Display system, display method, and display program

Abstract

To provide a display system, display method, and display program that facilitate the interpretation of the results of principal component analysis. [Solution] In an analysis system in which a server and a user terminal are connected via a network such as the Internet, the server 100 includes a principal component analysis unit 112 having a principal component acquisition unit and a score acquisition unit, a score change reception unit 113, a reconstruction data output unit 114, a difference evaluation unit 115, and a display control unit 116, which analyzes analysis target data such as spectral data, waveform data, graph data, secondary image data, and 3D image data transmitted from the user terminal, and transmits the analysis results to the user terminal.

Description

Technical Field

[0001] The present disclosure relates to a display system, a display method, and a display program.

Background Art

[0002] Patent Document 1 describes an information processing apparatus that performs principal component analysis (PCA, Principal Component Analysis) on measurement data of materials. The information processing apparatus described in Patent Document 1 performs principal component analysis on a plurality of measurement data, and generates principal components (PC, Principal Component) of the plurality of measurement data and principal component values (Principal Component Score) of the principal components. Note that the principal component value is also called the principal component score or the principal component obtained score.

[0003] When the information processing apparatus described in Patent Document 1 displays the principal component value of each of the generated plurality of measurement data on a display screen, each image representing the principal component of the plurality of measurement data is arranged obliquely with respect to the horizontal direction or the vertical direction on the display screen and displayed.

[0004] Then, for each pair of the first image and the second image among each of the images representing the principal components of the plurality of measurement data, the information processing apparatus described in Patent Document 1 displays a graph in which the principal component values of the plurality of measurement data are plotted at positions representing the intersections of the lines extending vertically from the first image and the lines extending horizontally from the second image.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] By the way, when attempting to interpret the results of principal component analysis of the data being analyzed, that is, the data being analyzed, it is necessary to understand which domain of the data being analyzed is most strongly reflected by each of the output principal components.

[0007] However, it is difficult to intuitively understand which data domain characteristics each principal component output as a result of the analysis strongly reflects. Therefore, properly interpreting the results of principal component analysis has not been easy for users.

[0008] In other words, a challenge in principal component analysis of data under analysis was that users could not easily identify data regions where the characteristics of each principal component were strongly expressed, making it difficult to interpret the analysis results. Patent Document 1 does not describe any technology capable of solving these problems.

[0009] This disclosure was made to solve these problems and aims to provide a display system, display method, and display program that can facilitate the interpretation of the results of principal component analysis. [Means for solving the problem]

[0010] The display system relating to this disclosure comprises a principal component acquisition unit, a score acquisition unit, a score change acceptance unit, a reconstructed data output unit, a difference evaluation unit, and a display control unit. The principal component acquisition unit acquires principal components, which are the results of principal component analysis of multiple data sets to be analyzed. The score acquisition unit acquires the principal component scores of the data sets to be analyzed that are to be displayed. The score change acceptance unit accepts changes to the principal component scores of the data sets to be analyzed that are to be displayed from the user. The reconstructed data output unit outputs reconstructed data, which is the data set to be analyzed that has been reconstructed based on the principal components and the changed principal component scores. The difference evaluation unit evaluates the difference between the reconstructed data and the reference data, which is the data set to be analyzed that has not been changed by the user. The display control unit displays at least one of the reconstructed data and the reference data, while highlighting the data areas that are evaluated to have a large difference.

[0011] With this configuration, users can easily identify data areas where changes due to changes in principal component scores are significant, i.e., data areas where the characteristics of each principal component are strongly expressed. As a result, the display system according to this embodiment can facilitate the interpretation of the analysis results of principal component analysis.

[0012] In the display system relating to this disclosure, the reference data may be reference reconstructed data obtained by reconstructing the data under analysis based on the principal components and the principal component scores that have not been modified by the user. With this configuration, the display system in this embodiment can more appropriately evaluate the difference between the reconstructed data and the reference data that arises from changes in the principal component score.

[0013] In the display system relating to this disclosure, the reference data may be the data to be analyzed. With this configuration, the display system in this embodiment can simplify the processing required to display the analysis results.

[0014] In the display system relating to this disclosure, the display control unit may display at least one of the reconstructed data and the reference data while color-coding the data area based on the evaluation results of the difference evaluation unit. With this configuration, users can easily see data areas with large differences.

[0015] In the display system relating to this disclosure, the display control unit may superimpose and display a heatmap that reflects the magnitude of the difference on at least one of the reconstructed data and the reference data. With this configuration, the display system relating to this disclosure allows users to understand in more detail how data changes due to changes in principal component scores.

[0016] In the display system relating to this disclosure, the display control unit may choose not to superimpose a heatmap in data areas where the difference is smaller than a predetermined size. With this configuration, the display system relating to this disclosure can display the results of principal component analysis more concisely.

[0017] In the display system relating to this disclosure, the principal component acquisition unit may acquire principal components that are the result of sparse principal component analysis. This configuration allows users to more precisely visualize the data regions where the characteristics of each principal component are strongly expressed.

[0018] In the display system relating to this disclosure, the display control unit may superimpose and display the reconstructed data and the reference data. This configuration allows users to easily see the differences between the reconstructed data and the reference data.

[0019] The method of display relating to this disclosure has the following configuration. The principal components are obtained as a result of principal component analysis of multiple data sets to be analyzed. Obtain the principal component scores of the data to be analyzed and displayed. The system accepts user requests to change the principal component scores of the analyzed data that are displayed. Based on the principal components and the modified principal component scores, the system outputs reconstructed data that reconstructs the data being analyzed. Evaluate the difference between the reconstructed data and the reference data. While emphasizing the data area according to the magnitude of the difference, display at least one of the reconstructed data and the reference data.

[0020] The display program according to the present disclosure causes a computer to execute the following operations. Obtain the principal components that are the results of principal component analysis of a plurality of analysis target data. Obtain the principal component scores of the analysis target data to be displayed. Accept from the user a change in the principal component scores of the analysis target data to be displayed. Based on the principal components and the changed principal component scores, output reconstructed data obtained by reconstructing the analysis target data. Evaluate the difference between the reconstructed data and the reference data. While emphasizing the data area according to the magnitude of the difference, display at least one of the reconstructed data and the reference data.

Advantages of the Invention

[0021] According to the present disclosure, it is possible to provide a display system, a display method, and a display program that facilitate the interpretation of the analysis results of principal component analysis.

Brief Description of the Drawings

[0022] [Figure 1] It is a block diagram showing the configuration of an analysis system according to the first embodiment. [Figure 2] It is a block diagram showing the hardware configuration of a server according to the first embodiment. [Figure 3] It is a block diagram showing the functional configuration of a server according to the first embodiment. [Figure 4] It is a block diagram showing the functional configuration of a server according to the first embodiment. [Figure 5] It is a schematic diagram for explaining the configuration of a score change reception unit according to the first embodiment. [Figure 6]This is a schematic diagram illustrating the configuration of the display control unit according to the first embodiment. [Figure 7] This is a flowchart illustrating the operation of the display system according to the first embodiment. [Figure 8] This is a schematic diagram illustrating the configuration of the display control unit according to the first embodiment. [Figure 9] This is a schematic diagram illustrating the configuration of the display control unit according to the first embodiment. [Figure 10] This is a schematic diagram illustrating the configuration of the display control unit according to the first embodiment. [Figure 11] This is a schematic diagram illustrating the configuration of the display control unit according to the first embodiment. [Figure 12] This block diagram shows the configuration of a display system according to another embodiment. [Figure 13] This block diagram shows the configuration of a display system according to another embodiment. [Modes for carrying out the invention]

[0023] <First Embodiment> (Display system configuration) The first embodiment of this disclosure will be described in detail below with reference to the drawings. First, the configuration of the display system according to this embodiment will be described in detail.

[0024] The display system according to this embodiment is a system configured as part of the analysis system according to this embodiment, and is a system for displaying the analysis results of the analysis system according to this embodiment to the user. Figure 1 is a block diagram showing the configuration of the analysis system according to the first embodiment. As shown in Figure 1, the analysis system 1 according to this embodiment has a server 100 and a user terminal 200 connected via a network N such as the Internet.

[0025] Analysis System 1 is a system for analyzing the data to be analyzed. The analysis system 1 according to this embodiment is typically provided as part of a data cloud service used in materials development and research and development, and is used as a system to promote so-called materials informatics (MI) and research and development using data science. In this case, the analysis system 1 stores various measurement data of, for example, a newly developed material. Then, based on instructions from the user, the analysis system 1 appropriately uses the stored measurement data as the data to be analyzed.

[0026] In analysis system 1, the user terminal 200 transmits the data to be analyzed to the server 100, and the server 100 analyzes the received data. The server 100 then transmits the analysis results to the user terminal 200, and the user terminal 200 displays the received analysis results. The data to be analyzed according to this embodiment is not particularly limited and can be any data that can be subjected to principal component analysis.

[0027] Examples of data analyzed by analysis system 1 include spectral data, waveform data, graph data, secondary image data, and three-dimensional image data. Examples of spectral data include those measured using methods such as nuclear magnetic resonance spectroscopy (NMR), infrared spectroscopy (IR), ultraviolet-visible spectroscopy (UV-vis), X-ray absorption spectroscopy (XAS), Raman spectroscopy, X-ray diffraction (XRD), small-angle X-ray scattering (SAXS), and mass spectrometry (MS). Furthermore, examples of two-dimensional image data include image data acquired using optical microscopes, scanning electron microscopes (SEM), transmission electron microscopes (TEM), and computed tomography (CT). Furthermore, examples of 3D image data include imaging data created by stacking tomographic images taken by computed tomography (CT), and model data created using CAD (Computer-Aided Design), etc. Wave-related data can include, for example, time-series data and displacement data. Examples of time-series data include acoustic data, vibration data, and stock price trends, but any data in which the numerical value changes with time can be used as a target. Examples of displacement data include the surface height and surface profile of a sample, but any data in which the numerical value changes with changes in coordinates and other parameters can be used as a target. Other data sources include, for example, cyclic voltammograms and gas chromatography (GC) charts. Furthermore, the data to be analyzed can also include, for example, coordinate data such as CIF (Crystallographic Information) files, or numerical data such as ingredient lists of compositions.

[0028] The user terminal 200 according to this embodiment is a terminal operated by a user, and is typically a computer device having a display device. The user terminal 200 transmits the data to be analyzed to the server 100 via the network N. The user terminal 200 then receives the analysis results of the data from the server 100 via the network N.

[0029] In this embodiment, the server 100 receives data to be analyzed from the user terminal 200 and analyzes the received data. The server 100 then transmits the analysis results to the user terminal 200 via the network N.

[0030] Figure 2 is a block diagram showing the hardware configuration of the server according to the first embodiment. As shown in Figure 2, the server 100 includes a processor 110, memory 120, storage device 130, input / output interface 140, network interface 150, and internal bus 160.

[0031] The internal bus 160 is a data transmission path for the processor 110, memory 120, storage device 130, input / output interface 140, and network interface 150 to send and receive data to and from each other. However, the method of connecting the processor 110 and the other components to each other is not limited to bus connection.

[0032] Memory 120 is the main memory, implemented using RAM (Random Access Memory), etc. Furthermore, the storage device 130 is an auxiliary storage device implemented using a hard disk, SSD (Solid State Drive), memory card, or ROM (Read Only Memory), etc. The storage device 130 stores a program for realizing a desired function.

[0033] The processor 110 is a variety of processor, such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an FPGA (field-programmable gate array). The processor 110 reads the program stored in the storage device 130 into the memory 120 and executes it, thereby performing the functions of each functional block as shown in Figures 3 and 4, which will be described later.

[0034] The input / output interface 140 is an interface for connecting the server 100 with input / output devices. For example, input devices such as keyboards and output devices such as display devices may be connected to the input / output interface 140. The network interface 150 is an interface for connecting the server 100 to the network.

[0035] The program, when loaded into a computer, includes a set of instructions (or software code) for causing the computer to perform one or more of the functions described in the embodiments. The program may be stored in a non-temporary computer-readable medium or a physical storage medium. Examples, but not limited to, include RAM, ROM, flash memory, SSD or other memory technologies, CD-ROM, DVD (digital versatile disc), Blu-ray® disc or other optical disc storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices. The program may be transmitted over a temporary computer-readable medium or a communication medium. Examples, but not limited to, include, a temporary computer-readable medium or a communication medium that includes electrically, optically, acoustically, or otherwise propagating signals.

[0036] Figures 3 and 4 are block diagrams showing the functions of the server 100 according to the first embodiment. As shown in Figure 3, the server 100 according to this embodiment includes, as functional blocks, an analysis target data acquisition unit 111, a principal component analysis unit 112, a score change reception unit 113, a reconstructed data output unit 114, a difference evaluation unit 115, and a display control unit 116. Furthermore, as shown in Figure 4, the principal component analysis unit 112 according to this embodiment includes a principal component acquisition unit 112a and a score acquisition unit 112b.

[0037] The data acquisition unit 111 acquires multiple data to be analyzed. More specifically, in this embodiment, the data acquisition unit 111 acquires multiple data to be analyzed from the user terminal 200 via the network N. The data acquisition unit 111 outputs the acquired multiple data to be analyzed to the principal component analysis unit 112.

[0038] Furthermore, the data acquisition unit 111 for analysis in this embodiment does not need to acquire multiple data sets for analysis in a single reception. For example, the data acquisition unit 111 to be analyzed according to this embodiment may acquire multiple data sets to be analyzed by acquiring the data in multiple steps.

[0039] In this case, the data acquisition unit 111 stores the acquired data in the storage device 130 each time it acquires data to be analyzed. Then, when performing principal component analysis as described later, it reads multiple data to be analyzed from the storage device 130 and outputs them to the principal component analysis unit 112. In this case, the data acquisition unit 111 does not need to read all the data to be analyzed stored in the storage device 130, but may read only the data to be analyzed specified by the user and output it to the principal component analysis unit 112. In other words, the data acquisition unit 111 may create a database containing the data to be analyzed. The data acquisition unit 111 according to this embodiment may be configured so that the user can appropriately select the data to be used for principal component analysis from the created database.

[0040] The principal component analysis unit 112 acquires multiple data points for analysis from the data acquisition unit 111. The principal component analysis unit 112 then performs principal component analysis on the acquired data points for analysis. As shown in Figure 4, the principal component analysis unit 112 includes a principal component acquisition unit 112a and a score acquisition unit 112b.

[0041] The principal component acquisition unit 112a acquires principal components, which are the results of principal component analysis of multiple data sets to be analyzed. In this embodiment, the principal component acquisition unit 112a acquires principal components by performing principal component analysis on multiple data sets to be analyzed. In other words, in this embodiment, the principal component acquisition unit 112a acquires principal components by calculating the principal components of multiple data sets to be analyzed. The principal component acquisition unit 112a outputs the acquired principal components to the reconstruction data output unit 114.

[0042] For example, the principal component acquisition unit 112a may convert each of the data to be analyzed into vector data. The principal component acquisition unit 112a may then perform principal component analysis on the converted vector data. Furthermore, if the data to be analyzed is spectral data, time-series data, waveform data, or graph data, the principal component acquisition unit 112a can convert the data to be analyzed into vector data having spectral values ​​or measured values ​​as components. Furthermore, if the data to be analyzed is image data, the principal component acquisition unit 112a can convert the data to be analyzed into vector data having pixel values ​​and the like as components.

[0043] The score acquisition unit 112b acquires the principal component scores of the data to be analyzed and displayed. More specifically, the score acquisition unit 112b first receives a specification from the user for the data to be analyzed and displayed on the user terminal 200, i.e., the data to be analyzed and displayed. Then, it calculates the principal component score for each principal component of the data to be analyzed and displayed.

[0044] However, the score acquisition unit 112b may calculate principal component scores not only for the data to be displayed, but also for all the data on which principal component analysis has been performed. The score acquisition unit 112b may then store the calculated principal component scores in the storage device 130. In this case, when the score acquisition unit 112b receives a specification from the user for the data to be analyzed to be displayed, it reads the corresponding principal component scores from the storage device 130. The score acquisition unit 112b then outputs the read principal component scores to the reconstructed data output unit 114 and the display control unit 116.

[0045] In this embodiment, the principal component acquisition unit 112a and the score acquisition unit 112b acquire the principal component and the principal component score by performing calculations, but the configuration of the principal component acquisition unit 112a and the score acquisition unit 112b according to this disclosure is not limited thereto. For example, the principal component acquisition unit 112a and the score acquisition unit 112b relating to this disclosure may acquire the principal component and the principal component score from an external device.

[0046] The score change reception unit 113 receives requests from users to change the principal component scores. The score change reception unit 113 outputs the principal component scores changed by the user to the reconstructed data output unit 114. The score change reception unit 113 may, for example, display a control screen on the user terminal 200 to receive changes to the principal component score from the user.

[0047] Figure 5 is a schematic diagram illustrating the configuration of the score change reception unit according to the first embodiment. More specifically, Figure 5 is an example of a control screen displayed on the user terminal 200 to receive changes to the principal component score from the user. As shown in Figure 5, the score change reception unit 113 may display a control screen P1 on the user terminal 200 that shows the principal component identifier P11, the bar P12 indicating the magnitude of the principal component score, and the principal component score value P13 side by side.

[0048] The principal component identifiers P11 are, for example, PC1, PC2, and PC3, as shown in Figure 5, and these identifiers represent the first principal component, the second principal component, and the third principal component, respectively.

[0049] Bar P12 visually represents the magnitude of the principal component score corresponding to the adjacent identifier P11. Here, bar P2 is configured so that the magnitude of the principal component score can be changed by dragging it by the user.

[0050] The principal component score value P13 indicates the principal component score value corresponding to the adjacent identifier P11 and bar P12. If the magnitude of the principal component score is changed by dragging bar P12, the value P13 displays the magnitude of the changed principal component score. Furthermore, the value P13 may be configured to allow the user to input a specific numerical value. In this case, the score change reception unit 113 changes the display of bar P12 according to the specific numerical value entered for value P13.

[0051] The score change reception unit 113 accepts the principal component score values ​​that have been changed by the user by manipulating the bar P12 or the value P13 on the control screen P1, as principal component scores changed by the user.

[0052] Let's return to the explanation of Figure 3. The reconstructed data output unit 114 outputs reconstructed data, which is the data to be analyzed, based on the principal components obtained from the principal component analysis and the principal component scores modified by the user. In other words, the reconstructed data output unit 114 outputs simulation data of the data to be analyzed, with the principal component scores modified.

[0053] Furthermore, the reconstructed data output unit 114 outputs reference reconstructed data based on the principal components and the principal component scores that have not been modified by the user. As will be described in detail later, the reference reconstructed data is data used for comparison with the reconstructed data described above.

[0054] Here, as an example of how the reconstruction data output unit 114 outputs reconstruction data, we will describe the case where the data to be analyzed is infrared absorption spectrum data. In this example, each infrared absorption spectrum is measured at 400 cm⁻¹. -1From 3500 cm -1 Measurement range up to 1 cm -1 The absorbance of the sample was measured at intervals of [specified interval], and the absorbance is normalized so that the maximum absorbance of each spectral data is set to 1. Furthermore, in this example, the aforementioned principal component analysis unit 112 outputs up to the 10th principal component of the infrared absorption spectrum being analyzed, and it is assumed that principal component scores are obtained for each component.

[0055] JPEG2026067010000002.jpg40166

[0056] JPEG2026067010000003.jpg40166

[0057] By the method described above, the reconstructed data output unit 114 can simulate the analysis data with modified principal component scores. The methods described above can be applied not only to infrared absorption spectra, but to any data that can be analyzed using principal component analysis. For example, if the data to be analyzed is image data, the above methods can be applied by appropriately converting the pixel values ​​and other elements contained in the image data into vector data. However, the method by which the reconstructed data output unit 114 simulates the data to be analyzed with modified principal component scores is not limited to the above; any method that allows for simulation without deviating from the purpose may be adopted.

[0058] The difference evaluation unit 115 acquires reconstructed data from the reconstructed data output unit 114. The difference evaluation unit 115 evaluates the difference between the reconstructed data and the reference data. The difference evaluation unit 115 outputs the difference evaluation result to the display control unit 116.

[0059] In this embodiment, the difference evaluation unit 115 uses the reference reconstruction data output by the reconstruction data output unit 114 as reference data. In other words, the difference evaluation unit 115 in this embodiment uses the simulation data output based on the principal components and the principal component scores that have not been modified by the user as reference data. With this configuration, the difference evaluation unit 115 according to this embodiment can more appropriately evaluate the difference between the reconstructed data and the reference data that arises due to changes in the principal component score.

[0060] However, the data that the differential evaluation unit 115 relating to this disclosure can use as reference data is not limited to reference reconstruction data. For example, the difference evaluation unit 115 in this disclosure may use the data to be analyzed acquired by the data to be analyzed unit 111 as reference data. In other words, the difference evaluation unit 115 in this disclosure may use the data to be analyzed that has not been reconstructed by the reconstructed data output unit 114 as reference data.

[0061] If the data to be analyzed is spectral data, the difference evaluation unit 115 may evaluate, for example, the difference in spectral values ​​as the difference. Furthermore, if the data to be analyzed is time-series data, waveform data, or graph data, the principal component acquisition unit 112a may evaluate the difference between the corresponding measured values ​​as a difference. Furthermore, if the data to be analyzed is image data, the difference evaluation unit 115 may evaluate, for example, the difference in pixel values ​​as the difference. Furthermore, the difference evaluation unit 115 may evaluate not only the magnitude of the difference in numerical values, but also, for example, the ratio of numerical values ​​or the rate of change as differences.

[0062] For example, the difference evaluation unit 115 may evaluate the difference between the reconstructed data and the reference data by outputting the difference between the reconstructed data and the reference data as numerical data. The difference evaluation unit 115 may also evaluate the difference between the reconstructed data and the reference data by comparing the difference between the reconstructed data and the reference data with a predetermined threshold.

[0063] The display control unit 116 obtains the difference evaluation result from the difference evaluation unit 115. The display control unit 116 displays at least one of the reconstructed data and the reference data, highlighting the data area according to the size of the difference. Figure 6 is a schematic diagram illustrating the configuration of the display control unit according to the first embodiment. More specifically, Figure 6 is a schematic diagram showing an example of the configuration of the display screen that the display control unit 116 according to this embodiment displays on the user terminal 200. Figure 6 illustrates the display screen when the data to be analyzed is an infrared absorption spectrum.

[0064] As shown in Figure 6, the display control unit 116 displays the reconstructed data and the reference data superimposed on the display screen P2. On the display screen P2, the reconstructed data is represented by a solid line, and the reference data is represented by a dashed line.

[0065] As shown in Figure 6, the display control unit 116 according to this embodiment highlights and displays data areas where the difference between the reconstructed data and the reference data is large. More specifically, the display control unit 116 according to this embodiment displays the reconstructed data and reference data while color-coding the data areas based on the evaluation results of the difference evaluation unit. In other words, the display control unit 116 according to this embodiment highlights data areas with large differences by coloring those areas. With this configuration, users can easily identify areas where data changes due to changes in principal component scores are significant, that is, data areas where the characteristics of each principal component are strongly expressed. As a result, the display system implemented as server 100 in this embodiment can facilitate the interpretation of the analysis results of principal component analysis.

[0066] Specifically, the display control unit 116 according to this embodiment displays a spectral chart showing reconstructed data, a spectral chart showing reference data, and color-coded images P21 and P22 superimposed on each other. Color-coded images P21 and P22 are superimposed on data regions where the difference between the reconstructed data and the reference data is greater than or equal to a predetermined threshold, and each color colors the data region. Color-coded image P21 is superimposed on data regions where the spectral value of the reconstructed data is smaller than the spectral value of the reference data, and color-coded image P22 is superimposed on data regions where the spectral value of the reconstructed data is larger than the spectral value of the reference data. With this configuration, users can gain a detailed understanding of how data changes due to changes in principal component scores. As a result, the server 100 according to this embodiment can assist in a more precise interpretation of the results of principal component analysis.

[0067] (Display system operation) Next, the operation of the display system, that is, the display method according to the first embodiment, will be described in detail. Figure 7 is a flowchart illustrating the operation of the display system according to the first embodiment. In the following description, Figures 1 to 6 will be referred to as appropriate.

[0068] In the processing procedure shown in Figure 7, the processor 110 of the server 100 reads the program stored in the storage device 130 into the memory 120 and executes it, thereby functioning as the data acquisition unit 111, principal component analysis unit 112, score change acceptance unit 113, reconstructed data output unit 114, difference evaluation unit 115, and display control unit 116.

[0069] In the display method relating to this implementation, first, the processor 110 acquires multiple data to be analyzed (step ST1). In other words, in step ST1, the processor 110 functions as a data acquisition unit 111 for analysis. For example, in step ST1, the processor 110 receives a selection of data to be analyzed from the user terminal 200. The processor 110 then retrieves the selected data from the storage device 130.

[0070] Next, the processor 110 performs principal component analysis on multiple data sets (step ST2). In other words, in step ST2, the processor 110 functions as a principal component analysis unit 112. More specifically, in step ST2, the processor 110 performs principal component analysis on multiple files to be analyzed acquired in step ST1 and obtains the principal components. In step ST2, the processor 110 also obtains the principal component scores of the data to be analyzed. However, the timing at which the processor 110 acquires the principal component scores is not limited to step ST2. For example, the processor 110 may calculate and acquire the principal component scores of the data to be analyzed and displayed in step ST3.

[0071] Next, the processor 110 receives a request from the user to change the principal component score (step ST3). In other words, in step ST3, the processor 110 functions as a score change reception unit 113. More specifically, in step ST3, the processor 110 displays the principal component score obtained in step ST2 to the user. The processor 110 then accepts a change to the displayed principal component score from the user.

[0072] Next, the processor 110 outputs reconstructed data based on the modified principal component scores (step ST4). In other words, in step ST4, the processor 110 functions as a reconstructed data output unit 114. The processor 110 according to this embodiment outputs reference reconstruction data. The timing of outputting the reference reconstruction data is not particularly limited and can be performed anytime between the end of step ST2 and the start of step ST5, which will be described later.

[0073] Next, the processor 110 evaluates the difference between the reference data and the reconstructed data (step ST5). In other words, in step ST2, the processor 110 functions as a difference evaluation unit 115.

[0074] Finally, the processor 110 highlights and displays the data area according to the size of the difference (step ST6), and the server 100 according to this embodiment completes the series of operations. In other words, in step ST6, the processor 110 is functioning as the display control unit 116. More specifically, in step ST6, the processor 110 displays a display screen P2, as illustrated in Figure 6, on the user terminal 200.

[0075] As described above, the display system according to this embodiment is implemented as a server 100. The display system according to this embodiment outputs reconstructed data based on the principal components of the data to be analyzed and the principal component scores modified by the user. The display system then highlights data areas where the difference between the reconstructed data and the reference data is large, and displays the reconstructed data and the reference data.

[0076] With this configuration, users can easily identify data areas where changes due to changes in principal component scores are significant, i.e., data areas where the characteristics of each principal component are strongly expressed. As a result, the display system according to this embodiment can facilitate the interpretation of the analysis results of principal component analysis.

[0077] <Variation> Figures 8 to 11 are schematic diagrams illustrating the configuration of the display control unit according to the first embodiment. More specifically, Figures 8 to 11 are schematic diagrams showing modified configurations of the display screen that the display control unit 116 according to this embodiment displays on the user terminal 200. As with Figure 6, Figures 8 to 11 illustrate the display screen when the data to be analyzed is an infrared absorption spectrum.

[0078] (Variation 1) For example, as shown in Figure 8, the display control unit 116 according to this embodiment may superimpose and display a heatmap that reflects the magnitude of the difference on at least one of the reconstructed data and the reference data. In other words, the display control unit 116 according to this embodiment may emphasize the data area by superimposing a heat map in which the color changes continuously according to the magnitude of the difference.

[0079] In Figure 8, the heatmap is color-coded so that the darker the color, the greater the difference between the reconstructed data and the reference data. With this configuration, users can gain a more detailed understanding of the magnitude of data changes resulting from changes in principal component scores. As a result, the server 100 according to this embodiment can assist in a more precise interpretation of the analysis results of principal component analysis.

[0080] Furthermore, the display control unit 116 according to this embodiment may superimpose a heat map such that, for example, the reconstructed data is warmer when it is larger than the reference data, and cooler when it is smaller than the reference data. With this configuration, users can gain a more detailed understanding of how data changes due to changes in principal component scores. As a result, the server 100 according to this embodiment can assist in a more precise interpretation of the results of principal component analysis.

[0081] Furthermore, the display control unit 116 according to this embodiment may choose not to superimpose a heatmap in data areas where the difference is smaller than a predetermined size. With this configuration, the server 100 according to this embodiment can simplify the screen displayed to the user. As a result, the server 100 according to this embodiment can display the results of principal component analysis in a more easily understandable format.

[0082] (Modification 2) Furthermore, for example, the display control unit 116 according to this embodiment may superimpose a color-coded image onto the memory portion indicating the data area, as shown in Figure 9. In other words, the display control unit 116 according to this embodiment may highlight the data area by color-coding the memory portion. This configuration also makes it easy for users to visualize data areas where the characteristics of each principal component are strongly expressed.

[0083] (Variation 3) Furthermore, for example, as shown in Figure 10, the display control unit 116 according to this embodiment may superimpose a color-coded image to reduce visibility in areas where the difference between the reconstructed data and the reference data is small. In other words, the display control unit 116 according to this embodiment may display data areas in which the characteristics of each principal component are strongly expressed relatively by reducing the visibility of data areas in which the characteristics of each principal component are not strongly expressed. This configuration also makes it easy for users to visualize data areas where the characteristics of each principal component are strongly expressed.

[0084] (Modification 4) Furthermore, for example, as shown in Figure 11, the display control unit 116 according to this embodiment may superimpose a frame-shaped image onto data areas where the difference between the reconstructed data and the reference data is greater than a predetermined threshold. In other words, the display control unit 116 according to this embodiment may emphasize data areas where the characteristics of each principal component are strongly expressed by surrounding them with a frame. This configuration also makes it easy for users to visualize data areas where the characteristics of each principal component are strongly expressed.

[0085] (Variation 5) In this embodiment, the principal component acquisition unit 112a and the score acquisition unit 112b may acquire principal components and principal component scores by analyzing multiple data to be analyzed using sparse principal component analysis. When principal components and principal component scores are obtained using sparse principal component analysis, the contribution of variables that do not reflect the characteristics of each principal component can be reduced. In this case, the difference between the reconstructed data and the reference data reflects the characteristics of each principal component more clearly, and as a result, users can more precisely visualize the data regions in which the characteristics of each principal component are strongly expressed.

[0086] <Other Embodiments> The display system according to the first embodiment is implemented as a server 100, but the configuration of the display system according to this disclosure is not limited thereto. For example, the display system according to this disclosure may be implemented by two or more computer devices.

[0087] Figures 12 and 13 are block diagrams showing the configuration of a display system according to another embodiment. More specifically, Figures 12 and 13 are block diagrams showing an example in which the display system according to this disclosure is implemented with a different device configuration than that of the first embodiment.

[0088] For example, the display system according to this disclosure may be implemented by a device configuration as shown in Figure 12. That is, the display system according to this disclosure may be implemented by a device configuration in which the user terminal 200 comprises an analysis target data acquisition unit 111, a principal component analysis unit 112, a score change reception unit 113, a reconstructed data output unit 114, a difference evaluation unit 115, and a display control unit 116.

[0089] In this case, the user terminal 200 may include various processors such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an FPGA (field-programmable gate array), memory implemented using RAM (Random Access Memory), and a storage device implemented using a hard disk, SSD (Solid State Drive), memory card, or ROM (Read Only Memory). The processor may then read a program stored in the storage device into memory and execute it, thereby performing the functions of an analysis target data acquisition unit 111, a principal component analysis unit 112, a score change acceptance unit 113, a reconstructed data output unit 114, a difference evaluation unit 115, and a display control unit 116.

[0090] Furthermore, the display system relating to this disclosure may be implemented with a device configuration as shown in Figure 13. In other words, the display system relating to this disclosure may be implemented with a device configuration in which the server 100 comprises an analysis target data acquisition unit 111 and a principal component analysis unit 112, and the user terminal 200 comprises a principal component acquisition unit 112a, a score acquisition unit 112b, a score change reception unit 113, a reconstructed data output unit 114, a difference evaluation unit 115, and a display control unit 116.

[0091] In this case, the server 100 performs principal component analysis on multiple data sets to be analyzed and sends the analysis results, namely the principal components and principal component scores, to the user terminal. Then, the user terminal 200 outputs reconstructed data based on the received principal components and principal component scores, and displays the reconstructed data and the reference data, highlighting the data area according to the magnitude of the difference between the reconstructed data and the reference data. In this case, the principal component acquisition unit 112a acquires the principal components by receiving them from the server 100 without performing principal component analysis. Similarly, the score acquisition unit 112b acquires the principal component scores by receiving the principal component scores from the server 100. The display system described in this disclosure can also be realized through such configurations.

[0092] Although the present invention has been described above in reference to the embodiments described above, the present invention is not limited to the configuration of the embodiments described above, and of course includes various modifications, alterations, and combinations that can be made by a person skilled in the art within the scope of the claims of the present patent application. [Explanation of symbols]

[0093] 1. Analysis System 100 servers 110 processors 120 memory 130 Storage Devices 140 Input / Output Interfaces 150 network interfaces 160 Internal Bus 111 Data acquisition unit for analysis 112 Principal component analysis section 112a Principal component acquisition section 112b Score acquisition section 113 Score Change Request Department 114 Reconstruction Data Output Unit 115 Difference Evaluation Unit 116 Display Control Unit 200 user terminals

Claims

1. A principal component acquisition unit that acquires principal components, which are the results of principal component analysis of multiple data sets to be analyzed, A score acquisition unit that acquires the principal component scores of the data to be analyzed and displayed, A score change reception unit that receives requests from users to change the principal component scores of the data to be analyzed and displayed, A reconstructed data output unit outputs reconstructed data obtained by reconstructing the data to be analyzed based on the principal components and the modified principal component scores. A difference evaluation unit that evaluates the difference between the reconstructed data and the reference data, The system includes a display control unit that displays at least one of the reconstructed data and the reference data while emphasizing the data area according to the magnitude of the difference, Display system.

2. The aforementioned reference data is reference reconstructed data obtained by reconstructing the data to be analyzed based on the principal components and the principal component scores that have not been modified by the user. The display system according to claim 1.

3. The aforementioned reference data is the data to be analyzed. The display system according to claim 1.

4. The display control unit displays at least one of the reconstructed data and the reference data, color-coding the data area according to the evaluation result of the difference evaluation unit. The display system according to any one of claims 1 to 3.

5. The display control unit overlays and displays a heatmap reflecting the magnitude of the difference on at least one of the reconstructed data and the reference data. The display system according to claim 4.

6. The display control unit shall not superimpose the heatmap in data areas where the difference is smaller than a predetermined size. The display system according to claim 5.

7. The principal component acquisition unit acquires principal components which are the result of sparse principal component analysis. The display system according to any one of claims 1 to 3.

8. The display control unit causes the reconstructed data and the reference data to be displayed superimposed. The display system according to any one of claims 1 to 3.

9. We obtain the principal components, which are the results of principal component analysis of multiple data sets to be analyzed. The principal component scores of the data to be analyzed and displayed are obtained. The system accepts changes from the user to the principal component scores of the data to be analyzed and displayed. Based on the aforementioned principal components and the modified principal component scores, reconstructed data is output, which is the data to be analyzed. The difference between the reconstructed data and the reference data is evaluated. The data area is highlighted according to the magnitude of the difference, and at least one of the reconstructed data and the reference data is displayed. Display method.

10. We obtain the principal components, which are the results of principal component analysis of multiple data sets to be analyzed. The principal component scores of the data to be analyzed and displayed are obtained. The system accepts changes from the user to the principal component scores of the data to be analyzed and displayed. Based on the aforementioned principal components and the modified principal component scores, reconstructed data is output, which is the data to be analyzed. The difference between the reconstructed data and the reference data is evaluated. The computer is instructed to perform an operation to display at least one of the reconstructed data and the reference data, while highlighting the data area according to the magnitude of the difference. Display program.

Images