Knowledge-creation assistance device and knowledge-creation assistance method

Abstract

A knowledge-creation assistance device (100) comprises: a knowledge-structuring unit (111) that, on the basis of a measurement analysis result of a sample outputted from an analysis device, generates pre-complementary structured knowledge that is configured to include one or more items pertaining to measurement; a missing-knowledge estimation unit (113) that adds an item to the pre-complementary structured knowledge to generate post-complementary structured knowledge; and a knowledge-editing unit (114) that receives an editing instruction for the post-complementary structured knowledge and generates confirmed structured knowledge. The missing-knowledge estimation unit (113) adds the item to the pre-complementary structured knowledge by using an estimation model that is a machine learning model in which an item included in the pre-complementary structured knowledge is used as an explanatory variable, and an item included in the confirmed structured knowledge is used as an objective variable.

Description

Knowledge creation support device and knowledge creation support method

[0001] The present invention relates to a knowledge creation support device and a knowledge creation support method for supporting the creation of structured knowledge based on measurement analysis results output by an analysis device.

[0002] Materials analysis requires a lot of know-how, and it is necessary to accumulate and utilize that know-how. Know-how is generated during analysis work, but it is difficult to allocate many man-hours to accumulating it. For this reason, there is a demand to reduce the man-hours required to accumulate know-how.

[0003] A technology related to know-how accumulation is a technology for similarity analysis and data enrichment using a knowledge source described in Patent Literature 1. A data enrichment service according to this technology can identify similar related data by comparing an input dataset with a reference dataset stored in a knowledge source. A similarity metric may be calculated to correspond to the semantic similarity of two or more datasets. Using the similarity metric, datasets can be identified based on the metadata attributes and data values ​​of these datasets, which can simplify indexing and high-performance searches for data values. The input dataset can be labeled with a category based on the dataset that best matches the input dataset. Additional information about the dataset can be obtained by querying the knowledge source using the similarity between the input dataset and a dataset provided by the knowledge source. Recommendations may be provided to users using the additional information.

[0004] Special table 2017-536601 publication

[0005] The technique described in U.S. Patent No. 6,277,633 provides a more sophisticated similarity metric that can be used to improve the automatic identification of highly relevant datasets that have semantic similarity to an input dataset. By identifying more similar related datasets, the input dataset may be enriched with data from related datasets. Enriching the input dataset allows users to understand and manage large amounts of data that can be difficult to manage.

[0006] However, because the knowledge source, which is structured knowledge, is not updated, there is a problem that acquired data, such as measurement and analysis results of new samples, cannot be expanded with high accuracy. The present invention has been made in view of this background, and an object of the present invention is to provide a knowledge creation support device and a knowledge creation support method that efficiently build structured knowledge based on acquired data.

[0007] In order to solve the above-mentioned problems, the knowledge creation support device of the present invention comprises a knowledge structuring unit that generates pre-complementation structured knowledge consisting of one or more items related to measurement based on the measurement and analysis results of a sample output from an analytical device, a missing knowledge estimation unit that adds items to the pre-complementation structured knowledge to generate complemented structured knowledge, and a knowledge editing unit that accepts editing instructions for the complemented structured knowledge and generates confirmed structured knowledge, wherein the missing knowledge estimation unit adds items to the pre-complementation structured knowledge using an estimation model that is a machine learning model that uses items included in the pre-complementation structured knowledge as explanatory variables and items included in the confirmed structured knowledge as objective variables.

[0008] According to the present invention, it is possible to provide a knowledge creation support device and a knowledge creation support method that improve the efficiency of building structured knowledge based on acquired data. Problems, configurations, and effects other than those described above will become clear from the description of the following embodiments.

[0009] FIG. 1 is a functional block diagram of a knowledge creation support device according to a first embodiment. FIG. 2 is a diagram showing the data configuration of measurement analysis results stored in a measurement analysis result database according to the first embodiment. FIG. 3 is a graph showing a DSC curve according to the first embodiment. FIG. 4 is a diagram showing the configuration of structured knowledge stored in a structured knowledge database according to the first embodiment. FIG. 5 is an example of pre-complementation structured knowledge according to the first embodiment. FIG. 6 is an example of post-complementation structured knowledge according to the first embodiment. FIG. 7 is a screen configuration diagram of a knowledge editing screen according to the first embodiment. FIG. 8 is a flowchart of knowledge creation support processing according to the first embodiment. FIG. 9 is a functional block diagram of a knowledge creation support device according to a second embodiment. FIG. 10 is a screen configuration diagram of an analysis project management screen according to the second embodiment. FIG. 11 is an example of an analysis report template according to the second embodiment. FIG. 12 is a hardware configuration diagram showing an example of a computer that realizes the functions of the knowledge creation support device according to the above-mentioned embodiments.

[0010] <<Outline of the Knowledge Creation Support Device>> An outline of the knowledge creation support device in a mode (embodiment) for carrying out the present invention will be described below. The knowledge creation support device generates pre-complemented structured knowledge (see FIG. 5 , described later), which is structured knowledge, based on measurement and analysis results (see FIG. 2 , described later) that are the results of measuring and analyzing a sample by an analytical device. Next, the knowledge creation support device generates post-complemented structured knowledge (see FIG. 6 , described later) based on the pre-complemented structured knowledge using an estimation model, which is a machine learning model.

[0011] Next, the user edits the completed structured knowledge. The knowledge creation support device stores the edited completed structured knowledge in the structured knowledge database as confirmed structured knowledge. Here, the estimation model is a machine learning model trained using the structured knowledge in the structured knowledge database as learning data, and is used to estimate some of the knowledge missing from the uncompleted structured knowledge.

[0012] With this knowledge creation support device, users can efficiently create and accumulate structured knowledge based on the measurement and analysis results of samples. As the accumulation of structured knowledge progresses, it is expected that the estimation accuracy of the estimation model will improve and the efficiency of creating structured knowledge will also improve. Below, an embodiment will be described using thermal analysis as an example.

[0013] <Configuration of Knowledge Creation Support Device> Figure 1 is a functional block diagram of a knowledge creation support device 100 according to a first embodiment. The knowledge creation support device 100 is a computer and includes a control unit 110, a storage unit 120, and an input / output unit 180. A user interface device 290, such as a display, keyboard, or mouse, and an analysis device 200 (thermal analysis device) that performs thermal analysis are connected to the input / output unit 180. The input / output unit 180 includes a communication device and is capable of receiving data from the analysis device 200 (see measurement analysis results 410 in Figure 2 described below). A media drive may also be connected to the input / output unit 180, enabling data exchange using a recording medium.

[0014] <<Knowledge Creation Support Device: Storage Unit>> The storage unit 120 is configured to include storage devices such as a ROM (Read Only Memory), a RAM (Random Access Memory), and an SSD (Solid State Drive). The storage unit 120 stores a material database 130, a measurement and analysis result database 140, a structured knowledge database 150, an estimation model database 121, and a program 128. The program 128 includes a description of processing procedures related to the functional units of the control unit 110 described below. Note that the various storage contents of the storage unit 120 may be stored in an external storage device such as a cloud server and read as needed.

[0015] <Storage Unit: Material Database> The material database 130 stores information related to materials that are the subject of structured knowledge 430 (see FIG. 4, which will be described later), such as material identification information, classification system, chemical structure, physical properties, etc., in association with each other. The materials analyzed by the analysis device 200 are also called specimens or samples.

[0016] <Memory unit: Measurement and analysis result database> The measurement and analysis result database 140 stores the measurement and analysis results 410 (see Figure 2 described below), which are the results of the analysis of the sample by the analysis device 200, in association with the identification information of the measurement and analysis results 410 and the identification information of the sample.

[0017] 2 is a diagram showing the data structure of measurement and analysis results 410 stored in the measurement and analysis result database 140 according to the first embodiment. The measurement and analysis results 410 are the results of measurement and analysis performed by thermally analyzing a material (specimen, sample) using the analytical device 200 (e.g., a differential scanning calorimeter). The measurement and analysis results 410 include items such as the measurement date and time, sample name, sample mass, temperature program, comments 413, peaks 414, and calorimetry results 415.

[0018] 3 is a graph 420 showing a DSC (Differential Scanning Calorimetry) curve according to the first embodiment. The vertical axis of the graph 420 represents heat flow, and the horizontal axis represents temperature. Note that the horizontal axis may also represent time. The graph 420 is a graph in which the temperature and heat quantity in the calorific measurement result 415 included in the measurement analysis result 410 are plotted. A peak 414 in the measurement analysis result 410 is the result (analysis feature) of the temperature and heat flow at the peak included in the graph 420 calculated by the analysis device 200 analyzing the calorific measurement result 415.

[0019] 4 is a diagram showing the configuration of structured knowledge 430 stored in the structured knowledge database 150 according to the first embodiment. The structured knowledge 430 is stored in association with the identification information of the corresponding sample and measurement analysis result 410, in addition to the identification information of the structured knowledge 430 itself.

[0020] The structured knowledge 430 is knowledge related to a specimen (sample) whose identification information is "1001" and whose name is "Polymer A." The structured knowledge 430 is configured to include configuration knowledge such as a sample 431 (specimen), a sample pan 432, measurement results 433, measurement conditions 434, analysis features 435, components 436, and physical property values ​​437. The sample pan 432 is a test container that stores a sample during thermal analysis measurement.

[0021] The configuration knowledge further includes items. For example, the configuration knowledge of the measurement conditions 434 includes items such as the start temperature and the end temperature. Furthermore, for example, the configuration knowledge of the components 436 includes items such as the name and the identification information of the material (sample) in the material database 130 (shown as "Material DB ID" in FIG. 4). In this way, the structured knowledge 430 is composed of configuration knowledge, which in turn is composed of items. For this reason, the structured knowledge 430 may be considered to be composed of items. In this embodiment, a pair of the name (label) and content (value) of an item is called an item. Sometimes, only the name or only the content of an item is called an item.

[0022] <Storage Unit: Estimation Model Database> Returning to Fig. 1 , the description of the storage unit 120 continues. Estimation models are stored in the estimation model database 121. The estimation models are machine learning models used to estimate items of structured knowledge 430 (see Fig. 4 ) that are not included in the measurement analysis results 410 (see Fig. 2 ) output by the analysis device 200, based on the measurement analysis results 410. Details of the estimation models will be described later.

[0023] <<Knowledge Creation Support Device: Control Unit>> The control unit 110 is configured to include a CPU (Central Processing Unit) and is equipped with a knowledge structuring unit 111, a learning unit 112, a missing knowledge estimation unit 113, and a knowledge editing unit 114. The control unit 110 may be configured to include a GPU (Graphics Processing Unit), an NPU (Neural (network) Processing Unit), an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), etc.

[0024] <<Control Unit: Knowledge Structuring Unit>> The knowledge structuring unit 111 generates pre-complemented structured knowledge 440 (see FIG. 5 , described later) based on the measurement analysis result 410 (see FIG. 2 ) output by the analysis device 200. FIG. 5 shows an example of the pre-complemented structured knowledge 440 according to the first embodiment. The correspondence between the items included in the structured knowledge that constitutes the pre-complemented structured knowledge 440 and the items included in the measurement analysis result 410 is determined in advance. The knowledge structuring unit 111 references this correspondence and generates the pre-complemented structured knowledge 440 based on the items included in the measurement analysis result 410.

[0025] As described above, the knowledge creation support device 100 includes a knowledge structuring unit 111 that generates pre-complemented structured knowledge 440 that includes one or more items related to the measurement based on the measurement analysis results 410 of the sample output from the analytical device 200.

[0026] <<Control Unit: Learning Unit>> Returning to FIG. 1 , the description of the control unit 110 will be continued. The learning unit 112 generates an estimation model used to complement missing constituent knowledge and items in the pre-complemented structured knowledge 440 (see FIG. 5 ) by comparing it with the structured knowledge 430 (see FIG. 4 ) stored in the structured knowledge database 150. For example, constituent knowledge such as sample pan 432 and ingredient 436 is included in the structured knowledge 430 but not in the pre-complemented structured knowledge 440. The learning unit 112 generates learning data based on the structured knowledge 430 stored in the structured knowledge database 150, and uses the learning data to train the estimation model for estimation.

[0027] For example, the learning unit 112 generates learning data using the comments included in the measurement results 433 as explanatory variables and the item of the sample pan 432 as a target variable. Next, the learning unit 112 uses this learning data to train and generate an estimation model used to estimate the material and type of the sample pan 432 based on the comments.

[0028] Furthermore, for example, the learning unit 112 generates learning data using items included in the measurement results 433, measurement conditions 434, and analysis feature quantities 435 as explanatory variables, and the name of the component 436, and the identification information (material DB ID) in the material database 130 and the physical property value 437 as objective variables. Next, the learning unit 112 uses this learning data to train and generate an estimation model used to estimate the component 436 based on the measurement results 433, measurement conditions 434, and analysis feature quantities 435.

[0029] The learning unit 112 may generate an estimation model that estimates the "ambient gas" and "ambient flow rate" included in the measurement conditions 434 based on the "comments" item included in the measurement results 433. The learning unit 112 may also generate an estimation model that estimates the component 436 item based on the analysis feature 435. The analysis feature 435 is included in the measurement analysis results 410 and is a feature calculated based on the changes in heat quantity and temperature over time, and the changes in heat quantity relative to temperature (see FIG. 3).

[0030] As explained above, the explanatory variable of the estimation model is the comment 413 included in the measurement analysis result 410 (see measurement result 433), and the objective variable of the estimation model is the item related to the atmosphere used during the thermal analysis of the sample (see "atmospheric gas" and "atmospheric flow rate" included in measurement condition 434). Furthermore, the explanatory variable of the estimation model is the comment 413 included in the measurement analysis result 410, and the objective variable of the estimation model is the item related to the sample pan 432 used during the thermal analysis of the sample. Another explanatory variable of the estimation model is the analysis feature 435 calculated based on the item related to the calorific value and temperature of the sample over time, and the objective variable of the estimation model is the item related to the components 436 of the sample.

[0031] <<Control Unit: Missing Knowledge Estimation Unit>> The missing knowledge estimation unit 113 generates the complemented structured knowledge 450 (see FIG. 6 ) based on the pre-complemented structured knowledge 440 (see FIG. 5 ) using an estimation model. FIG. 6 shows an example of the complemented structured knowledge 450 according to the first embodiment. Compared to the pre-complemented structured knowledge 440, the constituent knowledge of a sample pan 452, components 456, and physical property values ​​457 has been added. In addition, the items "atmospheric gas" and "atmospheric flow rate" that make up the constituent knowledge of measurement conditions 454 have been added. These constituent knowledge items and items have been complemented by the missing knowledge estimation unit 113 using an estimation model to estimate and add them based on the items included in the pre-complemented structured knowledge 440.

[0032] As described above, the knowledge creation support device 100 includes the missing knowledge estimation unit 113 that adds items to the pre-complemented structured knowledge 440 and generates the complemented structured knowledge 450. The missing knowledge estimation unit 113 adds items (item names and values) to the pre-complemented structured knowledge 440 using an estimation model, which is a machine learning model that uses items included in the pre-complemented structured knowledge 440 as explanatory variables and items included in the confirmed structured knowledge (see structured knowledge 430) as objective variables.

[0033] <<Control Unit: Knowledge Editing Unit>> Returning to Fig. 1, the explanation of the control unit 110 will be continued. The knowledge editing unit 114 outputs and displays a knowledge editing screen 460 (see Fig. 7 described later) on a display connected to the input / output unit 180. The knowledge editing screen 460 displays the contents of the complemented structured knowledge 450. Next, the knowledge editing unit 114 accepts an editing instruction for the complemented structured knowledge 450 from the user of the knowledge creation support device 100. When the knowledge editing unit 114 accepts an instruction to end editing, it stores the edited complemented structured knowledge 450 in the structured knowledge database 150 as confirmed structured knowledge.

[0034] 7 is a diagram showing the screen configuration of the knowledge editing screen 460 according to the first embodiment. The knowledge editing screen 460 displays items included in the supplemented structured knowledge 450 and a DSC curve graph (see FIG. 3). Items estimated and supplemented by the missing knowledge estimation unit 113 are marked with an "*". For example, for the estimated physical property value 457, "melting" and "Nylon6" are marked with an "*" as displayed in area 466 on the knowledge editing screen 460.

[0035] A user of the knowledge creation support device 100 checks the content displayed on the knowledge editing screen 460, edits incorrect information, and adds missing information. When the "Update" button at the bottom of the knowledge editing screen 460 is pressed, the knowledge editing unit 114 instructs the missing knowledge estimation unit 113 to complement the missing constituent knowledge or items based on the complemented structured knowledge 450 at the time the button was pressed. The knowledge editing unit 114 then displays the complemented structured knowledge 450 that has been completed by the missing knowledge estimation unit 113. When the "End Confirmation" button at the bottom of the knowledge editing screen 460 is pressed, the knowledge editing unit 114 stores the confirmed complemented structured knowledge 450 in the structured knowledge database 150 as confirmed structured knowledge.

[0036] As described above, the knowledge creation support device 100 includes the knowledge editing unit 114 that receives an instruction to edit the complemented structured knowledge 450 and generates confirmed structured knowledge (see structured knowledge 430).

[0037] <Knowledge Creation Support Processing> Fig. 8 is a flowchart of the knowledge creation support processing according to the first embodiment. With reference to Fig. 8, the processing from when the knowledge creation support device 100 receives the measurement analysis results from the analysis device 200 to when it acquires the structured knowledge 430 (confirmed structured knowledge) will be described.

[0038] In step S11, the knowledge structuring unit 111 receives the measurement analysis result 410 (see FIG. 2) output by the analysis device. In step S12, the knowledge structuring unit 111 generates pre-complementation structured knowledge 440 (see FIG. 5) based on the measurement analysis result 410.

[0039] In step S13, the missing knowledge estimation unit 113 uses the estimation model to add missing constituent knowledge and items in the pre-complemented structured knowledge 440 to generate complemented structured knowledge 450 (see FIG. 6). In step S14, the knowledge editing unit 114 displays a knowledge editing screen 460 (see FIG. 7) including the contents of the complemented structured knowledge 450. In step S15, the knowledge editing unit 114 accepts a correction instruction for the complemented structured knowledge 450 from the user and corrects it.

[0040] In step S16, if the "Update" button on the knowledge editing screen 460 is pressed (step S16 → YES), the knowledge editing unit 114 returns the process to step S13. More specifically, the knowledge editing unit 114 returns to step S13, and the missing knowledge estimation unit 113 complements the constituent knowledge or items that are missing in the complemented structured knowledge 450 at the time the "Update" button was pressed. If the "Update" button on the knowledge editing screen 460 is not pressed (step S16 → NO), the knowledge editing unit 114 proceeds to step S17.

[0041] In step S17, if the "End Confirmation" button on the knowledge editing screen 460 is pressed (step S17 → YES), the knowledge editing unit 114 proceeds to step S18. If the "End Confirmation" button on the knowledge editing screen 460 is not pressed (step S17 → NO), the knowledge editing unit 114 returns the process to step S15. In step S18, the knowledge editing unit 114 stores the completed structured knowledge 450 being edited in the structured knowledge database 150 as confirmed structured knowledge.

[0042] <Features of the Knowledge Creation Support Device> The knowledge creation support device 100 generates pre-complemented structured knowledge 440 (see FIG. 5) based on measurement and analysis results 410 (see FIG. 2) that are the results of measuring a sample by the analysis device 200. Next, the knowledge creation support device 100 generates complemented structured knowledge 450 (see FIG. 6) using an estimation model. The knowledge creation support device 100 then displays the complemented structured knowledge 450 on a knowledge editing screen 460 (see FIG. 7) and accepts correction instructions, and stores the complemented structured knowledge 450 in the structured knowledge database 150 as confirmed structured knowledge once the correction is complete.

[0043] The knowledge creation support device 100 allows users to efficiently create and accumulate structured knowledge 430 (see FIG. 4) based on the measurement and analysis results of samples. As the accumulation of structured knowledge 430 progresses, it is expected that the estimation accuracy of the estimation model will improve and the efficiency of creating structured knowledge will also improve.

[0044] Second Embodiment In the first embodiment, structured knowledge 430 is created based on one measurement analysis result 410 output by the analytical device 200 and stored in the structured knowledge database 150. A knowledge creation support device 100A (see FIG. 9 , described later) according to the second embodiment creates an analysis report template 480 (see FIG. 11 , described later) based on the measurement analysis result 410 of each sample for an analysis project including the analysis of one or more samples.

[0045] 9 is a functional block diagram of a knowledge creation support device 100A according to the second embodiment. Compared to the knowledge creation support device 100 of the first embodiment (see FIG. 1), an analysis project creation unit 115 and an analysis report creation unit 116 are added to the control unit 110. In addition, an analysis project database 160 is added to the storage unit 120.

[0046] The analysis project database 160 stores, in association with one another, the name of the analysis project, identification information of the sample related to the analysis project, identification information of the measurement and analysis results 410 of the sample, and identification information of the structured knowledge 430 (confirmed structured knowledge) of the sample.

[0047] The analysis project creation unit 115 creates an analysis project by accepting instructions from the user via an analysis project management screen 470 (see FIG. 10 described below). FIG. 10 is a screen configuration diagram of the analysis project management screen 470 according to the second embodiment. The analysis project management screen 470 has fields for inputting information about the analysis project, such as "project name" and "comments." In addition, a file name containing the measurement analysis results 410 is input into the "measurement analysis result file" field. When the user inputs this information and presses the "register" button, the analysis project creation unit 115 stores the information in the analysis project database 160.

[0048] A list of registered measurement analysis results 410 is displayed in area 473 of the analysis project management screen 470. When the "Edit" button is pressed, the analysis project creation unit 115 instructs the unit 115 to generate pre-complemented structured knowledge 440 (see FIG. 5) and post-complemented structured knowledge 450 (see FIG. 6) based on the corresponding measurement analysis result 410, and the knowledge editing screen 460 (see FIG. 7) is displayed. When editing of the structured knowledge is completed, the confirmed structured knowledge is stored in the structured knowledge database 150. When the "Delete" button is pressed, the analysis project creation unit 115 deletes the identification information of the measurement analysis result 410 stored in the corresponding analysis project database 160.

[0049] When a report format is specified in area 474 of the analysis project management screen 470 and the "Create" button is pressed, the analysis report creation unit 116 creates an analysis report template 480 (see FIG. 11 described later) in that format. FIG. 11 is an example of the analysis report template 480 according to the second embodiment. The analysis report template 480 includes information contained in the confirmed structured knowledge after editing by the user.

[0050] As described above, the knowledge creation support device 100A includes an analysis project creation unit 115 that associates an analysis project relating to the analysis of one or more samples, the measurement and analysis results of the samples, and the confirmed structured knowledge generated based on the measurement and analysis results. The knowledge creation support device 100A also includes an analysis report creation unit 116 that generates an analysis report template 480 for the analysis project, which includes information contained in the confirmed structured knowledge.

[0051] <Features of the Knowledge Creation Support Device> The knowledge creation support device 100A handles measurement and analysis results 410 of one or more samples as a single analysis project, and generates an analysis report template 480 that includes information contained in the confirmed structured knowledge of the samples. The user can create a final analysis report by checking and editing the analysis report template 480, thereby making analysis report creation more efficient.

[0052] <<Modification: Type of Analysis>> The above-described embodiment relates to thermal analysis, but the knowledge creation support device 100, 100A may support the creation of structured knowledge based on the output (measurement analysis result 410) of the analysis device 200 related to other analyses such as gas analysis, chromatographic analysis, elemental analysis, etc. In this case, the explanatory variables of the estimation model used by the missing knowledge estimation unit 113 are the measurement conditions and measurement results included in the measurement analysis result 410, and the objective variables are the components of the sample to be analyzed.

[0053] As described above, the explanatory variables of the estimation model are items related to the measurement conditions or measurement results included in the measurement analysis result 410, and the response variables of the estimation model are items related to the components of the sample.

[0054] Other Modifications Although several embodiments of the present invention have been described above, these embodiments are merely illustrative and do not limit the technical scope of the present invention. For example, the measurement analysis result 410 (see FIG. 2 ) may include, in addition to the measurement result, a result of analysis based on the measurement result (see peak 414), but is not limited to this. The knowledge creation support device 100, 100A may calculate the peak 414, which is an analysis feature, based on the calorific value measurement result 415.

[0055] The present invention can take on various other embodiments, and various modifications such as omissions and substitutions can be made without departing from the spirit of the present invention. These embodiments and modifications are included in the scope and spirit of the invention described in this specification, etc., and are also included in the invention described in the claims and their equivalents.

[0056] <Hardware Configuration> The knowledge creation support device 100, 100A according to the above-described embodiment is realized by a computer 900 having a configuration as shown in Fig. 12, for example. Fig. 12 is a hardware configuration diagram showing an example of the computer 900 that realizes the functions of the knowledge creation support device 100, 100A according to the above-described embodiment. The computer 900 includes a CPU 901, a ROM 902, a RAM 903, an SSD 904, an input / output interface 905 (referred to as an input / output I / F (Interface) in Fig. 12), a communication interface 906 (referred to as a communication I / F in Fig. 12), and a media interface 907 (referred to as a media I / F in Fig. 12). The computer 900 may include a hard disk drive (HDD) instead of the SSD 904, or may include an HDD in addition to the SSD 904.

[0057] The CPU 901 operates based on a program stored in the ROM 902 or the SSD 904, and performs control by the control unit 110 in Fig. 12. The ROM 902 stores a boot program executed by the CPU 901 when the computer 900 is started up, programs related to the hardware of the computer 900, and the like.

[0058] The CPU 901 controls an input device 910 such as a mouse or keyboard, and an output device 911 such as a display or printer, via an input / output interface 905. The CPU 901 acquires data from the input device 910 via the input / output interface 905, and outputs generated data to the output device 911.

[0059] The SSD 904 stores programs executed by the CPU 901 and data used by the programs. The communication interface 906 receives data from other devices (not shown) (e.g., the analysis device 200) via a communication network and outputs the data to the CPU 901, and also transmits data generated by the CPU 901 to other devices via the communication network.

[0060] The media interface 907 reads a program or data stored in the recording medium 912 and outputs it to the CPU 901 via the RAM 903. The CPU 901 loads the program from the recording medium 912 onto the RAM 903 via the media interface 907 and executes the loaded program. The recording medium 912 is an optical recording medium such as a DVD (Digital Versatile Disk), a magneto-optical recording medium such as an MO (Magneto Optical Disk), a magnetic recording medium, a conductive memory tape medium, a semiconductor memory, or the like.

[0061] For example, when the computer 900 functions as the knowledge creation support device 100, 100A according to the above-described embodiment, the CPU 901 of the computer 900 executes the program 128 (see FIG. 1 ) loaded onto the RAM 903, thereby realizing the functions of the knowledge creation support device 100, 100A. The CPU 901 reads and executes the program from the recording medium 912. Alternatively, the CPU 901 may read the program from another device via a communication network, or may install and execute the program 128 from the recording medium 912 onto the SSD 904.

[0062] 100, 100A Knowledge creation support device 111 Knowledge structuring unit 112 Learning unit 113 Missing knowledge estimation unit 114 Knowledge editing unit 115 Analysis project creation unit 116 Analysis report creation unit 121 Estimation model database 130 Material database 140 Measurement analysis result database 150 Structured knowledge database 160 Analysis project database 200 Analysis device 410 Measurement analysis result 430 Structured knowledge 440 Pre-complementation structured knowledge 450 Complementation structured knowledge 460 Knowledge editing screen 470 Analysis project management screen 480 Analysis report template

Claims

1. A knowledge creation support device comprising: a knowledge structuring unit that generates pre-complementation structured knowledge consisting of one or more items related to measurement based on the measurement and analysis results of a sample output from an analytical device; a missing knowledge estimation unit that adds items to the pre-complementation structured knowledge to generate complemented structured knowledge; and a knowledge editing unit that receives editing instructions for the complemented structured knowledge and generates confirmed structured knowledge, wherein the missing knowledge estimation unit adds items to the pre-complementation structured knowledge using an estimation model that is a machine learning model that uses items included in the pre-complementation structured knowledge as explanatory variables and items included in the confirmed structured knowledge as objective variables.

2. The knowledge creation support device according to claim 1, wherein the explanatory variables of the estimation model are items related to the measurement conditions or measurement results included in the measurement analysis results, and the objective variables of the estimation model are items related to the components of the sample.

3. The knowledge creation support device according to claim 1, wherein the analytical device is a thermal analysis device, the explanatory variables of the estimation model are comments included in the measurement analysis results, and the objective variables of the estimation model are items related to the atmosphere used during the thermal analysis of the sample.

4. The knowledge creation support device according to claim 1, wherein the analytical device is a thermal analysis device, the explanatory variables of the estimation model are comments included in the measurement analysis results, and the objective variables of the estimation model are items related to the sample pan used during the thermal analysis of the sample.

5. The knowledge creation support device according to claim 1, wherein the analytical device is a thermal analysis device, the explanatory variables of the estimation model are analytical features calculated based on items related to the calorific value and temperature of the sample over time, and the objective variables of the estimation model are items related to the components of the sample.

6. The knowledge creation support device according to claim 1, further comprising: an analysis project creation unit that associates an analysis project relating to the analysis of one or more of the samples, the measurement and analysis results of the samples, and confirmed structured knowledge created based on the measurement and analysis results; and an analysis report creation unit that creates an analysis report template for the analysis project, including information contained in the confirmed structured knowledge.

7. A knowledge creation support method in which a knowledge creation support device executes the steps of: generating pre-complementation structured knowledge consisting of one or more items related to measurement based on measurement and analysis results of a sample output from an analytical device; adding an item to the pre-complementation structured knowledge to generate complemented structured knowledge; and receiving an editing instruction for the complemented structured knowledge to generate confirmed structured knowledge, wherein in the step of generating complemented structured knowledge, an item is added to the pre-complementation structured knowledge using an estimation model that is a machine learning model that uses items included in the pre-complementation structured knowledge as explanatory variables and items included in the confirmed structured knowledge as objective variables.

Images