Information processing device, information processing method, and information processing program

Abstract

To provide an information processing device, an information processing method, and an information processing program capable of proposing an appropriate analysis method from among a plurality of analysis methods.SOLUTION: A processing unit 22 uses principal component analysis, one method of dimensionality reduction, as a statistical method to obtain a feature quantity by the principal component analysis for measurement data for each analysis method. The processing unit 22 calculates a correlation value of the feature quantity for each principal component of each analysis method as a value of quantification, creates a heat map 30 that visualizes the size of matrix-type numeric data in colors, and selects a necessary analysis method by eliminating one of duplicates as duplicates of information, which has a high correlation from the correlation value of the feature quantity for each principal component of each analysis method. A proposal unit 24 then sends the analysis method selected by the processing unit 22 to a user terminal 12, thereby proposing the necessary analysis method to the user from among the plurality of analysis methods.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent Document 1 discloses a measurement guide device that has a database that stores virtual measurement signals and measurement procedures obtained by simulation, a similarity search unit that extracts from the database a group of virtual measurement signals and their measurement procedures that are similar to the measurement results obtained by a measurement device, and a next point suggestion unit that selects one or more measurement procedures from the group of measurement procedures obtained by the similarity search unit and determines the next measurement point.

[0003] Patent Document 2 discloses an electronic device that selects one of a plurality of parameters generated by machine learning and uses the selected parameter to analyze spectral data indicating the spectral intensities of a plurality of spectral components in response to light reflected from a food to be analyzed. In detail, the electronic device selects one of the plurality of parameters depending on at least one of the type of food to be analyzed, the required accuracy of analysis, and the processing to be performed after the analysis. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2020-134156 [Patent Document 2] Japanese Patent Publication No. 2020-176951 Summary of the Invention [Problem to be solved by the invention]

[0005] Although there have been proposals for measurement procedures and selection of parameters for analysis, there have been no proposals for an appropriate analysis method from among multiple analysis methods.

[0006] The present invention has been made in consideration of the above-mentioned facts, and aims to provide an information processing device, an information processing method, and an information processing program that are capable of proposing an appropriate analysis method from among multiple analysis methods. [Means for solving the problem]

[0007] The information processing device according to the first aspect includes: Based on different measurement principles an acquisition unit that acquires measurement data measured by a plurality of analysis methods; a quantification processing unit that quantifies the measurement data acquired by the acquisition unit using a predetermined statistical method; and Based on different measurement principles and a suggestion unit that suggests a necessary analysis method from among a plurality of analysis methods.

[0008] According to the first aspect, the acquisition unit Based on different measurement principles Measurement data measured using a plurality of analysis methods is acquired.

[0009] The quantification processing unit quantifies the measurement data acquired by the acquisition unit using a predetermined statistical method.

[0010] Then, the proposing unit uses the processing result of the quantification processing unit to: Based on different measurement principles The necessary analysis method is proposed from among multiple analysis methods, which makes it possible to propose an appropriate analysis method from among multiple analysis methods.

[0011] In the information processing device of the second aspect, in the information processing device of the first aspect, the quantification processing unit obtains features for each analysis method using principal component analysis as the statistical method, and calculates the correlation value of the features for each analysis method as the quantification value.

[0012] According to the second aspect, it is possible to eliminate unnecessary analysis methods based on the correlation value.

[0013] An information processing device according to a third aspect is the information processing device according to the second aspect, wherein the quantification processing unit uses the correlation value to create a heat map that visualizes the size of matrix-type numeric data with colors.

[0014] According to the third aspect, the correlation of the feature amounts for each analysis method can be visually confirmed.

[0015] An information processing device according to a fourth aspect is the information processing device according to the second or third aspect, in which the quantification processing unit further calculates a correlation value between the feature amount and a performance value of the measurement object.

[0016] According to the fourth aspect, it is possible to propose an optimal analysis method.

[0017] An information processing method according to a fifth aspect includes: Based on different measurement principles Measurement data measured by a plurality of analysis methods is acquired, the measurement data is quantified using a predetermined statistical method, and the results of the quantification are used to Based on different measurement principles The necessary analysis method is proposed from among multiple analysis methods.

[0018] According to the fifth aspect, it is possible to provide an information processing method capable of proposing an appropriate analysis method from among a plurality of analysis methods.

[0019] An information processing program according to a sixth aspect includes: Based on different measurement principles Measurement data measured by a plurality of analysis methods is acquired, the measurement data is quantified using a predetermined statistical method, and the results of the quantification are used to Based on different measurement principles A process is executed to propose a necessary analysis method from among a plurality of analysis methods.

[0020] According to the sixth aspect, it is possible to provide an information processing program capable of proposing an appropriate analysis method from among a plurality of analysis methods. [Effects of the Invention]

[0021] As described above, according to the present invention, it is possible to provide an information processing device, an information processing method, and an information processing program that are capable of proposing an appropriate analysis method from among a plurality of analysis methods. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a diagram showing a schematic configuration of an information processing system according to an embodiment of the present invention; [Figure 2] FIG. 2 is a block diagram showing the configuration of the main electrical systems of a user terminal and a cloud server in the information processing system according to the present embodiment. [Figure 3] 10A and 10B are diagrams showing an example of determining feature amounts by performing principal component analysis on a microscope image and X-ray diffraction. [Figure 4] This is to explain the heat map created by the processing unit. [Figure 5] 10 is a flowchart showing an example of the flow of processing performed by a cloud server in the information processing system according to the first embodiment. [Figure 6] This is to illustrate a heat map containing performance values. [Figure 7] 10 is a flowchart showing an example of the flow of processing performed by a cloud server of an information processing system according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0023] An example of an embodiment of the present invention will be described in detail below with reference to the drawings. Note that components and processes that perform the same actions and functions are given the same reference numerals throughout the drawings, and duplicated descriptions may be omitted as appropriate. Furthermore, the present disclosure is not limited to the following embodiment, and can be implemented with appropriate modifications within the scope of the purpose of the present disclosure.

[0024] (First embodiment) An information processing system according to a first embodiment will be described below. Fig. 1 is a diagram showing a schematic configuration of the information processing system according to this embodiment.

[0025] The information processing system 10 includes a plurality of user terminals 12a, 12b, ..., 12n and a cloud server 14, which is an example of an information processing device. The plurality of user terminals 12a, 12b, ..., 12n and the cloud server 14 are connected via a network 16, such as a local area network (LAN) or the Internet. Note that, hereinafter, when referring to one user terminal, it will simply be referred to as the user terminal 12.

[0026] Each of user terminals 12a to 12n transmits measurement data related to materials measured using multiple analysis methods to cloud server 14. Then, cloud server 14 performs a process of quantifying the measurement data using a predetermined statistical method and proposing a necessary analysis method from among the multiple analysis methods using the quantification results. This allows users to avoid unnecessary analysis.

[0027] Specifically, each of the multiple user terminals 12a, 12b, ..., 12n is operated by a multiple number of different users. The users input measurement data obtained by measuring a material sample that they wish to analyze into the user terminal 12 that they operate. The users then operate the user terminal 12 to transmit the measurement data to a cloud server 14, which will be described later. The user terminal 12 transmits the measurement data input by the users to the cloud server 14 via a network 16.

[0028] On the other hand, the cloud server 14 has the functions of an acquisition unit 20, a processing unit 22 as an example of a quantification processing unit, a proposal unit 24, and a database 26.

[0029] The acquisition unit 20 acquires measurement data relating to materials measured by a plurality of analytical techniques from each of the plurality of user terminals 12a to 12n, and stores the data in a database 26. Examples of the plurality of analytical techniques include X-ray diffraction (XRD), small-angle X-ray scattering (SAXS), X-ray absorption fine structure (XAFS), microscopy, Raman spectroscopy, infrared spectroscopy (IR), nuclear magnetic resonance (NMR), and mass spectrometry (MS).

[0030] The processing unit 22 quantifies the measurement data stored in the database 26, and selects a necessary analysis method from among a plurality of analysis methods using the quantification results.

[0031] The proposing unit 24 transmits the analysis method selected by the processing unit 22 to the user terminal 12, thereby proposing a necessary analysis method.

[0032] Next, the main configuration of the electrical systems of the plurality of user terminals 12 and the cloud server 14 in the information processing system 10 according to this embodiment will be described.

[0033] 2 is a block diagram showing the main configuration of the electrical systems of the user terminal 12 and the cloud server 14 in the information processing system 10 according to this embodiment. Note that the user terminal 12 and the cloud server 14 basically have the configuration of a general computer, so the user terminal 12 will be described here as a representative.

[0034] As shown in FIG. 2, the user terminal 12 includes a CPU (Central Processing Unit) 12A, a ROM (Read Only Memory) 12B, a RAM (Random Access Memory) 12C, a storage 12D, and a communication I / F (Interface) unit 12E.

[0035] The CPU 12A is a central processing unit that controls the overall operation of the device by executing various programs. The ROM 12B stores various control programs, various parameters, etc. in advance. The RAM 12C is used as a work area when the CPU 12A executes various programs. The storage 12D is composed of various storage units such as a hard disk drive (HDD), a solid state drive (SSD), and a flash memory, and stores various data, application programs, etc. The communication I / F unit 12G is connectable to a network 16 such as a LAN, a WAN, or the Internet, and transmits and receives various data to and from other devices connected to the network 16. The above-mentioned units of the user terminal 12 are electrically connected to each other by a system bus 12F.

[0036] With the above configuration, the user terminal 12 uses the CPU 12A to access the ROM 12B, RAM 12C, and storage 12D, and to control the sending and receiving of communication data via the communication I / F unit 12G. Although not shown in Fig. 2, the user terminal 12 also includes an operation unit such as a keyboard, mouse, touch panel, etc., and a display unit that displays various information.

[0037] In addition, in the cloud server 14, the CPU 14A executes an information processing program stored in the ROM 12B or the storage 12D, thereby realizing the functions of the acquisition unit 20, the processing unit 22, the proposal unit 24, and the database 26 described above.

[0038] Next, a detailed description will be given of the process performed by processing unit 22 of cloud server 14 to quantify the measurement data and select a necessary analysis method from among a plurality of analysis methods using the quantification results.

[0039] The processing unit 22 quantifies the measurement data for each analysis method using a predetermined statistical method. In this embodiment, principal component analysis, which is one method of dimensionality reduction, is used as a statistical method to calculate feature quantities for each measurement data for each analysis method. That is, by performing principal component analysis on each measurement data of multiple analysis methods, multiple principal components (PC1, PC2, ...) of the analysis results are calculated as feature quantities. For example, as shown in FIG. 3, principal component analysis is performed on a microscopic image as an example of a statistical method to calculate the analysis results as feature quantities for the microscopic image. Furthermore, principal component analysis is performed on X-ray diffraction as an example of a statistical method to calculate multiple principal components (PC1, PC2, ...) of the analysis results as feature quantities. FIG. 3 is a diagram showing an example of calculating feature quantities by performing principal component analysis on each of a microscopic image and X-ray diffraction.

[0040] Furthermore, the processing unit 22 calculates the correlation value of the feature amount for each principal component of each analysis method as a quantification value, and creates a heat map in which the size of the matrix-type numeric data is visualized with color. For example, as shown in Fig. 4, by performing principal component analysis for each of analysis methods 1 to 4, the first to tenth principal components (PC1 to PC10) are arranged in a 40 x 40 matrix, and the correlation value for each principal component is calculated to create the heat map 30. Note that Fig. 4 is used to explain the heat map 30 created by the processing unit 22.

[0041] Furthermore, the processing unit 22 determines the correlation values ​​of the feature quantities for each principal component of each analysis method to be highly correlated (for example, correlations above a predetermined threshold) as duplicate information, and selects the necessary analysis method by eliminating one of the duplicates. That is, a high correlation means that the amount of information in the principal component of one analysis method can be explained by the amount of information in the principal component of the other analysis method, so the necessary analysis method can be selected by extracting only the principal component of one analysis method.

[0042] The proposing unit 24 then transmits the analysis method selected by the processing unit 22 to the user terminal 12 and displays the selection result on the user terminal 12, thereby proposing a necessary analysis method from among the multiple analysis methods to the user. This makes it possible to propose an appropriate analysis method from among the multiple analysis methods.

[0043] Next, specific processing performed by the cloud server 14 of the information processing system 10 according to this embodiment configured as described above will be described. Fig. 5 is a flowchart showing an example of the flow of processing performed by the cloud server 14 of the information processing system 10 according to this embodiment. Note that the processing in Fig. 5 starts, for example, when measurement data is uploaded from the user terminal 12 to the cloud server 14 and an instruction to propose an analysis method is issued.

[0044] In step 100, the CPU 14A acquires the measurement data, and the process proceeds to step 102. That is, the processing unit 22 reads out the measurement data that the acquisition unit 20 acquires from the user terminal 12 and stores in the database .

[0045] In step 102, the CPU 14A calculates the feature amounts and proceeds to step 104. That is, the processing unit 22 performs principal component analysis on the measurement data of each of the plurality of analysis methods, thereby calculating a plurality of principal components of the analysis results as feature amounts.

[0046] In step 104, the CPU 14A determines whether there are measurement results from other analysis methods. If the determination is affirmative, the process returns to step 100 to repeat the above-described process, and if the determination is negative, the process proceeds to step 106.

[0047] In step 106, the CPU 14A calculates a correlation value for each calculated feature amount, and the process proceeds to step 108. That is, the processing unit 22 calculates the correlation value of the feature amount for each principal component of each analysis method as a quantification value.

[0048] In step 108, CPU 14A creates a heat map 30 and proceeds to step 110. That is, processing unit 22 arranges correlation values ​​in a matrix as calculated feature quantities for each analysis method, and creates a heat map 30 in which the magnitude of the correlation values ​​is visualized with colors.

[0049] In step 110, the CPU 14A extracts a necessary analysis method based on the correlation value, and proceeds to step 112. That is, the processing unit 22 determines that highly correlated correlation values ​​of feature amounts for each principal component of each analysis method (for example, correlation values ​​equal to or greater than a predetermined threshold) are duplicated information, and selects a necessary analysis method by eliminating one of the duplicated information.

[0050] In step 112, the CPU 14A outputs the extracted analysis method and the created heat map 30, and then ends the series of processes. That is, the suggestion unit 24 transmits the analysis method selected by the processing unit 22 to the user terminal 12 and displays the selection result on the user terminal 12, thereby suggesting to the user a necessary analysis method from among multiple analysis methods. This makes it possible to suggest an appropriate analysis method from among multiple analysis methods.

[0051] (Second embodiment) Next, an information processing system according to a second embodiment will be described. Note that since the configuration itself is the same as that of the first embodiment, detailed description will be omitted and only the differences will be described.

[0052] In the above embodiment, the processing unit 22 creates the heat map 30 by calculating the correlation value for each principal component of the analysis method, but in this embodiment, the processing unit 22 further calculates the correlation value with the objective variable, such as the performance value of the object to be measured, and creates the heat map 30.

[0053] For example, as shown in Fig. 6, the first to tenth principal components (PC1 to PC10) of the principal component analyses of analysis methods 1 to 4 and the performance values ​​are arranged in a 41 x 41 matrix, and the correlation values ​​are calculated to create a heat map 30. Note that Fig. 6 is for explaining the heat map 30 including the performance values.

[0054] In this embodiment, an analysis method that has a high correlation with the performance value is selected and proposed to the user as an appropriate analysis method. For example, by selecting the analysis method with the highest correlation, the optimal analysis method can be proposed.

[0055] Next, specific processing performed by the cloud server 14 of the information processing system 10 according to this embodiment will be described. Fig. 7 is a flowchart showing an example of the flow of processing performed by the cloud server 14 of the information processing system 10 according to this embodiment. Note that the processing in Fig. 7 starts, for example, when measurement data is uploaded from the user terminal 12 to the cloud server 14 and an instruction to propose an analysis method is issued. Furthermore, the same processing as in Fig. 5 will be described using the same reference numerals.

[0056] In step 100, the CPU 14A acquires the measurement data, and the process proceeds to step 102. That is, the processing unit 22 reads out the measurement data that the acquisition unit 20 acquires from the user terminal 12 and stores in the database .

[0057] In step 102, the CPU 14A calculates the feature amounts and proceeds to step 104. That is, the processing unit 22 performs principal component analysis on the measurement data of each of the plurality of analysis methods, thereby calculating a plurality of principal components of the analysis results as feature amounts.

[0058] In step 104, the CPU 14A determines whether there are measurement results from other analysis methods. If the determination is affirmative, the process returns to step 100 to repeat the above-described process, and if the determination is negative, the process proceeds to step 105.

[0059] In step 105, the CPU 14A sets a performance value and proceeds to step 107. For example, the performance value stored in advance in the database 26 is read out and set.

[0060] In step 107, the CPU 14A calculates the correlation value including the performance value, and the process proceeds to step 109. That is, the processing unit 22 calculates the correlation value between the feature amount and the performance value of each analysis method as a quantification value.

[0061] In step 109, the CPU 14A creates a heat map 30 including the performance values, and proceeds to step 111. That is, the processing unit 22 arranges the correlation values ​​between the feature amounts of each analysis method and the performance values ​​in a matrix, and creates a heat map 30 in which the magnitude of the numerical correlation values ​​is visualized with colors.

[0062] In step 111, the CPU 14A extracts an analysis method that has a high correlation with the performance value, and proceeds to step 112. For example, an analysis method that has a correlation value equal to or greater than a predetermined value may be extracted, or an analysis method that has the highest correlation may be extracted.

[0063] In step 112, the CPU 14A outputs the extracted analysis method and the created heat map 30, and then ends the series of processes. That is, the suggestion unit 24 transmits the analysis method selected by the processing unit 22 to the user terminal 12 and displays the selection result on the user terminal 12, thereby suggesting to the user a necessary analysis method from among multiple analysis methods. This makes it possible to suggest an appropriate analysis method from among multiple analysis methods. Furthermore, if the analysis method with the highest correlation is extracted in step 111, it becomes possible to suggest the optimal analysis method.

[0064] In the above embodiment, the measurement data for each analysis method is quantified by performing principal component analysis, but the measurement data may be quantified using other statistical methods, such as PPCA, kernel PCA, ICA, LSA, LDA, UMAP, t-SNE, VAE, or other statistical methods of dimension reduction.

[0065] In the above embodiment, the database 26 is provided in the cloud server 14, and the measurement data is stored in the database 26 of the cloud server 14, but this is not limitative. For example, the database 26 may be provided in a separate server.

[0066] Furthermore, although the processing performed by the cloud server 14 in each of the above embodiments has been described as software processing performed by executing a program, this is not limited to this. For example, the processing may be performed by hardware such as a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array). Alternatively, the processing may be a combination of both software and hardware. Furthermore, if the processing is software, the program may be stored in various storage media and distributed.

[0067] Furthermore, the present invention is not limited to the above, and it goes without saying that various modifications can be made without departing from the spirit of the present invention. [Explanation of symbols]

[0068] 10 Information Processing Systems 14 Cloud Server 20 Acquisition Department 22 Processing section (quantification processing section) 24 Proposal Department 30 Heatmaps

Claims

1. An acquisition unit that acquires measurement data measured by a plurality of analysis methods based on different measurement principles; a quantification processing unit that quantifies the measurement data acquired by the acquisition unit using a predetermined statistical method; a proposal unit that proposes a necessary analysis method from among a plurality of analysis methods based on the different measurement principles, using a processing result of the quantification processing unit; An information processing device comprising:

2. The information processing apparatus according to claim 1 , wherein the quantification processing unit obtains a feature amount for each of the analysis methods by principal component analysis as the statistical method, and calculates a correlation value of the feature amount for each of the analysis methods as the quantification value.

3. The information processing device according to claim 2 , wherein the quantification processing unit uses the correlation value to create a heat map that visualizes the size of matrix-type numeric data with colors.

4. The information processing apparatus according to claim 2 , wherein the quantification processing unit further calculates a correlation value between the feature amount and a performance value of the measurement target.

5. The computer Acquire measurement data measured using multiple analytical methods based on different measurement principles, quantifies the measurement data using a predetermined statistical method; An information processing method that performs processing to propose a necessary analysis method from a plurality of analysis methods based on the different measurement principles using the results of the quantification.

6. On the computer, Acquire measurement data measured using multiple analytical methods based on different measurement principles, quantifies the measurement data using a predetermined statistical method; an information processing program for executing a process of proposing a necessary analysis method from among a plurality of analysis methods based on the different measurement principles, using the results of the quantification;

Images