the future data (step S37). Then, the end point value prediction unit 50 assigns the target value T of the response variable Y in the future data to the differential model M, and finds the end point value Y(i+1) of the response variable Y in the future data that is closest to the target value T (step S38).

[0129] Meanwhile, the specification end value prediction unit 50A inputs the end values ​​X(i+1) and Y(i+1) of the explanatory variable X and the response variable Y included in the past data into the specification difference model MA as the start values ​​X(i) and Y(i) of the explanatory variable X and the response variable Y in the future data (step S47). Then, the specification end value prediction unit 50A assigns the target value T of the response variable Y in the future data to the specification difference model MA, and finds the end value Y(i+1) of the response variable Y in the future data that is closest to the target value T (step S48). Steps S47 and S48 may be executed before steps S37 and S38, or may be executed simultaneously with steps S37 and S38.

[0130] Next, the selection unit 60A compares the difference d between the end point value Y(i+1) of the objective variable Y calculated by the end point value prediction unit 50 and the target value T of the objective variable Y with the difference dA between the end point value Y(i+1) of the objective variable Y calculated by the standard end point value prediction unit 50A and the target value T of the objective variable Y. Specifically, it is determined whether the difference d is smaller than the difference dA (step S50).

[0131] If the difference d is smaller than the difference dA (Yes in step S50), the selection unit 60A uses the end point value Y(i+1) of the objective variable Y calculated by the end point value prediction unit 50 to calculate the end point value X(i+1) of the explanatory variable X (step S51). The output unit 104 displays the end point value X(i+1) of the explanatory variable X calculated by the end point value prediction unit 50 (step S52).

[0132] On the other hand, if the difference d is greater than the difference dA (No in step S50), the selection unit 60A uses the end point value Y(i+1) of the objective variable Y determined by the specification end point value prediction unit 50A to determine the end point value X(i+1) of the explanatory variable X (step S53). The output unit 104 displays the end point value X(i+1) of the explanatory variable X determined by the specification end point value prediction unit 50A (step S54). By performing these steps S11 to S54, multiple data can be analyzed with high accuracy.

[0133] [Another example of data analysis method] Another example of the data analysis method according to the second embodiment will be described with reference to Fig. 19. In this other example, when an end point value X(i+1) of a predetermined explanatory variable X is input, an end point value Y(i+1) of a response variable Y, which is an output, is predicted.

[0134] Fig. 19 is a flowchart showing another example of the data analysis method according to embodiment 2. Steps S11 to S26 are the same as those in the data analysis method shown in Fig. 18, and therefore description thereof will be omitted.

[0135] In this example, the end point value prediction unit 50 inputs the end point values ​​X(i+1) and Y(i+1) of the explanatory variable X and the response variable Y included in the past data into the differential model M as the start point values ​​X(i) and Y(i) of the explanatory variable X and the response variable Y in the future data (step S37). Then, the end point value prediction unit 50 inputs the end point value X(i+1) of the explanatory variable X in the future data into the differential model M, thereby obtaining the end point value Y(i+1) of the response variable Y in the future data (step S38a).

[0136] Meanwhile, the standard end value prediction unit 50A inputs the end values ​​X(i+1) and Y(i+1) of the explanatory variable X and the response variable Y contained in the past data into the standard difference model MA as the start values ​​X(i) and Y(i) of the explanatory variable X and the response variable Y in the future data (step S47).Then, the standard end value prediction unit 50A inputs the end value X(i+1) of the explanatory variable X in the future data into the standard difference model MA to obtain the end value Y(i+1) of the response variable Y in the future data (step S48a).

[0137] The output unit 104 displays the end point values ​​Y(i+1) of both the objective variables Y predicted by the end point value prediction unit 50 and the standard end point value prediction unit 50A (step S55). By performing these steps, multiple data can be analyzed with high accuracy.

[0138] [Effects, etc.] A data analysis apparatus 1A according to this embodiment further includes a standardized difference value calculation unit 30A, a standardized difference model derivation unit 40A, and a standardized end value prediction unit 50A, in addition to the components of the data analysis apparatus 1. The standardized difference value calculation unit 30A defines predetermined standard values ​​Sx and Sy as the start values ​​in a data interval (i, i+1), and calculates an explanatory variable difference value ΔX when the standardized value Sx is used, which is the difference between the standard value Sx of an explanatory variable X included in the i-th data and the end value X(i+1) of the explanatory variable X included in the i+1-th data, and an objective variable difference value ΔY when the standardized value Sy is used, which is the difference between the standardized value Sy of an objective variable Y included in the i-th data and the end value Y(i+1) of the objective variable Y included in the i+1-th data. The standard differential model derivation unit 40A derives a standard differential model MA that indicates the relationship between the explanatory variable difference values ​​ΔX and the dependent variable difference values ​​ΔY when the standard values ​​Sx and Sy are used, based on a plurality of explanatory variable difference values ​​ΔX and dependent variable difference values ​​ΔY when the standard values ​​Sx and Sy are used. The standard end point value prediction unit 50A predicts at least one of the end point value X(i+1) of the explanatory variable X and the end point value Y(i+1) of the dependent variable Y in future data using the standard differential model MA.

[0139] In this way, by generating a model based on the difference between the standard value and the end value, it is possible to cancel at least some of the uncertainties contained in the data. Therefore, it is possible to derive a standard difference model MA in which the uncertainties are suppressed. This makes it possible to accurately analyze multiple data sets and accurately predict at least one of the end value X(i+1) of the explanatory variable X and the end value Y(i+1) of the target variable Y in future data.

[0140] The data analysis apparatus 1A may further include a selection unit 60A that compares the difference d between the end point value Y(i+1) of the objective variable Y calculated by the end point value prediction unit 50 and the target value T of the objective variable Y with the difference dA between the end point value Y(i+1) of the objective variable Y calculated by the standard end point value prediction unit 50A and the target value T of the objective variable Y, and selects the end point value Y(i+1) of the objective variable Y with the smaller difference.

[0141] This makes it possible to predict the end point value Y(i+1) of the response variable Y in future data with even greater accuracy.

[0142] Furthermore, the selection unit 60A may select the end point value prediction unit 50 or the standard end point value prediction unit 50A that has determined the end point value Y(Y+1) of the response variable Y for which the above difference is small, and the end point value prediction unit 50 or the standard end point value prediction unit 50A selected by the selection unit 60A may predict the end point value X(i+1) of the explanatory variable X based on the end point value Y(i+1) of the response variable Y selected by the selection unit 60A.

[0143] This makes it possible to predict the end value X(i+1) of the explanatory variable X in future data with even greater accuracy.

[0144] (Other embodiments) While the data analysis device and the like according to the present disclosure have been described above based on the embodiments, the present disclosure is not limited to these embodiments. As long as they do not deviate from the gist of the present disclosure, various modifications conceivable by a person skilled in the art to the embodiments and other forms constructed by combining some of the components of the embodiments are also included within the scope of the present disclosure.

[0145] Fig. 20 shows an example of the explanatory variable difference values ​​and the response variable difference values ​​calculated for the data set Ds in Fig. 3. Fig. 20 shows the difference values ​​between two variables of the same type that were produced on different days, one day apart. In another embodiment, as shown in Fig. 20, data analysis may be performed based on the explanatory variable difference values ​​for each of physical properties 1 to 3 and process condition 1, and the response variable difference values ​​for each of inspections 1 and 2.

[0146] Fig. 21 shows an example in which the explanatory variable difference values ​​and the response variable difference values ​​are calculated using standard values ​​for the data set Ds in Fig. 3. Fig. 21 shows an example in which the difference values ​​are calculated using standard values ​​for data from the data set Ds where production has started or resumed. As another embodiment, data analysis may be performed including the explanatory variable difference values ​​and the response variable difference values ​​calculated using standard values ​​such as those shown in Fig. 21.

[0147] For example, in the first embodiment, the data interval setting unit 20 sets data intervals for data arranged in chronological order, but this is not limiting. The data interval setting unit 20 may set data intervals for data arranged in chronological order, skipping intervals. FIG. 22 is a diagram illustrating another example of multiple data intervals set for multiple data. FIG. 22 illustrates an example in which data intervals are set by extracting data numbers 1, 3, 5, and 7 from the data illustrated in FIG. 2. In this case, data numbers 1, 3, 5, and 7 are rearranged in chronological order and renumbered as 1, 2, 3, and 4, and each data interval is set so that two adjacent data intervals in chronological order have common data. Specifically, in the rearranged data interval (1, 2), the end point value X(i+1) of the explanatory variable X is set to 6, and in the data interval (2, 3), the start point value X(i) of the explanatory variable X is set to 6.

[0148] The data interval setting unit 20 may also use averaged data obtained by averaging time-series data over an arbitrary interval, or may use post-processed data obtained by performing a predetermined calculation on time-series data.

[0149] Furthermore, in the first embodiment, the end point value of the previous data is used as the start point value of the next data, but the end point value of the previous data does not necessarily have to be used as the start point value of the next data. For example, when first collecting data, there is no previous data, so in this case, a standard value may be used as the start point value of the data. In other words, it is not necessary for all data used in generating the differential model M to be actual input / output data, and the differential model M may be generated using standard values ​​for some of the data.

[0150] Furthermore, in the first embodiment, an example was shown in which the multiple data are configured with one explanatory variable X and one response variable Y, but there may be two or more explanatory variables. For example, when there are two explanatory variables, the differential model M may be derived using explanatory variables X1 and X2 as the explanatory variables.

[0151] Furthermore, in the first embodiment, an example has been shown in which a plurality of data includes one kind of objective variable. However, for example, if a plurality of data has two or more kinds of objective variables, data analysis according to the present embodiment may be performed on each of the two or more kinds of objective variables.

[0152] For example, the data analysis device may be specifically configured as a computer system including a microprocessor, ROM, RAM, a hard disk drive, a display unit, a keyboard, and a mouse. A data analysis program is stored in the RAM or the hard disk drive. The microprocessor operates in accordance with the data analysis program, causing the data analysis device to achieve its functions. Here, the data analysis program is configured by combining multiple instruction codes that indicate commands to the computer to achieve a predetermined function.

[0153] Furthermore, some or all of the components constituting the data analysis device may be configured as a single system LSI (Large Scale Integration). A system LSI is an ultra-multifunctional LSI manufactured by integrating multiple components on a single chip, and specifically, is a computer system configured to include a microprocessor, ROM, RAM, etc. A computer program is stored in the RAM. The system LSI achieves its functions when the microprocessor operates in accordance with the computer program.

[0154] Furthermore, some or all of the components constituting the data analysis device may be configured as an IC card or a standalone module that can be attached to or detached from a computer. The IC card or module is a computer system composed of a microprocessor, ROM, RAM, etc. The IC card or module may include the ultra-multifunctional LSI described above. The IC card or module achieves its functions when the microprocessor operates according to a computer program. This IC card or module may be tamper-resistant.

[0155] The present disclosure may also be a data analysis method executed by the data analysis device. The data analysis method may be realized by a computer executing a data analysis program, or may be realized by a digital signal comprising the data analysis program.

[0156] Furthermore, the present disclosure may be configured as a computer-readable non-transitory recording medium containing the data analysis program or the digital signal. Examples of the recording medium include a flexible disk, a hard disk, a CD-ROM, an MO, a DVD, a DVD-ROM, a DVD-RAM, a Blu-ray (registered trademark) Disc (BD), and a semiconductor memory. The data analysis program may be configured as the digital signal recorded on a non-transitory recording medium.

[0157] The present disclosure may also be configured by transmitting the data analysis program or the digital signal via a telecommunications line, a wireless or wired communication line, a network such as the Internet, or data broadcasting, etc.

[0158] The present disclosure may also provide a computer system including a microprocessor and a memory, the memory storing a data analysis program, and the microprocessor operating in accordance with the data analysis program.

[0159] The data analysis program or the digital signal may also be implemented by another independent computer system by recording it on the non-transitory recording medium and transferring it, or by transferring the data analysis program or the digital signal via the network or the like.

[0160] The data analysis system may also be configured with a server and a terminal carried by a user that is connected to the server via a network.

[0161] According to the data analysis device and the like of the present disclosure, multiple pieces of data can be analyzed with high accuracy. [Industrial Applicability]

[0162] The data analysis device disclosed herein can be applied to data analysis such as highly accurate prediction of a response variable. Furthermore, because it can predict and calculate conditions that satisfy a target value of a response variable with high accuracy, it can be applied to data analysis such as, for example, calculating and specifying optimal manufacturing conditions using data. It can also be used, for example, to support manufacturing operations. [Explanation of symbols]

[0163] 1. 1A data analysis equipment 10 Data Acquisition Section 20 Data section setting section 30 Differential value calculation unit 30A standard difference value calculation section 40 Differential model derivation part 40A Standard Differential Model Derivation Section 50 End point value prediction section 50A standard end point value prediction section 60A selection section 101 Input section 102 Arithmetic circuit 103 memory 104 Output section 105 Storage section 105a Program 105b Temporary Data 106 databases 107 Communications Department 500 Manufacturing control equipment 900 Data Analysis System Ds dataset d, dA difference M-difference model MA standard difference model Sx, Sy standard values T target value X(i), Y(i) starting value X(i+1), Y(i+1) end point values X explanatory variables Y response variable ΔX Explanatory variable difference value ΔY: Objective variable difference value (i,i+1) data interval

Claims

1. A data analysis device for analyzing a plurality of data items each consisting of one or more explanatory variables and one response variable, a data acquisition unit that acquires the plurality of data; a difference value calculation unit that calculates, in a data section having an i-th data (i is an integer of 1 or more) at a predetermined date and time among the plurality of data as a start value and an i+1-th data at a date and time after the predetermined date and time as an end value, an explanatory variable difference value that is a difference between the start value of the explanatory variable included in the i-th data and the end value of the explanatory variable included in the i+1-th data, and an objective variable difference value that is a difference between the start value of the objective variable included in the i-th data and the end value of the objective variable included in the i+1-th data; a differential model derivation unit that derives a differential model indicating a relationship between the explanatory variable difference values ​​and the dependent variable difference values ​​based on a plurality of the explanatory variable difference values ​​and the dependent variable difference values; A data analysis device comprising:

2. further comprising a data interval setting unit that sets the data intervals for the plurality of data acquired by the data acquisition unit, the difference value calculation unit calculates the explanatory variable difference value and the objective variable difference value for each of the data intervals set by the data interval setting unit. The data analysis device according to claim 1 .

3. further comprising an end point value prediction unit that predicts at least one of the end point value of the explanatory variable and the end point value of the response variable in the future data using the differential model. The data analysis device according to claim 1 or 2.

4. the end point value prediction unit inputs the end point values ​​of the explanatory variables and the dependent variable included in the past data into the differential model as the start point values ​​of the explanatory variables and the dependent variable in the future data, and assigns a target value of the dependent variable in the future data to the differential model, thereby obtaining the end point value of the explanatory variables in the future data. The data analysis device according to claim 3 .

5. the end point value prediction unit calculates the end point value of the explanatory variable in the future data when the end point value of the dependent variable in the future data is closest to the target value in the differential model; The data analysis device according to claim 4 .

6. the end point value prediction unit inputs the end point values ​​of the explanatory variables and the dependent variable included in the past data into the differential model as the start point values ​​of the explanatory variables and the dependent variable in the future data, and inputs the end point values ​​of the explanatory variables in the future data into the differential model, thereby obtaining the end point value of the dependent variable in the future data. The data analysis device according to claim 3 .

7. further comprising an output unit that displays at least one of the end point value of the explanatory variable and the end point value of the target variable in the future data. The data analysis device according to any one of claims 3 to 6.

8. moreover, The starting point value in the data section is set to a predetermined standard value, and the value included in the i-th data is a standardized difference value calculation unit that calculates an explanatory variable difference value when the standardized value is used, the explanatory variable difference value being the difference between the standardized value of the explanatory variable included in the i data and the end point value of the explanatory variable included in the i+1 data, and an objective variable difference value when the standardized value is used, the objective variable difference value being the difference between the standardized value of the objective variable included in the i data and the end point value of the objective variable included in the i+1 data; a standardized differential model derivation unit that derives a standardized differential model that indicates a relationship between the explanatory variable difference values ​​and the dependent variable difference values ​​when the standardized values ​​are used, based on the plurality of explanatory variable difference values ​​and the dependent variable difference values ​​when the standardized values ​​are used; a standardized end-point value prediction unit that predicts at least one of the end-point values ​​of the explanatory variables and the end-point values ​​of the response variables in the future data using the standardized difference model; The data analysis device according to any one of claims 3 to 7, comprising:

9. moreover, a selection unit that compares a difference between the end point value of the objective variable calculated by the end point value prediction unit and a target value of the objective variable, and a difference between the end point value of the objective variable calculated by the standard end point value prediction unit and the target value of the objective variable, and selects the end point value of the objective variable with the smaller difference; The data analysis device according to claim 8 .

10. the selection unit selects the endpoint value prediction unit or the standard endpoint value prediction unit that has calculated the endpoint value of the response variable with the smallest difference; the end point value prediction unit or the standard end point value prediction unit selected by the selection unit predicts the end point value of the explanatory variable based on the end point value of the response variable selected by the selection unit. The data analysis device according to claim 9 .

11. A data analysis method in which a computer analyzes a plurality of data consisting of one or more explanatory variables and one target variable, a step of the computer acquiring the plurality of data; a step in which the computer calculates, in a data interval having an i-th data (i is an integer of 1 or more) at a predetermined date and time among the plurality of data as a start value and an i+1-th data at a date and time after the predetermined date and time as an end value, an explanatory variable difference value which is a difference between the start value of the explanatory variable included in the i-th data and the end value of the explanatory variable included in the i+1-th data, and a dependent variable difference value which is a difference between the start value of the dependent variable included in the i-th data and the end value of the dependent variable included in the i+1-th data; deriving, by the computer, a difference model indicating a relationship between the explanatory variable difference values ​​and the dependent variable difference values ​​based on a plurality of the explanatory variable difference values ​​and the dependent variable difference values; a step of predicting, by the computer, at least one of an end-point value of the explanatory variable and an end-point value of the response variable in the future data using the differential model; Data analysis methods, including:

12. A program for causing a computer to execute the data analysis method according to claim 11.

Images