A user terminal that displays the results of data analysis, a program for the user terminal, and a server that communicates with the user terminal

Abstract

To provide an interface that allows charts and tables that are made by changing only some of certain condition settings to be simply added to charts and tables that are displayed based on the certain condition settings, in a system that displays the results of data analysis in charts and tables.SOLUTION: A user terminal obtains a first set of setting values that defines multiple types of analysis conditions in response to setting operations by a user regarding multiple types of analysis conditions. The user terminal then displays analysis results according to the obtained first set of setting values as a view V1 on a display. When the user performs a copy operation targeting the view V1, the user terminal displays analysis results according to a second set of setting values that have the same values as the first set of setting values, together with the first view as a view V2 on the display. When the user operates to partially change the second set of setting values, the user terminal further reflects analysis results according to the changed second set of setting values in the second view.SELECTED DRAWING: Figure 15

Description

[Technical Field]

[0001] The present invention relates to a user terminal that displays the results of data analysis, a program for the user terminal, and a server that communicates with the user terminal. [Background technology]

[0002] Patent Document 1 describes a system that displays the results of data analysis in the form of diagrams (e.g., graphs, tables). In this system, the layout of the displayed diagrams is determined in advance by the system and cannot be changed by user operation. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-42704 Summary of the Invention [Problem to be solved by the invention]

[0004] The inventors of the present application considered enabling users to change the layout of charts in a system that displays the results of data analysis in charts, and came up with the idea that there may be cases where users wish to add a chart that changes only part of the chart settings that are displayed based on certain condition settings.

[0005] In view of the above, the present invention aims to provide an interface in a system that displays the results of data analysis in charts and graphs, which allows users to easily add charts that change only part of the condition settings to charts that are displayed based on certain condition settings. [Means for solving the problem]

[0006] The invention described in claim 1 to achieve the above object is: A user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a display (23) for displaying images to the user; 、 an operation device (24) that accepts the user's operation; a control circuit (25); The control circuit a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that defines a plurality of types of analysis conditions in response to a setting operation performed by the user on the operation device for the plurality of types of analysis conditions; a view display unit (S250, S320, S420, S520) that displays an analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit on the display as a first view (V1); The view display unit, based on the user's operation of copying the first view using the operation device, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of changing a part of the second setting value set using the operation device, reflects the analysis result corresponding to the changed second setting value set in the second view. 、 The setting value acquisition unit is capable of simultaneously accepting multiple setting values ​​for one type of analysis condition among the multiple types of analysis conditions as part of the first setting value set, and the view display unit is a user terminal that includes multiple charts and graphs corresponding to the multiple setting values ​​in the first view as analysis results according to the first setting value set. The invention described in claim 5 is a user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), and includes a display (23) that displays an image to the user; an operation device (24) that accepts the user's operation; a control circuit (25); The control circuit a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that defines a plurality of types of analysis conditions in response to a setting operation performed by the user on the operation device for the plurality of types of analysis conditions; a view display unit (S250, S320, S420, S520) that displays an analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit on the display as a first view (V1); the view display unit, when the user has performed a copy operation on the operation device targeting the first view, causes the display to display, as a second view (V2), an analysis result corresponding to a second setting value set having values ​​identical to those of the first setting value set together with the first view, and, when the user has performed an operation on the operation device to change a part of the second setting value set, causes the view display unit to reflect, in the second view, the analysis result corresponding to the second setting value set after the change; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; The setting value acquisition unit is a user terminal that displays on the display a candidate setting unit (67) that shows candidates for new learning targets to be selected in the active learning in a tabular format, enables the user to set values ​​for individual cells in the tabular display, and further enables tables created by a function other than the function of the setting value acquisition unit to be reflected all at once in the tabular display. The invention described in claim 6 is a user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a display (23) for displaying images to a user; an operation device (24) that accepts the user's operation; a control circuit (25); The control circuit a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that defines a plurality of types of analysis conditions in response to a setting operation performed by the user on the operation device for the plurality of types of analysis conditions; a view display unit (S250, S320, S420, S520) that displays an analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit on the display as a first view (V1); the view display unit, when the user has performed a copy operation on the operation device targeting the first view, causes the display to display, as a second view (V2), an analysis result corresponding to a second setting value set having values ​​identical to those of the first setting value set together with the first view, and, when the user has performed an operation on the operation device to change a part of the second setting value set, causes the view display unit to reflect, in the second view, the analysis result corresponding to the second setting value set after the change; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; the first set of setting values ​​includes values ​​indicating a search range for explanatory variables used in the Gaussian process regression, the values ​​being candidates for new learning targets to be used in the Gaussian process regression; The setting value acquisition unit is a user terminal that, when the user specifies explanatory variables to be used in the Gaussian process regression, sets a value indicating the search range of the explanatory variables based on values ​​of the explanatory variables in training data that have already been used in the Gaussian process regression.

[0007] Also, claims 11 The invention described in is a program used in a user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; causing the user terminal to function as a view display unit (S250, S320, S420, S520) that displays the analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); The view display unit, based on the user's operation of copying the first view using the operation device, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of changing a part of the second setting value set using the operation device, reflects the analysis result corresponding to the changed second setting value set in the second view. 、 the setting value acquisition unit is capable of simultaneously accepting a plurality of setting values ​​for one type of analysis condition among the plurality of types of analysis conditions as part of the first setting value set; the view display unit includes in the first view a plurality of charts corresponding to the plurality of setting values, respectively, as analysis results according to the first setting value set. It is a program. The invention described in claim 12 is a program used in a user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; causing the user terminal to function as a view display unit (S250, S320, S420, S520) that displays the analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, when the user has performed a copy operation on the operation device targeting the first view, causes the display to display, as a second view (V2), an analysis result corresponding to a second setting value set having values ​​identical to those of the first setting value set together with the first view, and, when the user has performed an operation on the operation device to change a part of the second setting value set, causes the view display unit to reflect, in the second view, the analysis result corresponding to the second setting value set after the change; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; The setting value acquisition unit is a program that displays on the display a candidate setting unit (67) that shows candidates for new learning targets selected in the active learning in a tabular format, enables the user to set values ​​for individual cells in the tabular display, and further enables tables created by a function other than the function of the setting value acquisition unit to be reflected in the tabular display all at once. The invention described in claim 13 is a program used in a user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; causing the user terminal to function as a view display unit (S250, S320, S420, S520) that displays the analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, when the user has performed a copy operation on the operation device targeting the first view, causes the display to display, as a second view (V2), an analysis result corresponding to a second setting value set having values ​​identical to those of the first setting value set together with the first view, and, when the user has performed an operation on the operation device to change a part of the second setting value set, causes the view display unit to reflect, in the second view, the analysis result corresponding to the second setting value set after the change; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; the first set of setting values ​​includes values ​​indicating a search range for explanatory variables used in the Gaussian process regression, the values ​​being candidates for new learning targets to be used in the Gaussian process regression; The setting value acquisition unit is a program that, when the user specifies explanatory variables to be used in the Gaussian process regression, sets values ​​indicating the search range of the explanatory variables based on values ​​of the explanatory variables in training data that have already been used in the Gaussian process regression.

[0008] Also, claims 14 The invention described in A server that communicates with a user terminal (1), a communication interface circuit (11) for communicating with the user terminal; a storage medium (12) for storing data (12b) to be analyzed; a control circuit (13) that analyzes the data and transmits the analysis result to the user terminal using the communication interface circuit; The storage medium (12) stores a program (12a) to be executed on the user terminal, the control circuit transmits the program to the user terminal using the interface circuit; The program causes the user terminal to a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; functioning as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); The view display unit, based on the user's operation of copying the first view using the operation device, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of changing a part of the second setting value set using the operation device, reflects the analysis result corresponding to the changed second setting value set in the second view. 、 the setting value acquisition unit is capable of simultaneously accepting a plurality of setting values ​​for one type of analysis condition among the plurality of types of analysis conditions as part of the first setting value set; The view display unit is a server that includes, in the first view, a plurality of charts corresponding to the plurality of setting values, respectively, as analysis results according to the first setting value set. The invention described in claim 15 is a server that communicates with a user terminal (1), a communication interface circuit (11) for communicating with the user terminal; a storage medium (12) for storing analysis target data (12b); a control circuit (13) that analyzes the analysis target data and transmits the analysis result to the user terminal using the communication interface circuit; The storage medium (12) stores a program (12a) to be executed on the user terminal, the control circuit transmits the program to the user terminal using the communication interface circuit; The program causes the user terminal to a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; functioning as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, based on the user's operation of the operation device to copy the first view, causes the display to display, together with the first view, an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2), and further, based on the user's operation of the operation device to change a part of the second setting value set, causes the second view to reflect the analysis result corresponding to the changed second setting value set; the first view includes a diagram displaying a new learning target selected by active learning based on Gaussian process regression; The setting value acquisition unit is a server that displays on the display a candidate setting unit (67) that shows candidates for new learning targets selected in the active learning in a tabular format, enables the user to set values ​​for individual cells in the tabular display, and further enables tables created by a function other than the function of the setting value acquisition unit to be reflected in the tabular display all at once. The invention described in claim 16 is a server that communicates with a user terminal (1), a communication interface circuit (11) for communicating with the user terminal; a storage medium (12) for storing analysis target data (12b); a control circuit (13) that analyzes the analysis target data and transmits the analysis result to the user terminal using the communication interface circuit; The storage medium (12) stores a program (12a) to be executed on the user terminal, the control circuit transmits the program to the user terminal using the communication interface circuit; The program causes the user terminal to a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; functioning as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, when the user has performed a copy operation on the operation device targeting the first view, causes the display to display, as a second view (V2), an analysis result corresponding to a second setting value set having values ​​identical to those of the first setting value set together with the first view, and, when the user has performed an operation on the operation device to change a part of the second setting value set, causes the view display unit to reflect, in the second view, the analysis result corresponding to the second setting value set after the change; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; the first set of setting values ​​includes values ​​indicating a search range for explanatory variables used in the Gaussian process regression, the values ​​being candidates for new learning targets to be used in the Gaussian process regression; The setting value acquisition unit is a server that, when the user specifies explanatory variables to be used in the Gaussian process regression, sets values ​​indicating the search range of the explanatory variables based on values ​​of the explanatory variables in training data that have already been used in the Gaussian process regression.

[0009] In this way, in order to add a chart that has only part of the condition settings changed from a chart that is displayed based on certain condition settings, a copy operation and an operation to change part of the second set of setting values ​​are performed, which is easier than setting the entire second set of setting values.

[0010] The reference symbols in parentheses attached to each component indicate an example of the correspondence between the component and the specific components described in the embodiments described below. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a configuration diagram of a material analysis system. [Figure 2] 1 is an example of a hardware configuration of a server. [Figure 3] 1 is an example of a hardware configuration of a user terminal. [Figure 4] FIG. 2 is a diagram illustrating the functional configuration of a server and a user terminal. [Figure 5] FIG. 10 is a diagram illustrating an example of a data management screen. [Figure 6] FIG. 10 is a diagram illustrating an example of a data registration screen. [Figure 7] FIG. 10 is a diagram illustrating an example of an analysis start screen. [Figure 8] FIG. 10 is a diagram showing an example of an analysis result display screen in a state where no view is displayed. [Figure 9] FIG. 10 is a diagram showing an example of an analysis result display screen in a state where the analysis result is not displayed. [Figure 10] FIG. 10 is a diagram showing an example of an analysis condition setting screen. [Figure 11] 10 is a flowchart of a process executed during view display. [Figure 12] 10 is a flowchart executed while an analysis condition setting screen is displayed. [Figure 13] FIG. 10 is a diagram showing an example of an analysis result display screen in a state where the analysis result has been displayed. [Figure 14]10 is a flowchart of a process for copying a view that is executed while a view is being displayed. [Figure 15] FIG. 10 is a diagram showing a view in which a diagram has been added by copying. [Figure 16] FIG. 10 is a diagram showing a state in which the display content of a diagram or table added by copying has been changed. [Figure 17] FIG. 10 is a diagram showing a chart displayed when two setting values ​​are set for each of two types of analysis conditions. [Figure 18] 10 is a flowchart of a process executed during view display to move a view. [Figure 19] 10 is a flowchart of a process executed during view display to interactively manipulate a chart. [Figure 20] FIG. 10 is a diagram showing an analysis condition setting screen for Gaussian process regression when random mode is selected. [Figure 21] 10 is a flowchart of a process executed while the analysis condition setting screen for Gaussian process regression is displayed when the random mode is selected. [Figure 22] FIG. 10 is a diagram showing an analysis condition setting screen for Gaussian process regression when a user-defined mode is selected. [Figure 23] 10 is a flowchart of a process executed while a screen for setting analysis conditions for Gaussian process regression is displayed when a user-defined mode is selected. [Figure 24] 10 is a chart showing the analysis results of Gaussian process regression. [Figure 25] 10 is a diagram showing the results of classification analysis. [Figure 26] FIG. 1 is a network diagram showing the results of statistical causal search. [Figure 27] FIG. 10 shows a graph and table that work together. [Figure 28] FIG. 10 is a diagram showing a statistical table. [Figure 29] FIG. 1 shows a pair plot. [Figure 30] FIG. 10 is a diagram showing a heat map. [Figure 31] This figure shows the regression results in a two-dimensional graph and a bar graph. [Figure 32] FIG. 10 is a diagram showing a state in which multiple views are arranged within one canvas. DETAILED DESCRIPTION OF THE INVENTION

[0012] An embodiment of the present disclosure will be described below. As shown in Fig. 1, a material analysis system according to this embodiment includes a server 1 and a user terminal 2. The server 1 and the user terminal 2 are capable of communicating with each other via a communication network 4. The communication network 4 may include a wide area network such as the Internet. The communication network 4 may also include a wired network or a wireless network.

[0013] Each of the multiple user terminals 2 communicates with the server 1 via a communication network 4 in accordance with user operations. During this communication, each user terminal 2 sometimes transmits to the server 1 experimental data related to materials for supporting materials development, and sometimes obtains from the server 1 analysis results based on experimental data related to materials already registered in the server 1 and displays them to the user. Types of experiments indicated by the experimental data include, but are not limited to, X-ray spectroscopy, X-ray diffraction measurement, infrared spectroscopy, X-ray imaging, infrared imaging, and microscope imaging. The experimental data corresponds to the data to be analyzed.

[0014] The server 1 is a material analysis server that receives experimental data transmitted from multiple user terminals 2 and performs analysis using the received experimental data. Machine learning (e.g., deep learning) is used for the analysis. The server 1 then transmits the analysis results to one of the multiple user terminals 2 in response to a request transmitted from the user terminal 2 via a communication network 4. The user terminal 2 displays the received analysis results on a display for the operator of the user terminal 3, as described above.

[0015] 2, the server 1 includes a communication interface circuit 11, a storage medium 12, and a control circuit 13. The communication interface circuit 11 is an interface circuit for communicating with the user terminal 2 via the communication network 4.

[0016] The storage medium 12 is a non-volatile storage medium such as an SSD or HDD. A program executed by the control circuit 13 is recorded in the storage medium 12. Data can be read from and written to the storage medium 12 by the control circuit 13. The storage medium 12 is a non-transitory tangible storage medium.

[0017] The control circuit 13 is a circuit having an arithmetic circuit, a volatile storage medium such as RAM, etc. The arithmetic circuit performs various processes by executing various programs recorded in the storage medium 12, and uses the volatile storage medium as a working area during this process. The volatile storage medium is a non-transient tangible storage medium.

[0018] 3, each user terminal 2 has a communication interface circuit 21, a storage medium 22, a display 23, an operation device 24, and a control circuit 25. The communication interface circuit 21 is an interface circuit for communicating with the server 1 via the communication network 4.

[0019] The storage medium 22 is a non-volatile storage medium such as an SSD or HDD. A program executed by the control circuit 25 is recorded in the storage medium 22. Data can be read from and written to the storage medium 22 by the control circuit 25. The storage medium 22 is a non-transitory tangible storage medium.

[0020] The display 23 is a device such as a liquid crystal display that displays images to the user. The display 23 displays images under the control of the control circuit 25. The operation device 24 is a device such as a keyboard, mouse, or touch panel that receives user operations and outputs a signal to the control circuit 25 according to the content of the received operation.

[0021] The control circuit 25 is a circuit having an arithmetic circuit, a volatile storage medium such as RAM, etc. The arithmetic circuit performs various processes by executing various programs recorded in the storage medium 22, etc., and uses the volatile storage medium as a working area during this process. The volatile storage medium is a non-transient tangible storage medium.

[0022] 4, control circuit 13 of server 1 functions as Web service unit 13a and back-end unit 13b. Specifically, control circuit 13 functions as Web service unit 13a and back-end unit 13b by executing a Web service program and a back-end program (not shown) recorded on storage medium 12. However, as another example, control circuit 13 may have a dedicated circuit for Web service unit 13a and a dedicated circuit for back-end unit 13b. In this case, these dedicated circuits may be hardware circuits whose circuit configuration is not programmable, or may be programmable logic circuits whose circuit configuration is programmable.

[0023] Furthermore, the control circuit 25 of the user terminal 2 functions as a client unit 25a or the like. Specifically, the control circuit 25 functions as the client unit 25a by executing a client program (not shown) recorded in the storage medium 22. However, as another example, the control circuit 25 may have a dedicated circuit for the client unit 25a. In that case, this dedicated circuit may be a hardware circuit whose circuit configuration is not programmable, or may be a programmable logic circuit whose circuit configuration is programmable.

[0024] The client unit 25a of the user terminal 2 functions as a web browser and displays a canvas, view, etc., which will be described below, within the display range of the web browser (i.e., within the frame of the web browser window) of the entire display screen of the display 23. Specifically, the client unit 25a transmits a request to the server 1 according to the HTTP protocol in response to a user's operation on the operation device 24. Then, the client unit 25a receives a response (e.g., HTML source code) in response to the request, and causes the display 23 to display a display in accordance with the received response (e.g., a display in accordance with the HTML source code).

[0025] Web service unit 13a functions as a Web server. When Web service unit 13a receives a request from client unit 25a of user terminal 2, it acquires a response corresponding to the received request. The response may be acquired by reading one that has been previously recorded on storage medium 12, by creating a response based on information included in the request, or by combining these methods. Web service unit 13a then transmits the acquired response to client unit 25a.

[0026] The response sent from the Web service unit 13a to the client unit 25a in this manner includes a program such as the HTML source code described above. Apart from the HTML source code, the response may also include a program such as Javascript (registered trademark) code. Such a program, such as the HTML source code or Javascript code, may include the UI program 12a pre-recorded in the storage medium 12 of the server 1.

[0027] This UI program 12a is sent by the Web service unit 13a to the client unit 25a as part of a response. The client unit 25a then executes the received UI program 12a by functioning as a Web browser and a JavaScript interpreter. That is, the control circuit 25 realizes part of the client unit 25a by executing the UI program 12a. Although HTML is a data description language, its source code defines the content to be displayed by the client unit 25a, and therefore is categorized as a program in this specification. The UI program 12a may also include information such as CSS.

[0028] By executing the UI program 12a, the client unit 25a communicates with the backend unit 13b of the server 1 via the communication interface circuit 21 in response to user operations on the operation device 24, as described below, and causes the display 23 to display various information.

[0029] The back-end unit 13b communicates with the client unit 25a via the communication interface circuit 11, and performs various processes such as analysis based on requests received from the client unit 25a. For example, the back-end unit 13b performs various analyses using the experimental data 12b recorded in the storage medium 12, and transmits the analysis results to the client unit 25a. For communication between the back-end unit 13b and the client unit 25a, for example, the well-known WebSocket technology or other technologies may be used.

[0030] The operation of the material analysis system having such a configuration will be described below. First, the operation of registering the experiment data 12b from the user terminal 2 to the server 1 will be described, and then the operation of displaying the analysis results of the experiment data 12b on the user terminal 2 will be described.

[0031] [Experimental data registration] First, when a user operates the operation device 24 at one of the user terminals 2 to perform a predetermined operation, the client unit 25a transmits a request to the Web service unit 13a based on the operation. The Web service unit 13a then transmits a response including HTML source and JavaScript code for predetermined data management to the client unit 25a. Both the HTML source and the JavaScript code are programs executed by the control circuit 25.

[0032] Upon receiving this response, the client unit 25a displays a data management screen, such as the one shown in Fig. 5, on the display 23 in accordance with the description of the HTML source and Javascript code in the received response. This data management screen has a data management button D1, an analysis button D2, an experiment data list D11, a character input field D12, a filter application button D13, and a data registration button D14. The data management button D1, analysis button D2, filter application button D13, and data registration button D14 are button images that the user can select using the operation device 24.

[0033] When the data management button D1 is selected, the client unit 25a sends a request for this data management screen to the Web service unit 13a. Specifically, the client unit 25a sends a request including the URL of the data management screen to the Web service unit 13a.

[0034] When the analysis button D2 is selected, the client unit 25a transmits a request for an analysis screen (described later) to the Web service unit 13a. Specifically, the client unit 25a transmits a request including the URL of the analysis screen to the Web service unit 13a.

[0035] The experiment data list D11 is a table that lists multiple pieces of experiment data 12b recorded on the storage medium 12 of the server 1. Information on the multiple pieces of experiment data 12b is included in the response received from the Web service unit 13a. Each row in the experiment data list D11 in FIG. 5, except for the first row, corresponds to one piece of experiment data 12b. In the example of FIG. 5, the following information is displayed for each experiment: the name of the data, whether or not it is owned by this user, the owner, access rights, the date and time of modification, and a description.

[0036] The character input field D12 is a field where a character string to be used as a filter can be input. When the user inputs a character string (for example, "aluminum") using the operation device 24 and then selects the filter application button D13, the client unit 25a displays only the experimental data 12b whose name includes the character string in the experimental data list D11.

[0037] When the data registration button D14 is selected, the client unit 25a sends a request for a data registration screen to the Web service unit 13a. Specifically, the client unit 25a sends a request including the URL of the data registration screen to the Web service unit 13a.

[0038] The Web service unit 13a then sends a response including an HTML source for displaying a data registration screen to the client unit 25a. The client unit 25a, which receives this response, displays a data registration screen such as the one shown in FIG. 6 on the display 23 in accordance with the description of the HTML source in the received response.

[0039] This data management screen includes a data management button D1, an analysis button D2, a name input field D151 in which the name of the experimental data to be registered can be input, and a description input field D152 in which a description of the experimental data can be input. The data management screen also includes a file selection field D153 in which the file of the experimental data can be selected from the storage medium 22, and a drop-down list D154 in which access permissions for the experimental data can be set. The data management screen also includes an owner selection field D155 and an owner group selection field D156 in which the owner and owner group of the experimental data can be selectively set or deleted, and a save button D157 for confirming the settings input on this management screen.

[0040] The user uses the operation device 24 to input auxiliary information for the experimental data to be registered into the name input field D151, description input field D152, drop-down list D154, owner selection field D155, and owner group selection field D156. The user then uses the operation device 24 to perform an operation to select the experimental data in the file selection field D153. The user then uses the operation device 24 to select the save button D157. The client unit 25a then transmits a request including the input information for the experimental data and the file of the experimental data to the Web service unit 13a.

[0041] Upon receiving this request, the Web service unit 13a records the received experimental data 12b file and its associated information in the storage medium 12. By repeating this operation, multiple experimental data 12b and their associated information are registered in the storage medium 12 of the server 1.

[0042] [Displaying analytical results by regression of experimental data] Next, the display of analysis results based on regression of experimental data will be described. Suppose that one or more sets of experimental data 12b are stored in the storage medium 12 of the server 1, and a user performs a predetermined operation for analysis on one of the plurality of user terminals 2. An example of the predetermined operation for analysis is the operation of selecting the above-mentioned analysis button D2 using the operation device 24.

[0043] The Web service unit 13a then sends a response including HTML source and JavaScript code for starting a predetermined data analysis to the client unit 25a. Upon receiving this response, the client unit 25a displays a data analysis start screen, such as the one shown in FIG. 7, on the display 23 in accordance with the description of the HTML source and JavaScript code in the received response. This data management screen includes a data management button D1, an analysis button D2, an analysis result list D21, a character input field D22, a filter application button D23, and a new analysis button D24. The data management button D1, analysis button D2, filter application button D23, and new analysis button D24 are button images that the user can select using the operation device 24.

[0044] The analysis result list D21 is a table that lists the results of analyses that have already been performed. The analysis results are recorded in the storage medium 12 of the server 1. Each row in the analysis result list D21 in FIG. 7, except for the first row, corresponds to one analysis result. In the example of FIG. 7, the name of the analysis, whether or not it is owned by this user, the owner, access rights, the date and time of modification, and a description are displayed for each experiment.

[0045] Each analysis result in these lists can be selected by the user's operation using the operation device 24. When selected, the client unit 25a sends a request specifying the selected analysis result to the Web service unit 13a. Upon receiving this request, the Web service unit 13a sends a response to the client unit 25a that includes the analysis result data specified in the received request, as well as HTML source code and JavaScript code for displaying the data. Upon receiving this response, the client unit 25a displays the analysis result on the display 23 in accordance with the content of the response.

[0046] The character input field D22 is a field where a character string to be used as a filter can be input. When the user inputs a character string using the operation device 24 and then selects the filter application button D23, the client unit 25a displays only analysis results whose names include the character string in the analysis result list D21.

[0047] When the new analysis button D24 is selected, the client unit 25a sends a request for an analysis result display screen to the Web service unit 13a. Specifically, the client unit 25a sends a request including the URL of the analysis result display screen to the Web service unit 13a.

[0048] Web service unit 13a then reads out the HTML source and JavaScript code for displaying the analysis result display screen from storage medium 12. This HTML source and JavaScript code are UI program 12a shown in Fig. 4. Web service unit 13a then sends a response including this UI program 12a to client unit 25a.

[0049] Upon receiving this response, the client unit 25a displays an analysis result display screen, such as the one shown in Fig. 8, on the display 23 in accordance with the description of the UI program 12a in the received response. This analysis result display screen has a data management button D1, an analysis button D2, a data selection item D25, an add view button D26, and a canvas D30.

[0050] The canvas D30 is an area for displaying a view, which will be described later. The canvas D30 occupies a predetermined area in the display 23. For example, if the client unit 25a uses one window in the display 23 as its display area, the canvas D30 occupies a predetermined area in that window. Note that the canvas D30 may be displayed with its area reserved and a frame indicating its boundary visible, or alternatively, the area may simply be reserved without the reserved area being visible.

[0051] The data selection item D25 is a drop-down list. When the user operates the operation device 24 to specify the data selection item D25, the client unit 25a displays a selectable list of the experimental data 12b recorded in the storage medium 12 of the server 1 on the display 23. The list of the experimental data 12b recorded in the storage medium 12 of the server 1 is the list included in the received UI program 12a.

[0052] When the user selects one of the experimental data 12b from the list using the operation device 24, the client unit 25a displays the name of the experimental data 12b in the position of the data selection item D25.

[0053] 8, the selected experimental data 12b is experimental data relating to the composition and mechanical properties of an aluminum alloy. That is, the selected experimental data 12b includes a plurality of experimental results, each of which includes, as variables, data on the contents of a plurality of materials constituting the product (an aluminum alloy) and the mechanical properties (e.g., tensile strength) of the product.

[0054] The add view button D26 is a button image that the user can select using the operation device 24. A view is a display unit that presents experimental results pictorially, rather than just textually, using one or more diagrams or tables. A view may consist of a single graph, a single table, multiple graphs, multiple tables, or one or more graphs and one or more tables. Note that a diagram or table means "figure or table."

[0055] When the add view button D26 is selected, the client unit 25a displays a view type selection screen (not shown) on the display 23. This view type selection screen is a screen that displays a list of multiple selectable view types. For example, regression, classification, statistical causal discovery, Gaussian process regression, statistical data table, scatter plot, histogram, pair plot, and heat map each correspond to one type of view.

[0056] When the user selects one of the list of view types using the operation device 24, the client unit 25a displays the view V1 in the canvas D30 as exemplified in Fig. 9. The view V1 corresponds to the first view.

[0057] The view V1 includes a main display area V11 that displays diagrams (e.g., graphs, tables) as analysis results, a delete button V12 for deleting the view V1, and a set button V13 for calling up an analysis condition setting screen, etc. The view V1 shown in Fig. 9 is a view that is displayed when regression is selected as the view type.

[0058] When the user selects the delete button V12 using the operation device 24, the client unit 25a deletes the view V1. When the user selects the set button V13 using the operation device 24, the client unit 25a displays the analysis condition setting screen shown in FIG.

[0059] At this time, the processing executed by the client unit 25a is the processing specified by the UI program 12a, as shown in FIG. 11. That is, while a view is being displayed, the client unit 25a first repeatedly determines in step S110 whether the setting button V13 has been selected until it determines that it has been selected, and if it determines that it has been selected, proceeds to step S120. Furthermore, in step S120, the client unit 25a selects an analysis condition setting screen according to the displayed view. Different types of views result in different analysis condition setting screens. The UI program 12a includes information about the correspondence between the types of views and the types of analysis condition setting screens, and in step S120, this information is used to select the analysis condition setting screen. Furthermore, in step S130, the client unit 25a causes the display 23 to display the selected analysis condition setting screen.

[0060] 10, the analysis condition setting screen for regression is a screen for accepting setting operations for multiple types of analysis conditions used for regression. Specifically, the analysis condition setting screen includes multiple setting sections 41 to 50 and a decision button 51. The setting sections 41 to 50 are forms for accepting setting operations for setting values ​​for the analysis conditions, and the decision button 51 is a form for saving those setting values ​​as one setting value set.

[0061] Specifically, the standardization setting unit 41 is a form for setting whether or not to standardize variables of the experimental data 12b in regression. The preprocessing setting unit 42 is a form for selecting the type of preprocessing for the experimental data (e.g., eliminating missing data, etc.). The validation setting unit 43 is a form for selecting the type of validation, such as cross-validation. Different types of validation result in different forms regarding which of the multiple experimental results included in the experimental data 12b are used as a learning dataset during analysis and which are used for accuracy evaluation.

[0062] The algorithm setting unit 44 is a form for selecting a regression algorithm (e.g., linear regression, Ridge regression, Lasso regression). The hyperparameter setting unit 45 is a form for selecting a hyperparameter in regression (e.g., the number of intermediate layers in a neural network). The metric setting unit 46 is a form for selecting an accuracy evaluation index. The dependent variable setting unit 47 is a form for selecting a dependent variable from the selected experimental data.

[0063] The explanatory variable batch setting section 48 is a form for selecting whether or not to set all variables in the selected experimental data 12b other than the objective variable set in the objective variable setting section 47 as explanatory variables. The explanatory variable individual setting section 49 is a form for individually selecting explanatory variables from the selected experimental data 12b. The descriptor setting section 50 is a form for selecting the form of change of the explanatory variable selected in the explanatory variable batch setting section 48 or the individual explanatory variable setting section 49 (for example, changing the explanatory variable x to Log(x)).

[0064] Among these setting sections 41 to 50, the algorithm setting section 44, hyperparameter setting section 45, individual explanatory variable setting section 49, and descriptor setting section 50 are multi-select drop-down format forms. A multi-select drop-down format form is a form that allows multiple setting values ​​to be selected simultaneously for one type of analysis condition.

[0065] The processing executed by the client unit 25a while the analysis condition setting screen is displayed is that shown in Fig. 12 of the UI program 12a. That is, while the analysis condition setting screen is displayed, the client unit 25a first determines in step S210 whether a setting operation has been performed on any of the setting units 41 to 50, and if no setting operation has been performed, proceeds to step S220. Then in step S220, it determines whether the decision button 51 has been selected, and if not, returns to step S210. That is, if no setting operation has been performed on any of the setting units 41 to 50 and the decision button 51 has not been selected, a loop of steps S210 and S220 is repeated.

[0066] In this loop, when the user performs a setting operation on any of the setting units 41 to 50 using the operation device 24, the client unit 25a proceeds from step S210 to step S230. Then, in step S230, the setting value of the analysis condition corresponding to the target setting unit is set according to the content of the setting operation. For example, as shown in FIG. 10, when a setting operation is performed on the algorithm setting unit 44 to set two conditions, Linear and Ridge, the client unit 25a sets the setting value of the analysis condition, Algorithm, to Linear regression and Ridge regression. After step S230, the client unit 25a proceeds to step S220.

[0067] Therefore, until the decision button 51 is selected, a loop of steps S210, S230, and S220 is executed each time a setting operation is performed on the setting units 41 to 50. As a result, the user sets the setting values ​​of the analysis conditions corresponding to each setting unit by performing a setting operation on each of the setting units 41 to 50 as necessary. All of the setting values ​​of the analysis conditions of the setting units 41 to 50 set in this way correspond to one setting value set.

[0068] If the user operates the enter button 51 during the loop of steps S210, S230, and S220 or during the loop of steps S210 and S220, the client unit 25a proceeds from step S220 to step S240, and requests an analysis that reflects the current set of setting values ​​from the server 1. Specifically, the client unit 25a transmits a query to the backend unit 13b, the query including information specifying the selected experimental data 12b (for example, a name), the type of analysis to be performed, and the current set of setting values.

[0069] The back-end unit 13b receives this query and executes the analysis type included in the query on the experimental data 12b specified in the query under the analysis conditions defined by the setting value set included in the query.The back-end unit 13b then transmits an answer including data on the execution results of this analysis (i.e., the analysis results) to the client unit 25a.For example, if a regression analysis is performed as the analysis, the analysis results include information such as a regression model that predicts the correspondence between explanatory variables and response variables, and data points that indicate the values ​​of the explanatory variables and response variables in each experiment.

[0070] Upon receiving this answer, the client unit 25a proceeds from step S240 to step S250. Then, in step S250, an image of a chart is generated using the data of the analysis results in the answer, and the image of the generated chart is displayed on the display 23 as the main display section V11 in the view V1, as shown in Fig. 13. Specifically, points corresponding to each experiment in the experimental data 12b are plotted on a graph in which the horizontal axis represents the actual value of the dependent variable and the vertical axis represents the predicted value of the dependent variable obtained by regression analysis.

[0071] In the RAM of the control circuit 25, the view V1 displayed in this way is associated with information specifying the experimental data 12b used for the view V1, the type of analysis, and a set of setting values.

[0072] 14 among the processes defined by the UI program 12a. That is, in step S310, the client unit 25a repeatedly determines whether the user has performed a predetermined copy operation on the view V1 using the operation device 24 until it determines that the user has performed the operation. The predetermined copy operation on the view V1 may be, for example, clicking the view V1 with a pointing device such as a mouse and performing the well-known Ctrl-C and Ctrl-V operations on the keyboard, or may be another operation. The copy operation may involve five or fewer key presses or mouse clicks on the operation device 24.

[0073] If it is determined that a predetermined copy operation has been performed on view V1, the client unit 25a proceeds to step S320. In step S320, a view V2 having the same main display area V11 as the selected view V1 is displayed in the canvas D30 of the display 23, as shown in FIG. 15. The view V2 corresponds to the second view. The experimental data designation, analysis type, and setting value set associated with this view V2 are the same as those of view V1. In other words, if the setting value set associated with view V1 is the first setting value set and the setting value set associated with view V2 is the second setting value set, the first setting value set and the second setting value set have the same values.

[0074] This view V2 also includes a main display section V21, a delete button V22, and a set button V23, which have the same functions as the main display section V11, the delete button V12, and the set button V13, respectively.

[0075] When the user selects the setting button V23 using the operation device 24, the client unit 25a performs the same process as in Fig. 11 to display an analysis condition setting screen similar to Fig. 10 on the display 23. At this time, the values ​​of the second set of setting values ​​are reflected on the analysis condition setting screen.

[0076] Then, with the analysis condition setting screen for view V2 displayed in this way, client unit 25a performs the same processing as in Fig. 12. At this time, it is assumed that the user performs a setting operation to change the setting value for a desired setting unit among setting units 41 to 50 using operation device 24. Then, client unit 25a proceeds from step S210 to step S230, and sets the setting value of the analysis condition corresponding to the target setting unit in accordance with the content of the setting operation.

[0077] For example, if a user performs a setting operation on the hyperparameter setting unit 45 to change the setting value from Default to Middle, the client unit 25a changes the setting value of the hyperparameter from Default to Middle. That is, the number of intermediate layers of the neural network used in the regression is increased. This changes the value of the second setting value set. Note that the change in the setting value of the analysis condition may be made to a setting unit other than the hyperparameter setting unit 45. Furthermore, the change in the setting value of the analysis condition may be made to not only one but also multiple setting units.

[0078] After the setting values ​​of the analysis conditions corresponding to the target setting section are set, when the user selects the OK button 51, the client unit 25a proceeds from step S220 to step S240. Then, as described above, in step S240, the client unit 25a requests the backend unit 13b of the server 1 to perform an analysis that reflects the current set of setting values. Then, in step S250, the analysis results are obtained from the backend unit 13b, and the display content of the main display unit V21 of the view V2 is updated based on the analysis results, as shown in FIG.

[0079] The analysis result chart displayed in view V1 and the analysis result chart displayed in view V2 have different hyperparameters (e.g., the number of hidden layers), so the analysis results are different and therefore the displayed content is also different.

[0080] Since views V1 and V2 are simultaneously displayed on the display 23, the user can view both views V1 and V2. This allows the user to easily compare the differences in the analysis results. For example, the user can visually determine which of the hyperparameters adopted in view V1 and the hyperparameters adopted in view V2 is superior. In this way, the user can view the results of different analyses performed on the same experimental data 12b, making it easy to find the better analysis. In other words, analyses using multiple similar analysis conditions are made more efficient.

[0081] 13 and 16 show examples of the display of views V1 and V2 when a single value is set for each of the setting value sets. However, if multiple setting values ​​are set for one of the setting value sets, multiple charts corresponding to the multiple setting values ​​may be displayed in one view.

[0082] For example, Fig. 17 shows an example in which two setting values ​​are set for each of two types of analysis conditions for view V1. In this example, Ridge and ElasticNet are set for the algorithm setting section 44, and Default and Middle are set for the hyperparameter setting section 45 on the analysis condition setting screen in Fig. 10. A single value is set for each of the other setting sections.

[0083] Upon receiving this set of setting values, the back-end unit 13b calculates data of four analysis results and transmits them to the client unit 25a. The client unit 25a then controls the display 23 to display the received data of the four analysis results in one view V1.

[0084] 17, graphs V11a, V11b, V11c, and V11d of 2×2=4 analysis results, which are combinations of the setting values ​​of the two algorithm setting units 44 and the setting values ​​of the two hyperparameter setting units 45, are displayed on the main display unit V11. More specifically, the graphs V11a to V11d are displayed in a matrix on the display unit V11.

[0085] At this time, the client unit 25a may highlight the analysis result with the highest accuracy evaluation index (i.e., the highest accuracy) among these four analysis results on the display 23 in a format different from that of the other three analysis results. A method for highlighting the analysis result in a format different from that of the other three analysis results may be, for example, a method of displaying it in a different color, a method of displaying it in a different size, or a method of displaying it in a different darkness. Alternatively, as shown in FIG. 17, a method of adding a marker MP to distinguish it from other diagrams may be used. In this way, it is possible to see at a glance which combination of algorithm and hyperparameter is optimal.

[0086] 17, the client unit 25a may display a save button V14, a display formula button V15, and a download button V16 in the view V1, which can be selected by the user using the operation device 24.

[0087] When the save button V14 is selected, the client unit 25a sends a query requesting that the view V1 be saved to the back-end unit 13b. Then, in accordance with the query, the back-end unit 13b records the data already received regarding the view V1 (i.e., the designation of the experimental data 12b, the type of analysis, and the set of setting values) and the analysis results regarding the view V1 in the storage medium 12. The analysis results recorded in this manner are displayed in the analysis result list D21 on the screen of FIG. 7, which has already been described.

[0088] When the display equation button V15 is selected, the client unit 25a displays the regression equation of the regression model included in the analysis result transmitted from the back-end unit 13b on the display 23. When the download button V16 is selected, the client unit 25a records the regression coefficients included in the analysis result transmitted from the back-end unit 13b in the storage medium 22.

[0089] During the display of the view V1, the client unit 25a also performs the process shown in Fig. 18 among the processes defined by the UI program 12a. That is, in step S410, the client unit 25a repeatedly determines whether or not the user has performed a predetermined movement operation on the view V1 using the operation device 24 until it determines that the user has performed the operation.

[0090] The predetermined movement operation for the view V1 may be, for example, an operation of clicking and holding the view V1 with a pointing device such as a mouse, moving the view V1 to a desired position within the canvas D30, and then releasing the click, i.e., a drag-and-drop operation. Alternatively, the predetermined movement operation for the view V1 may be some other operation. By this movement operation, the view V1 can be moved to any position within the canvas D30 in increments of one pixel on the display 23. Note that not only the view V1 but also other views within the canvas D30 can be moved by such a movement operation.

[0091] Furthermore, while the view V1 is being displayed, the client unit 25a can change the display size of the view V1 through some of the processes defined by the UI program 12a, in accordance with a predetermined operation by the user using the operation device 24. This also applies to views other than the view V1.

[0092] Furthermore, the display format of the chart in the view V1 can be changed without changing the content of the analysis result by the user's operation using the operation device 24. Specifically, while the view V1 is being displayed, the client unit 25a also performs the process shown in Fig. 19 among the processes defined by the UI program 12a.

[0093] In the processing of this figure, in step S510, the client unit 25a repeats determining whether the user has performed a predetermined display format change operation on a predetermined chart (e.g., a two-dimensional graph) in the view V1 using the operation device 24 until it determines that the operation has been performed.

[0094] As a predetermined display mode change operation for view V1, for example, a user may select mark W1 in Fig. 5 with a pointing device such as a mouse and then drag the target diagram. When this operation is performed, the client unit 25a performs a parallel movement of the display range (for example, the range of values ​​on the X-axis and the Y-axis) within the diagram in step S520.

[0095] 5 with a pointing device such as a mouse, and then selecting a portion of the target diagram using a rectangular shape. When this operation is performed, the client unit 25a expands the range of coordinates selected by the rectangular shape within the diagram to cover the entire display range of the diagram in step S520.

[0096] Furthermore, a predetermined display mode change operation targeting view V1 is, for example, an operation of selecting mark W3 in FIG. 5 with a pointing device such as a mouse to select a partial range of the target diagram. When this operation is performed, client unit 25a emphasizes (e.g., highlights) data points within the range in step S520. Here, data points refer to data included in the analysis results and represented by points in the diagram. Also, for example, when mark W4 in FIG. 5 is selected with a pointing device such as a mouse, client unit 25a can save an image of the target diagram in storage medium 22 in step S520.

[0097] In this way, the processing of steps S510 and S520 by the client unit 25a makes it possible to interactively change the display form inside each diagram or table by user operation. In steps S510 and S520, the client unit 25a does not send a query to the server 1. That is, the client unit 25a does not request analysis from the backend unit 13b again and does not receive the analysis results again. Therefore, the display form inside each diagram or table can be interactively changed by user operation, without the data of the analysis results being changed. Furthermore, the appearance of the display inside the diagram or table can be adjusted only on the user terminal 2 side, without requiring reanalysis on the server 1.

[0098] [Display of analysis results using Gaussian process regression of experimental data] Next, the display of analysis results by Gaussian process regression of experimental data will be described. This analysis is also active learning for searching for the next desirable experiment. Suppose the user selects the add view button D26 when the client unit 25a is displaying the analysis result display screen exemplified in FIG. 8, FIG. 13, etc. on the display 23. Then, based on this operation, the client unit 25a displays a view type selection screen (not shown) on the display 23 as described above. Here, suppose the user performs an operation to select Gaussian process regression.

[0099] Then, the client unit 25a displays a view V1 corresponding to the Gaussian process regression on the canvas of the display 23. The configuration of this view is the same as the above-mentioned view V1. When the user selects the above-mentioned setting button V13 for this view using the operation device 24, the client unit 25a displays an analysis condition setting screen shown in FIG. 20 on the display 23.

[0100] At this time, the processing executed by the client unit 25a is, among the processing prescribed by the UI program 12a, the processing shown in Fig. 11. That is, in step S120, the client unit 25a selects an analysis condition setting screen for Gaussian process regression as an analysis condition setting screen corresponding to the displayed view, and further in step S130, causes the display 23 to display the selected analysis condition setting screen.

[0101] As shown in Fig. 20, the analysis condition setting screen for Gaussian process regression is a screen for accepting setting operations for multiple types of analysis conditions used for Gaussian process regression. Specifically, the analysis condition setting screen includes multiple setting sections 61 to 67 and an OK button 68. The setting sections 61 to 67 are forms for accepting setting operations for setting values ​​for the analysis conditions, and the OK button 68 is a form for saving these setting values ​​as one setting value set. Of these setting sections 61 to 67, the explanatory variable setting section 65 and the descriptor setting section 66 are forms in a multi-select drop-down format.

[0102] Specifically, the acquisition function setting unit 61 is a form for setting the type of acquisition function in Gaussian process regression. The preprocessing setting unit 62 is a form for selecting the type of preprocessing for experimental data (e.g., eliminating missing data, etc.). The direction setting unit 63 is a form for setting the direction of optimization, i.e., whether to optimize in the direction of maximizing or minimizing the objective variable.

[0103] The objective variable setting section 64 is a form for selecting an objective variable from the selected experimental data. The explanatory variable setting section 65 is a form for individually selecting explanatory variables from the selected experimental data. The descriptor setting section 66 is a form for selecting a form of change of the explanatory variable selected in the explanatory variable setting section 65 (for example, changing the explanatory variable x to Log(x)).

[0104] Furthermore, the candidate setting unit 67 is a form for setting parameters of candidate experiments for the recommended experiment when identifying an experiment to be recommended through active learning using Gaussian process regression. When the recommended experiment is separately conducted and the experimental results are additionally registered as experimental data 12b in the storage medium 12 of the server 1, the experimental data 12b is used as a new learning subject in active learning based on Gaussian process regression. As a result, the accuracy of the analysis results obtained by Gaussian process regression is efficiently improved.

[0105] Specifically, the candidate setting unit 67 includes a plurality of radio buttons 67a and 67b and a table input field 67c. Either one of the radio buttons 67a and 67b can be exclusively selected by the user's operation using the operation device 24. When the radio button 67a is selected, the client unit 25a enters a random mode, displays an analysis condition setting screen for Gaussian process regression as shown in FIG. 20, and executes processing as shown in FIG. 21. When the radio button 67b is selected, the client unit 25a enters a user definition mode, displays an analysis condition setting screen for Gaussian process regression as shown in FIG. 22, and executes processing as shown in FIG. 23.

[0106] In the form on the screen of Fig. 20 and the form on the screen of Fig. 22, the setting sections 61 to 66 and the decision button 68 have the same functions, but the function of the candidate setting section 67 is partially different. Specifically, in the candidate setting section 67, the functions of the radio buttons 67a and 67b are the same, but the function of the table input field 67c is different. In addition, both the processing of Fig. 21 and the processing of Fig. 23 are defined by part of the UI program 12a.

[0107] In the random mode, first, in step S610, the client unit 25a accepts a change operation for a setting unit other than the candidate setting unit 67. That is, if such a change operation is performed, the change is reflected in the setting value set and the process proceeds to step S615, and if such a change operation is not performed, the process proceeds to step S615 without changing the setting value set.

[0108] In step S615, the client unit 25a identifies the maximum and minimum values ​​in the experimental data 12b currently being analyzed for each of one or more explanatory variables set in the explanatory variable setting unit 65, and sets these values ​​as the upper and lower limits of the search range, respectively. The upper and lower limits of the search range correspond to values ​​that indicate the search range.

[0109] The client unit 25a then determines the search interval based on the upper and lower limits. For example, the search interval is determined by dividing the difference between the upper and lower limits by a predetermined value (e.g., 10 or 100). Furthermore, as shown in FIG. 20, the upper and lower limits and search intervals for each explanatory variable are displayed in a table format in the table input field 67c. For example, a row is displayed for each explanatory variable, and in each row, the values ​​are arranged in ascending order of the column, in the order of the name of the target explanatory variable, the upper limit, the lower limit, and the search interval. In this way, the search range for recommended experiments can be determined using the experimental data 12b already used in this Gaussian process regression, allowing for efficient interpolative searches for recommended experiments.

[0110] For example, suppose that at a certain timing, the user performs a setting operation to increase or decrease the number of explanatory variables, and the client unit 25a increases or decreases the number of explanatory variables accordingly in step S610. Then, in step S615, the client unit 25a increases or decreases the number of rows in the table in accordance with the increase or decrease in the explanatory variables. Specifically, if one of the explanatory variables is deleted, the row corresponding to the deleted explanatory variable is deleted. Then, if a new explanatory variable is added, a row corresponding to the added explanatory variable is added, and its upper limit, lower limit, and search interval are calculated and displayed. In this example, multiple rows correspond to multiple one-dimensional cell arrays.

[0111] Following step S615, in step S620, it is determined whether or not the user has selected a specific cell in the table displayed in the table input field 67c using the operation device 24. If it is determined that the cell has been selected, the process proceeds to step S625, and if it is determined that the cell has not been selected, the process proceeds to step S630.

[0112] In step S625, input of changes to the value of the selected cell (e.g., upper limit value, lower limit value, search interval) is accepted, and the accepted changes are reflected in the setting value set, and the display content of the cell is changed. In this way, the value of each cell can be manually changed to a value desired by the user.

[0113] Next, in step S630, it is determined whether tabular data has been input into the table input field 67c by the user performing a copy and paste operation using the operation device 24. If it is determined that data has been input, the process proceeds to step S635, and if it is determined that data has not been input, the process proceeds to step S640.

[0114] For example, suppose that another program (e.g., spreadsheet software) created by a function other than that of the client unit 25a is being executed in the control circuit 25 of the user terminal 2. In this case, when tabular data created in the processing of the other program by the control circuit 25 is input into the table input field 67c by a user's copy-and-paste operation, the determination result in step S630 becomes YES. The tabular data created in the processing of the other program includes two or more cells adjacent to each other in a one-dimensional or two-dimensional matrix.

[0115] In step S635, the input tabular data is reflected in the table input field 67c. As a result, the multiple cells in the table input field 67c are changed according to the input tabular format, thereby changing the setting value set and changing the display content of the table input field 67c. In this way, multiple cells in the table input field 67c are changed all at once by copy and paste operations, reducing the effort required for the user to perform setting operations. After step S635, the process proceeds to step S640.

[0116] In step S640, it is determined whether or not the decision button 68 has been selected, and if not, the process returns to step S610, whereas if selected, the process proceeds to step S240. Steps S240 and onwards will be described later.

[0117] In the user definition mode, the client unit 25a first accepts a change operation for the setting units other than the candidate setting unit 67 in step S710, similar to step S610 in FIG.

[0118] In the next step S715, the client unit 25a displays in the table input field 67c a table having columns the number of which corresponds to one or more explanatory variables set in the explanatory variable setting unit 65. Specifically, a row is displayed for each candidate for a recommended experiment, and in each row, a cell is displayed for each explanatory variable, and these cells are arranged in an increasing column order.

[0119] Therefore, for example, suppose that at a certain timing, the user performs a setting operation to increase or decrease the number of explanatory variables, and the client unit 25a increases or decreases the number of explanatory variables accordingly in step S710. Then, in step S715, the client unit 25a increases or decreases the number of columns in the table in accordance with the increase or decrease in the explanatory variables. Specifically, if one of the explanatory variables is deleted, the column corresponding to the deleted explanatory variable is deleted, and if a new explanatory variable is added, the column corresponding to the added explanatory variable is added. Note that in this example, multiple columns correspond to multiple one-dimensional cell arrays.

[0120] Following step S715, in step S720, it is determined whether or not the user has selected a specific cell in the table displayed in the table input field 67c using the operation device 24. If it is determined that the cell has been selected, the process proceeds to step S725, and if it is determined that the cell has not been selected, the process proceeds to step S730.

[0121] In step S725, an input of a change to the value of the selected cell (that is, the value of the explanatory variable) is accepted, and the accepted change is reflected in the setting value set, and the display content of the cell is changed.

[0122] 21, it is determined whether tabular data has been input into the table input field 67c by the user performing a copy and paste operation using the operation device 24. If it is determined that data has been input, the process proceeds to step S735, and if it is determined that data has not been input, the process proceeds to step S740.

[0123] In step S735, the input tabular data is reflected in the table input field 67c. As a result, the multiple cells in the table input field 67c are changed according to the input tabular format, thereby changing the setting value set and changing the display content of the table input field 67c. In this way, multiple cells in the table input field 67c are changed all at once by copy and paste operations, reducing the effort required for the user to perform setting operations. After step S735, the process proceeds to step S740.

[0124] In step S740, it is determined whether or not the decision button 68 has been selected. If not, the process returns to step S710; if selected, the process proceeds to step S240.

[0125] In the random mode or the user-defined mode, as described above, when the user selects the decision button 68 using the operation device 24, the client unit 25a proceeds to step S240 in Fig. 21 or 23. The processing content of this step S240 is the same as the processing content of step S240 in Fig. 12.

[0126] That is, in step S240 in the random mode and the user-defined mode, an analysis that reflects the current set of setting values ​​is requested. Specifically, a query including information (e.g., name) specifying the selected experimental data 12b, the type of analysis to be performed (i.e., Gaussian process regression), and the current set of setting values ​​set by the setting units 61 to 67 is sent to the back-end unit 13b.

[0127] The back-end unit 13b that receives this query executes the type of analysis included in the query on the experimental data 12b specified in the query under the analysis conditions defined by the setting value set included in the query.

[0128] Specifically, a regression model is calculated using Gaussian process regression, and the value of the acquisition function is calculated for various values ​​of the explanatory variables, and the candidate with the largest value of the acquisition function is identified as the recommended experiment from among the candidates for recommended experiments set by the setting unit 67. Then, an answer including data on the execution result of this analysis (i.e., the analysis result) is sent to the client unit 25a.

[0129] In addition, when a random mode query is received, the backend unit 13b repeatedly calculates the value of the acquisition function for each of the multiple explanatory variables shown in the table input field 67c, shifting the value by the search interval from the lower limit value to the upper limit value shown in the table input field 67c.

[0130] In addition, when a query in user-defined mode is received, the backend unit 13b repeatedly calculates the value of the acquisition function for each of the multiple explanatory variables shown in the table input field 67c and for each of the multiple candidates shown in the table input field 67c.

[0131] Upon receiving this answer, the client unit 25a proceeds from step S240 to step S250. The processing content of this step S240 is the same as the processing content of step S240 in FIG.

[0132] That is, in step S250 in the random mode or user-defined mode, the client unit 25a generates a diagram image using the analysis result data in the answer. Then, as shown in Fig. 24, the generated diagram image is displayed on the display 23 as the main display section in the view. In the example of Fig. 24, the displayed diagram is a table in which the top candidates for which the values ​​of the acquisition function have been calculated are arranged in ascending order of the values ​​of the acquisition function, with the rows increasing. Each row displays a variable (e.g., the content of a material constituting a product) for the target candidate.

[0133] The operations for interactive manipulation of the displayed view, changing the set of setting values, etc. are the same as those for the views V1 and V2 displayed in the regression.

[0134] The back-end unit 13b can also perform analyses other than regression and Gaussian process regression. The client unit 25a can set a set of setting values ​​for each of the analyses that can be performed by the back-end unit 13b through the user's setting operation using the operation device 24, as described above. The client unit 25a can receive analysis results based on these setting value sets from the back-end unit 13b, as with regression and Gaussian process regression, and can display a chart showing the received analysis results in a view within the canvas D30.

[0135] An example of a chart displayed in the view is a graph showing the results of classification in machine learning using a neural network or the like, as shown in Figure 25. This graph shows the actual measured values ​​and predicted values ​​of a response variable, called a class, which should be classified into 0, 1, or 2, as points.

[0136] There is also a network diagram showing the results of statistical causal exploration, as shown in Fig. 26. For this network diagram, the user can operate a pull-down menu using the operation device 24 to select the layout of the network diagram from a plurality of preset layouts.

[0137] Furthermore, examples of charts displayed in the view as described above include a combination of a scatter plot 72 and a table 73, as shown in FIG. 27. These are displayed in the same view. In the scatter plot 72, each of a plurality of data points is represented by a dot in a two-dimensional graph as an analysis result of the experimental data 12b. The table 73 consists of a plurality of rows corresponding one-to-one to the plurality of data points. Each row indicates the parameter value at the corresponding data point. The correspondence between the dots in the scatter plot 72 and the rows in the table 73 is included in the analysis result data received from the back-end unit 13b.

[0138] When the view of Fig. 27 is displayed, if the user selects any dot in the scatter plot 72 using the operation device 24, the client unit 25a proceeds from step S510 to step S520 in the processing of Fig. 19. Then, in step S520, the row in the table 73 corresponding to the selected dot is highlighted. Conversely, if the user selects a row in the table 73 using the operation device 24, the client unit 25a proceeds from step S510 to step S520 in the processing of Fig. 19. Then, in step S520, the dot in the scatter plot 72 corresponding to the selected row is highlighted. In this way, the highlighting of data points is linked in multiple charts in the same view, making it easy to check the analysis results.

[0139] Furthermore, one of the charts displayed in the view as described above is a statistical table as shown in Fig. 28. This table shows statistical quantities obtained as a result of analyzing the experimental data 12b in the back-end unit 13b. In the example of Fig. 28, statistical quantities such as the average content, standard deviation, minimum value, 50th percentile value, and maximum value are shown for each component constituting a certain material.

[0140] Furthermore, examples of diagrams displayed in the view as described above include pair plots such as those shown in Figure 29. Figure 29 illustrates a pair plot for two variables (for example, the contents of two components that make up a material), but pair plots for three or more variables may also be displayed. Note that a pair plot is a matrix of scatter plots and histograms, but the diagram displayed in the view may simply be a histogram.

[0141] Another example of a chart displayed in the view as described above is a heat map as shown in Figure 30. In Figure 30, correlation coefficients between variables are displayed in a matrix using numbers and density. Note that the correlation coefficients may be displayed using color instead of density.

[0142] Furthermore, a single view showing the regression results may include a two-dimensional graph of the predicted and actual values ​​of the dependent variable, as well as a bar graph showing the importance of each variable (for example, regression coefficient, feature importance), as shown in Fig. 31. Furthermore, multiple such graphs may be arranged as multiple views within a single canvas D30, as shown in Fig. 32.

[0143] 25 to 32, the client unit 25a makes it possible to interactively change the internal display form of each diagram or table by user operation through the processes of steps S510 and S520 in FIG.

[0144] As described above, when a setting operation for a plurality of types of analysis conditions is performed, the client unit 25a acquires a first set of setting values ​​corresponding to the plurality of types of analysis conditions in accordance with the setting operation. Then, the client unit 25a displays the analysis results corresponding to the acquired first set of setting values ​​on the display 23 as view V1.

[0145] Furthermore, based on a copy operation targeting view V1, client unit 25a displays the analysis results corresponding to a second set of setting values, which have the same values ​​as the first set of setting values, as view V2 together with view V1 on display 23. Furthermore, based on an operation to change part of the second set of setting values, client unit 25a reflects the analysis results corresponding to the changed second set of setting values ​​in view V2.

[0146] In this way, in order to add a chart that has only part of the condition settings changed to a chart that is displayed based on certain condition settings, a copy operation and an operation to change part of the second set of setting values ​​are performed. Therefore, the operation is simpler than when setting the entire second set of setting values. This is because the amount of user operation required for the copy operation on the operation device 24 is sufficiently smaller (for example, half or less) than the amount of user operation required to set the entire second set of setting values.

[0147] (1) The client unit 25a can also simultaneously accept multiple setting values ​​for one type of analysis condition among multiple types of analysis conditions as part of the first setting value set, for example, in a multi-select drop-down format, and include multiple charts corresponding to the multiple setting values ​​for the same type of analysis condition in the view V1 as analysis results according to the first setting value set.

[0148] In this way, by simply setting multiple setting values ​​for the same type of analysis conditions within one setting value set, multiple corresponding charts can be displayed. This reduces the effort required to display multiple charts compared to creating multiple setting value sets. Furthermore, displaying multiple charts with some setting values ​​different makes it easier for users to evaluate the analysis results.

[0149] (2) Furthermore, view V1 includes a diagram displaying a new learning object selected by active learning based on Gaussian process regression, as shown in Fig. 24. Furthermore, client unit 25a displays, on display 23, a candidate setting unit 67 that displays, in a tabular format, candidates for a new learning object selected by the active learning, and enables the user to set values ​​for individual cells in the tabular display. Furthermore, client unit 25a enables tables created by a function other than the function of the setting value acquisition unit to be reflected in the tabular display all at once.

[0150] In this way, a table creation function separate from that of the client unit 25a can be seamlessly used for setting learning target candidates in the client unit 25a. Conversely, the control circuit 25 may input the display content of the table input field 67c of the candidate setting unit 67 as a table for the processing of the separate program described above by copy and paste processing.

[0151] (3) View V1 also includes a chart displaying a new learning target selected by active learning based on Gaussian process regression. The first set of setting values ​​includes values ​​indicating a search range for the explanatory variables used in the Gaussian process regression, which are candidates for the new learning target to be used in the Gaussian process regression. When the user specifies an explanatory variable, client unit 25a sets the value indicating the search range for the explanatory variable based on the value of the explanatory variable in the learning data already used in the Gaussian process regression.

[0152] In this way, the search range for a new learning object can be determined using the training data that has already been used in this Gaussian process regression, so that the search for a new learning object can be carried out efficiently.

[0153] (4) View V1 also includes a diagram displaying a new learning object selected by active learning based on Gaussian process regression. Multiple explanatory variables used in Gaussian process regression can be set for the first set of setting values. Client unit 25a also displays, on display 23, a candidate setting unit 67 that displays candidates for a new learning object selected by active learning in a tabular format. In the tabular format, multiple one-dimensional cells (rows or columns) correspond one-to-one to multiple explanatory variables.

[0154] Furthermore, when the number of explanatory variables is changed by a user operation, the client unit 25a changes the number of one-dimensional cell arrays in the same way. In this way, the number of one-dimensional cell arrays (i.e., the number of rows or columns) changes in conjunction with the change in the number of explanatory variables, thereby saving the user time and effort.

[0155] (5) The view V1 also includes one or more of the following charts: regression, classification, statistical causal discovery, Gaussian process regression, active learning, a statistical list, a scatter plot, a histogram, a pair plot, and a heat map. The internal display format of the one or more charts can be interactively changed by a user operation without sending a query to the server 1 for re-analysis. This allows the user terminal 2 to adjust the internal display appearance of the chart.

[0156] (6) Furthermore, the client unit 25a can move the position of the view V1 in units of one pixel on the display 23 in response to a user's operation on the operation device 24. This allows the position of the view V1 to be adjusted with a high degree of freedom.

[0157] (7) The client unit 25a also functions as a web browser and displays the view V1 within the display range of the web browser. Since web browsers are available for many operating systems, this improves user convenience.

[0158] (8) The analysis results corresponding to each set of setting values ​​are the results of material analysis in materials development support. In the field of materials development support, a suitable development style is to compare the analysis results for multiple sets of setting values ​​with slightly different analysis condition settings.

[0159] Note that the features of view V1 in (1) to (8) above are similarly possessed by view V2 and other views. Furthermore, in the above embodiment, the control circuit 25 of the user terminal 2 executes steps S230, S610 to S635, and S710 to S735, thereby functioning as a setting value acquisition unit. Furthermore, the control circuit 25 executes steps S250, S320, S420, and S520, thereby functioning as a view display unit.

[0160] (Other embodiments) The present invention is not limited to the above-described embodiments and can be modified as appropriate. The above-described embodiments are not unrelated to each other and can be combined as appropriate unless the combination is clearly impossible. In the above-described embodiments, the elements constituting the embodiments are not necessarily essential unless expressly stated as essential or clearly considered essential in principle. In the above-described embodiments, when numerical values ​​such as the number, value, amount, and range of components of the embodiments are mentioned, they are not limited to the specific number unless expressly stated as essential or clearly limited to a specific number in principle. In particular, when multiple values ​​are exemplified for a certain quantity, values ​​between those multiple values ​​can be adopted unless otherwise specified or clearly impossible in principle. In the above-described embodiments, when the shape, positional relationship, etc. of components are mentioned, they are not limited to the shape, positional relationship, etc., unless expressly stated or limited to a specific shape, positional relationship, etc. in principle. The present invention also allows the following modifications and modifications within equivalent ranges to the above-described embodiments. The following modifications can be independently applied or inapplicable to the above-described embodiments. That is, any combination of the modified examples described below can be applied to the above embodiment.

[0161] The control unit and the method described herein may be implemented by a special-purpose computer configured by configuring a processor and memory programmed to perform one or more functions embodied in a computer program. Alternatively, the control unit and the method described herein may be implemented by a special-purpose computer configured by configuring a processor with one or more dedicated hardware logic circuits. Alternatively, the control unit and the method described herein may be implemented by one or more special-purpose computers configured by combining a processor and memory programmed to perform one or more functions with a processor configured with one or more hardware logic circuits. Furthermore, the computer program may be stored as instructions executed by a computer on a computer-readable non-transitory tangible storage medium.

[0162] (Variation 1) In the above embodiment, the client unit 25a functions as a Web client, but the client unit 25a does not have to be a Web client, and in that case, the Web service unit 13a does not have to be a Web server.

[0163] (Variation 2) In the above embodiment, the data analyzed by the server 1 and the analysis results transmitted to the user terminal are experimental data 12b related to materials for supporting materials development. Therefore, the diagrams displayed by the client unit 25a are also analysis results related to materials for supporting materials development. However, the data analyzed by the server 1 and the analysis results displayed by the client unit 25a are not limited to experimental results for supporting materials development, and may be any analysis target data.

[0164] (Variation 3) In the above embodiment, the server 1 is configured with one computer. However, the server 1 may be configured with multiple computers that communicate with each other. In that case, the control circuit 13 that configures the server 1 may be distributed among the multiple computers. The same applies to the communication interface circuit 11 and the storage medium 12.

[0165] (Variation 4) In the above embodiment, the client unit 25a communicates with the backend unit 13b without going through the Web service unit 13a. However, the client unit 25a may communicate with the backend unit 13b via the Web service unit 13a. For example, the Web service unit 13a may function as a Web server in a three-tier Web structure, and the backend unit 13b may function as an application server in the three-tier Web structure.

[0166] (Features of the invention) [Claim 1] A user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a display (23) for displaying images to the user; an operation device (24) that accepts the user's operation; a control circuit (25); The control circuit a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that defines a plurality of types of analysis conditions in response to a setting operation performed by the user on the operation device for the plurality of types of analysis conditions; a view display unit (S250, S320, S420, S520) that displays an analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit on the display as a first view (V1); the view display unit, based on the user performing a copy operation on the operation device targeting the first view, causes the display to display, together with the first view, an analysis result corresponding to a second setting value set having values ​​identical to those of the first setting value set as a second view (V2), and further, based on the user performing an operation on the operation device to change a part of the second setting value set, causes the view display unit to reflect, in the second view, the analysis result corresponding to the changed second setting value set.

[0167] [Claim 2] the setting value acquisition unit is capable of simultaneously accepting a plurality of setting values ​​for one type of analysis condition among the plurality of types of analysis conditions as part of the first setting value set; 2. The user terminal according to claim 1, wherein the view display unit includes, in the first view, a plurality of charts corresponding to the plurality of setting values, respectively, as analysis results according to the first set of setting values.

[0168] [Claim 3] the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; 3. The user terminal according to claim 1, wherein the setting value acquisition unit displays on the display a candidate setting unit (67) that displays candidates for new learning targets to be selected in the active learning in a tabular format, enables the user to set values ​​for individual cells in the tabular display, and further enables a table created by a function other than the function of the setting value acquisition unit to be reflected in the tabular display all at once.

[0169] [Claim 4] the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; the first set of setting values ​​includes values ​​indicating a search range for explanatory variables used in the Gaussian process regression, the values ​​being candidates for new learning targets to be used in the Gaussian process regression; 3. The user terminal according to claim 1, wherein, when the user specifies explanatory variables to be used in the Gaussian process regression, the setting value acquisition unit sets values ​​indicating the search range of the explanatory variables based on values ​​of the explanatory variables in training data already used in the Gaussian process regression.

[0170] [Claim 5] the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; a plurality of explanatory variables to be used in the Gaussian process regression can be set for the first setting value set; 3. The user terminal according to claim 2, wherein the setting value acquisition unit displays on the display a candidate setting unit (67) that presents candidates for new learning targets to be selected in the active learning in a tabular format, and in the tabular format, corresponds a plurality of one-dimensional cells, which are a plurality of rows or a plurality of columns, to a plurality of explanatory variables, respectively, and when the number of the plurality of explanatory variables is changed by an operation of the user, similarly changes the number of the plurality of one-dimensional cell arrangements.

[0171] [Claim 6] 6. The user terminal of claim 1, wherein the first view includes one or more of a regression, classification, statistical causal discovery, Gaussian process regression, active learning, a statistical summary, a scatter plot, a histogram, a pair plot, and a heat map, and the internal display form of the one or more charts can be interactively changed by a user operation.

[0172] [Claim 7] 7. The user terminal according to claim 1, wherein the first view is movable in units of one pixel on the display in response to a user's operation on the operation device.

[0173] [Claim 8] 8. The user terminal according to claim 1, wherein the view display unit functions as a web browser and displays the first view within a display area of ​​the web browser.

[0174] [Claim 9] 9. The user terminal according to claim 1, wherein the analysis results corresponding to the first set of setting values ​​are results of analysis of materials in material development support.

[0175] [Claim 10] A program used in a user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; causing the user terminal to function as a view display unit (S250, S320, S420, S520) that displays the analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, based on the user performing a copy operation on the operation device targeting the first view, causes the display to display, together with the first view, an analysis result corresponding to a second setting value set having values ​​identical to those of the first setting value set as a second view (V2), and further, based on the user performing an operation on the operation device to change a part of the second setting value set, causes the second view to reflect the analysis result corresponding to the changed second setting value set.

[0176] [Claim 11] A server that communicates with a user terminal (1), a communication interface circuit (11) for communicating with the user terminal; a storage medium (12) for storing analysis target data (12b); a control circuit (13) that analyzes the data and transmits the analysis result to the user terminal using the communication interface circuit; The storage medium (12) stores a program (12a) to be executed on the user terminal, the control circuit transmits the program to the user terminal using the interface circuit; The program causes the user terminal to a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; functioning as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, based on the user performing a copy operation on the operation device targeting the first view, causes the display to display, together with the first view, an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2); and, based on the user performing an operation on the operation device to change a part of the second setting value set, causes the view display unit to reflect, in the second view, the analysis result corresponding to the changed second setting value set. [Explanation of symbols]

[0177] 1 server 2. User terminal 25 Control circuit 12a UI Program 12b Experimental Data 13a Web Services Department 13b Back-end section

Claims

1. A user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a display (23) for displaying images to a user; an operation device (24) that accepts an operation by the user; a control circuit (25); The control circuit a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires, based on a setting operation performed by the user on the operation device regarding a plurality of types of analysis conditions, a first setting value set that defines the plurality of types of analysis conditions in accordance with the setting operation; a view display unit (S250, S320, S420, S520) that displays an analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit on the display as a first view (V1), the view display unit, based on the user's operation of the operation device to copy the first view, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of the operation device to change a part of the second setting value set, reflects the analysis result corresponding to the changed second setting value set in the second view; the setting value acquisition unit is capable of simultaneously receiving a plurality of setting values ​​for one type of analysis condition among the plurality of types of analysis conditions as part of the first setting value set; The view display unit includes, in the first view, a plurality of charts corresponding to the plurality of setting values, respectively, as analysis results according to the first setting value set.

2. the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; The user terminal of claim 1, wherein the setting value acquisition unit displays on the display a candidate setting unit (67) that displays in tabular form candidates for new learning targets to be selected in the active learning, and enables the user to set values ​​for individual cells in the tabular display, and further enables tables created by a function other than the function of the setting value acquisition unit to be reflected in the tabular display all at once.

3. the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; the first set of setting values ​​includes values ​​indicating a search range for explanatory variables used in the Gaussian process regression, the values ​​being candidates for new learning targets to be used in the Gaussian process regression; 2. The user terminal according to claim 1, wherein, when the user specifies explanatory variables to be used in the Gaussian process regression, the setting value acquisition unit sets values ​​indicating the search range of the explanatory variables based on values ​​of the explanatory variables in training data already used in the Gaussian process regression.

4. the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; a plurality of explanatory variables to be used in the Gaussian process regression can be set for the first setting value set; 2. The user terminal of claim 1, wherein the setting value acquisition unit displays on the display a candidate setting unit (67) that presents candidates for new learning targets to be selected in the active learning in a tabular format, and in the tabular format, corresponds a plurality of one-dimensional cells, which are a plurality of rows or a plurality of columns, to the plurality of explanatory variables, respectively, and when the number of the plurality of explanatory variables is changed by the user's operation, similarly changes the number of the plurality of one-dimensional cell arrangements.

5. A user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a display (23) for displaying images to a user; an operation device (24) that accepts an operation by the user; a control circuit (25); The control circuit a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires, based on a setting operation performed by the user on the operation device regarding a plurality of types of analysis conditions, a first setting value set that defines the plurality of types of analysis conditions in accordance with the setting operation; a view display unit (S250, S320, S420, S520) that displays an analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit on the display as a first view (V1), the view display unit, based on the user's operation of the operation device to copy the first view, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of the operation device to change a part of the second setting value set, reflects the analysis result corresponding to the changed second setting value set in the second view; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; The setting value acquisition unit displays on the display a candidate setting unit (67) that displays candidates for new learning targets to be selected in the active learning in a tabular format, and enables the user to set values ​​for individual cells in the tabular display, and further enables tables created by a function other than the function of the setting value acquisition unit to be reflected in the tabular display all at once.

6. A user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a display (23) for displaying images to a user; an operation device (24) that accepts an operation by the user; a control circuit (25); The control circuit a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires, based on a setting operation performed by the user on the operation device regarding a plurality of types of analysis conditions, a first setting value set that defines the plurality of types of analysis conditions in accordance with the setting operation; a view display unit (S250, S320, S420, S520) that displays an analysis result corresponding to the first set of setting values ​​acquired by the setting value acquisition unit on the display as a first view (V1), the view display unit, based on the user's operation of the operation device to copy the first view, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of the operation device to change a part of the second setting value set, reflects the analysis result corresponding to the changed second setting value set in the second view; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; the first set of setting values ​​includes values ​​indicating a search range for explanatory variables used in the Gaussian process regression, the values ​​being candidates for new learning targets to be used in the Gaussian process regression; the setting value acquisition unit, when the user specifies explanatory variables to be used in the Gaussian process regression, sets values ​​indicating the search range of the explanatory variables based on values ​​of the explanatory variables in training data already used in the Gaussian process regression.

7. 7. The user terminal of claim 1, wherein the first view includes one or more diagrams of regression, classification, statistical causal discovery, Gaussian process regression, a learning object selected by active learning, a statistical list, a scatter plot, a histogram, a pair plot, and a heat map, and the internal display form of the one or more diagrams can be interactively changed by the user's operation.

8. 7. The user terminal according to claim 1, wherein the first view is movable in units of one pixel on the display in response to an operation of the user on the operation device.

9. 7. The user terminal according to claim 1, wherein the view display unit functions as a web browser and displays the first view within a display area of ​​the web browser.

10. 7. The user terminal according to claim 1, wherein the analysis results corresponding to the first set of setting values ​​are results of analysis of materials in materials development support.

11. A program used in a user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; and causing the user terminal to function as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, based on the user's operation of the operation device to copy the first view, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of the operation device to change a part of the second setting value set, reflects the analysis result corresponding to the changed second setting value set in the second view; the setting value acquisition unit is capable of simultaneously receiving a plurality of setting values ​​for one type of analysis condition among the plurality of types of analysis conditions as part of the first setting value set; The view display unit includes, in the first view, a plurality of charts corresponding to the plurality of setting values, respectively, as analysis results according to the first setting value set.

12. A program used in a user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; and causing the user terminal to function as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, based on the user's operation of the operation device to copy the first view, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of the operation device to change a part of the second setting value set, reflects the analysis result corresponding to the changed second setting value set in the second view; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; The setting value acquisition unit displays on the display a candidate setting unit (67) that shows candidates for new learning targets to be selected in the active learning in a tabular format, and enables the user to set values ​​for individual cells in the tabular display, and further enables tables created by a function other than the function of the setting value acquisition unit to be reflected in the tabular display all at once.

13. A program used in a user terminal that receives analysis results from a server (1) that acquires and analyzes analysis target data (12b), a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; and causing the user terminal to function as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, based on the user's operation of the operation device to copy the first view, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of the operation device to change a part of the second setting value set, reflects the analysis result corresponding to the changed second setting value set in the second view; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; the first set of setting values ​​includes values ​​indicating a search range for explanatory variables used in the Gaussian process regression, the values ​​being candidates for new learning targets to be used in the Gaussian process regression; the setting value acquisition unit, when the user specifies explanatory variables to be used in the Gaussian process regression, sets values ​​indicating the search range of the explanatory variables based on values ​​of the explanatory variables in training data already used in the Gaussian process regression.

14. A server that communicates with a user terminal (1), a communication interface circuit (11) for communicating with said user terminal; a storage medium (12) for storing analysis target data (12b); a control circuit (13) that analyzes the analysis target data and transmits the analysis result to the user terminal using the communication interface circuit; The storage medium (12) stores a program (12a) to be executed on the user terminal, the control circuit transmits the program to the user terminal using the communication interface circuit; The program causes the user terminal to a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; and functioning as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, based on the user's operation of the operation device to copy the first view, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of the operation device to change a part of the second setting value set, reflects the analysis result corresponding to the changed second setting value set in the second view; the setting value acquisition unit is capable of simultaneously receiving a plurality of setting values ​​for one type of analysis condition among the plurality of types of analysis conditions as part of the first setting value set; The view display unit includes, in the first view, a plurality of charts corresponding to the plurality of setting values, respectively, as analysis results according to the first setting value set.

15. A server that communicates with a user terminal (1), a communication interface circuit (11) for communicating with said user terminal; a storage medium (12) for storing analysis target data (12b); a control circuit (13) that analyzes the analysis target data and transmits the analysis result to the user terminal using the communication interface circuit; The storage medium (12) stores a program (12a) to be executed on the user terminal, the control circuit transmits the program to the user terminal using the communication interface circuit; The program causes the user terminal to a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; and functioning as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, based on the user's operation of the operation device to copy the first view, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of the operation device to change a part of the second setting value set, reflects the analysis result corresponding to the changed second setting value set in the second view; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; The setting value acquisition unit displays on the display a candidate setting unit (67) that shows candidates for new learning targets to be selected in the active learning in a tabular format, and enables the user to set values ​​for individual cells in the tabular display, and further enables tables created by a function other than the function of the setting value acquisition unit to be reflected in the tabular display all at once.

16. A server that communicates with a user terminal (1), a communication interface circuit (11) for communicating with said user terminal; a storage medium (12) for storing analysis target data (12b); a control circuit (13) that analyzes the analysis target data and transmits the analysis result to the user terminal using the communication interface circuit; The storage medium (12) stores a program (12a) to be executed on the user terminal, the control circuit transmits the program to the user terminal using the communication interface circuit; The program causes the user terminal to a setting value acquisition unit (S230, S610 to S635, S710 to S735) that acquires a first setting value set that determines a plurality of types of analysis conditions in response to a setting operation performed by a user on an operation device (24) for the plurality of types of analysis conditions; and functioning as a view display unit (S250, S320, S420, S520) that displays the analysis result according to the first set of setting values ​​acquired by the setting value acquisition unit as a first view (V1) on a display (23); the view display unit, based on the user's operation of the operation device to copy the first view, displays an analysis result corresponding to a second setting value set having the same values ​​as the first setting value set as a second view (V2) on the display together with the first view, and further, based on the user's operation of the operation device to change a part of the second setting value set, reflects the analysis result corresponding to the changed second setting value set in the second view; the first view includes a chart displaying a new learning target selected by active learning based on Gaussian process regression; the first set of setting values ​​includes values ​​indicating a search range for explanatory variables used in the Gaussian process regression, the values ​​being candidates for new learning targets to be used in the Gaussian process regression; the setting value acquisition unit, when the user specifies explanatory variables to be used in the Gaussian process regression, sets values ​​indicating the search range of the explanatory variables based on values ​​of the explanatory variables in training data already used in the Gaussian process regression.

Images