Analysis system, analysis method, and program product

Abstract

This invention provides an analysis system, analysis method, and program product that can assist in analyses that flexibly utilize the experimenter's insights. The analysis system disclosed herein includes a text acquisition unit, an extraction unit, and a difference evaluation unit. The text acquisition unit acquires text data corresponding to multiple samples. The extraction unit extracts sample-related information from the text data. The difference evaluation unit evaluates the differences between the multiple extracted information pieces.

Description

Technical Field

[0001] This disclosure relates to analytical systems, analytical methods, and program products. Background Technology

[0002] Patent Document 1 describes a data analysis system for analyzing measurement data of analyzed materials. The data analysis system described in Patent Document 1 receives measurement data via a communication device, processes the measurement data using a learned machine learning model, and outputs the analysis results.

[0003] Patent Document 1: Japanese Patent Application Publication No. 2021-092467

[0004] It should be noted that experimenters and developers will gain various insights during the experiments. These insights may contain information that is important for the analysis.

[0005] However, it is difficult to incorporate such insights into the analysis. Therefore, the analytical systems involved in the prior art cannot fully and flexibly utilize such insights.

[0006] In other words, existing analytical systems suffer from the problem of not being able to fully and flexibly utilize the insights gained by the experimenter during the experiment. Patent Document 1 does not disclose a technology that can solve this problem. Summary of the Invention

[0007] This disclosure was made to address such a problem, with the aim of providing an analysis system, analysis method, and program product that can assist in analysis that flexibly utilizes the experimenter's insights.

[0008] The analysis system disclosed herein includes a text acquisition unit, an extraction unit, and a difference evaluation unit. The text acquisition unit acquires text data corresponding to multiple samples. The extraction unit extracts sample-related information from the text data. The difference evaluation unit evaluates the differences between the extracted information.

[0009] Based on this configuration, the analysis system can, for example, extract differences between samples whose importance the user did not recognize. As a result, the analysis system 1 according to this embodiment can assist in analysis that flexibly utilizes the experimenter's insights.

[0010] In the analytical system disclosed herein, the extracted information may include at least one of information related to the experimental conditions of the sample and information related to the physical properties of the sample.

[0011] In the analytical system disclosed herein, the extraction unit can extract at least one of the items related to the experimental conditions of the sample and the physical properties of the sample.

[0012] The analysis system disclosed herein may also include a notification department, which evaluates user notification difference evaluation items as having differences among multiple extracted information items.

[0013] The analysis system disclosed herein may further include a group generation unit, which generates a first sample group and a second sample group based on the user's selection of samples. Furthermore, a difference evaluation unit can evaluate the differences between the extracted information corresponding to the first sample group and the extracted information corresponding to the second sample group.

[0014] The analytical system disclosed herein may further include: an analytical object data acquisition unit for acquiring analytical object data, wherein the analytical object data includes at least one of measurement data obtained by measuring a sample and numerical data related to the sample; and a display control unit for displaying multiple analytical object data. Furthermore, the group generation unit can process sample selection by allowing a user to select the displayed analytical object data.

[0015] The analytical system disclosed herein may further include an analytical object data acquisition unit and a group generation unit. The analytical object data acquisition unit acquires analytical object data, which includes at least one of measurement data obtained by measuring samples and numerical data related to the samples. The group generation unit performs clustering processing on at least one of the analytical object data and the analysis results of the analytical object data, and generates a first sample group and a second sample group based on the results of the clustering processing. Furthermore, a difference evaluation unit evaluates the differences between the extracted information corresponding to the first sample group and the extracted information corresponding to the second sample group.

[0016] The analytical methods disclosed herein involve the following steps.

[0017] Obtain the text data corresponding to each of the multiple samples.

[0018] Extract information related to the sample from text data.

[0019] Evaluate the differences between multiple extracted information.

[0020] The program products disclosed herein cause a computer to perform the following actions.

[0021] Obtain the text data corresponding to each of the multiple samples.

[0022] Extract information related to the above samples from the above text data.

[0023] The differences between the various extracted information items are evaluated.

[0024] According to this disclosure, analytical systems, analytical methods, and program products that can assist in analyses that flexibly utilize the experimenter's insights can be provided.

[0025] The above and other objectives, features and advantages of this disclosure will become more fully understood from the detailed description and accompanying drawings given below. Attached Figure Description

[0026] Figure 1 This is a block diagram illustrating the configuration of the analysis system according to the first embodiment.

[0027] Figure 2 This is a block diagram illustrating the configuration of the server according to the first embodiment.

[0028] Figure 3 This is a block diagram illustrating the configuration of the server according to the first embodiment.

[0029] Figure 4 This is a schematic diagram used to explain the configuration of the server according to the first embodiment.

[0030] Figure 5 This is a flowchart illustrating the operation of the analysis system according to the first embodiment. Detailed Implementation

[0031] <First Embodiment>

[0032] (Analysis of the system's structure)

[0033] Hereinafter, the first embodiment of this disclosure will be described in detail with reference to the accompanying drawings. First, the configuration of the analysis system according to this embodiment will be described in detail.

[0034] Figure 1 This is a block diagram illustrating the configuration of the analysis system according to the first embodiment. For example... Figure 1 As shown, the analysis system 1 according to this embodiment is connected to a server 100 and a user terminal 200 via a network N such as the Internet.

[0035] The analysis system 1 involved in this embodiment is typically provided as part of a data cloud service used in materials development and research development, and is used as a system to promote research and development using so-called materials informatics (MI) and data science.

[0036] Analysis system 1 stores various data related to experimental samples. Moreover, analysis system 1 analyzes the stored data based on instructions from the user.

[0037] The various data related to the experimental samples mentioned here include the data of the analysis objects and text data, which will be described later.

[0038] In analysis system 1, user terminal 200 sends various data related to the experimental samples to server 100, and server 100 analyzes the received data. Furthermore, server 100 sends the analysis results to user terminal 200, and user terminal 200 displays the received analysis results.

[0039] The user terminal 200 involved in this embodiment is a user-operated terminal, typically a computer device with a display device.

[0040] User terminal 200 sends various data related to the experimental samples to server 100 via network N. Furthermore, user terminal 200 receives analysis results of the various data related to the experimental samples from server 100 via network N.

[0041] In this embodiment, the server 100 receives various data related to the experimental samples from the user terminal 200 via network N and analyzes the received data. Furthermore, the server 100 sends the analysis results to the user terminal 200 via network N.

[0042] Figure 2 This is a block diagram illustrating the hardware configuration of the server according to the first embodiment.

[0043] like Figure 2 As shown, server 100 includes processor 110, memory 120, storage device 130, input / output interface 140, network interface 150, and internal bus 160.

[0044] The internal bus 160 is a data transmission path for the processor 110, memory 120, storage device 130, input / output interface 140, and network interface 150 to send and receive data with each other. However, the method of interconnecting the processor 110 and the others is not limited to bus connection.

[0045] Memory 120 is a main storage device implemented using RAM (Random Access Memory) or the like. Storage device 130 is an auxiliary storage device implemented using a hard disk, SSD (Solid State Drive), memory card, or ROM (Read Only Memory) or the like. Storage device 130 stores programs used to perform the desired functions.

[0046] Processor 110 is any processor, such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), or FPGA (Field-Programmable Gate Array). Processor 110 executes the program described later by reading it from storage device 130 into memory 120 and executing it. Figure 3 The functions of each module are shown in the diagram.

[0047] Input / output interface 140 is an interface used to connect server 100 to input / output devices. For example, input devices such as keyboards and output devices such as displays can be connected to input / output interface 140.

[0048] Network interface 150 is an interface used to connect server 100 to a network.

[0049] The program includes a set of commands (or software code) for causing a computer to perform one or more functions described in the implementation when read by the computer. The program may be stored on a non-transitory computer-readable medium or a physical storage medium. Without limitation, examples include RAM, ROM, flash memory, SSD or other memory technologies, CD-ROM, DVD (digital versatile disc), Blu-ray disc or other optical disc storage, cassette tape, magnetic tape, disk storage, or other magnetic storage devices. The program may also be transmitted on a transient computer-readable medium or communication medium. Without limitation, examples include electrical, optical, acoustic, or other forms of propagation signals.

[0050] Figure 3 This is a block diagram illustrating the functionality of the server according to the first embodiment. For example... Figure 3 As shown, the server 100 in this embodiment includes an analysis object data acquisition unit 111, a display control unit 112, a group generation unit 113, a text acquisition unit 114, an extraction unit 115, a difference evaluation unit 116, and a notification unit 117 as functional modules.

[0051] The analysis object data acquisition unit 111 acquires analysis object data. More specifically, the analysis object data acquisition unit 111 of this embodiment acquires multiple analysis object data from the user terminal 200 via network N. The analysis object data acquisition unit 111 outputs the acquired analysis object data to the display control unit 112.

[0052] Furthermore, the analysis object data acquisition unit 111 can also store the acquired analysis object data in the storage device 130 whenever analysis object data is acquired. Moreover, when performing the analysis described later, the analysis object data can be read from the storage device 130 and output to the display control unit 112.

[0053] In addition, in this case, the analysis object data acquisition unit 111 does not need to read all the analysis object data stored in the storage device 130, but can only read the specified analysis object data received by the user and output it to the display control unit 112.

[0054] In other words, the analysis object data acquisition unit 111 can create a database storing analysis object data. Moreover, the analysis object data acquisition unit 111 according to this embodiment can be configured in such a way that the user can appropriately select data as the object of analysis from the created database.

[0055] Here, the analytical object data acquired by the analytical object data acquisition unit 111 according to this embodiment is at least one of measurement data obtained by measuring the sample and numerical data related to the sample.

[0056] In other words, the analytical object data acquisition unit 111 according to this embodiment acquires analytical object data as at least one of measurement data obtained by measuring the sample and numerical data related to the sample.

[0057] The measurement data obtained by measuring the sample can be raw data output from the measuring device that measured the sample, or data that has undergone analytical processing. The measurement data obtained by measuring the sample can be any data recorded in a computer-readable format.

[0058] Measurement data obtained by measuring samples can include, for example, spectral data, waveform data, graph data, two-dimensional image data, three-dimensional image data, etc.

[0059] Examples of spectral data include those obtained using nuclear magnetic resonance spectroscopy (NMR), infrared spectroscopy (IR), ultraviolet / visible spectroscopy (UV-vis), X-ray absorption spectroscopy (XAS), Raman spectroscopy, X-ray diffraction (XRD), small-angle X-ray scattering (SAXS), and mass spectrometry (MS).

[0060] In addition, as two-dimensional image data, examples include image data captured using optical microscopes, scanning electron microscopes (SEM), transmission electron microscopes (TEM), and computed tomography (CT).

[0061] In addition, examples of three-dimensional image data include imaging data created by overlaying tomographic images taken using computed tomography (CT) scans, and model data created using CAD (Computer-Aided design).

[0062] Furthermore, waveform data can include, for example, time-series data and displacement data. Time-series data can include, for example, acoustic data and vibration data; any data whose values ​​change over time can be used. Displacement data can include, for example, the surface height and surface profile of the sample; any data whose values ​​change with changes in coordinates and other parameters can be used.

[0063] In addition, other data such as cyclic voltammetry, gas chromatography (GC) spectra, and coordinate data such as CIF (Crystallographic Information) files can be cited.

[0064] Numerical data related to a sample can be, for example, numerical data that defines the experimental conditions of the sample, or numerical data that represents the composition of the sample.

[0065] For example, when the sample is a composition, the numerical data related to the sample may be the content percentage of each component contained in the composition. Alternatively, when the sample is a product, the numerical data related to the sample may be values ​​related to reaction conditions such as reaction temperature, reaction time, the weight of the substrate used in the reaction, and the reaction scale.

[0066] Furthermore, the data to be analyzed can be data belonging to both measurement data obtained by measuring a sample and numerical data related to the sample. For example, physical property values ​​of a sample can be cited as data belonging to both measurement data obtained by measuring a sample and numerical data related to the sample.

[0067] Examples of physical properties that can be used as samples include mechanical properties such as strength, hardness, toughness, and wear resistance; physical properties such as density, electrical conductivity, magnetism, thermal conductivity, and coefficient of thermal expansion; and chemical properties such as corrosion resistance.

[0068] In addition, the data to be analyzed is not limited to data that is only associated with the sample; for example, it may also include the performance values ​​of products and modules made using the sample.

[0069] Examples of performance values ​​for products and modules made using samples include the photoelectric conversion efficiency of solar cells with photoelectric conversion layers using samples as materials, and the durability of vehicles manufactured using samples as body materials.

[0070] In other words, the analysis object data acquired by the analysis object data acquisition unit 111 may be data that directly or indirectly defines the composition of the sample as the analysis object, or data that directly or indirectly evaluates the performance of the sample as the analysis object.

[0071] In this embodiment, the data of the analysis object is data that records information related to the experimental content in a form other than an article.

[0072] The display control unit 112 displays multiple analysis object data. More specifically, the display control unit 112 according to this embodiment displays multiple analysis object data on the display device of the user terminal 200.

[0073] The display control unit 112 displays the data of the object of analysis in a manner that allows the user to select.

[0074] For example, when the data to be analyzed is spectral data, waveform data, etc., the display control unit 112 can display multiple data sets of the data being analyzed in an overlapping manner. Furthermore, the display control unit 112 can display the data to be analyzed in a way that allows the user to select the overlapping data sets through clicks or other operations.

[0075] Figure 4 This is a schematic diagram used to explain the configuration of the server according to the first embodiment. More specifically, Figure 4 This is a schematic diagram showing an example of a screen displayed on the user terminal 200 by the display control unit 112 according to this embodiment.

[0076] exist Figure 4 In this configuration, the display control unit 112 displays three spectral data points, D1, D2, and D3, overlapping each other. Here, the user confirms the screen displayed on the user terminal 200 and selects the data to be analyzed.

[0077] For example, confirmed Figure 4 The user of the screen shown focuses on the fact that spectral data D1 has different characteristics from spectral data D2 and D3.

[0078] In this case, the user selects spectral data by grouping the samples corresponding to spectral data D1 and the samples corresponding to spectral data D2 or D3 into different groups, as described later by the group generation unit 113.

[0079] Although details will be described later, the analysis system involved in this embodiment is able to search for differences between samples belonging to different groups from text data that corresponds to the samples.

[0080] Here, the differences searched by analysis system 1 are those that the user has failed to pay attention to or may not have noticed. By searching for and highlighting such differences, analysis system 1 enables the user to recognize their importance. As a result, analysis system 1 according to this embodiment can assist in analysis that flexibly utilizes the experimenter's insights.

[0081] The group generation unit 113 generates a first sample group and a second sample group based on the selection of user-accepted samples. The group generation unit 113 outputs identification information of the samples belonging to each group to the difference evaluation unit 116.

[0082] More specifically, the group generation unit 113 in this embodiment accepts sample selection by having the user select the analysis object data displayed by the display control unit 112. For example, the group generation unit 113 can group the samples displayed by the display control unit 112 that correspond to the analysis object data selected by the user into a first sample group. Furthermore, the group generation unit 113 can group analysis object data that has not been selected by the user into a second sample group.

[0083] Furthermore, although the group generation unit 113 in this embodiment groups multiple samples into two groups, the configuration of the group generation unit in this disclosure is not limited to this.

[0084] For example, the group generation unit involved in this disclosure can also divide multiple samples into more than three groups.

[0085] The group generation unit 113 in this embodiment groups samples based on user selection, but the group generation unit in this disclosure may also group samples without relying on user instructions.

[0086] In this case, the group generation unit 113 may, for example, perform clustering processing on the data of the analysis object or the analysis results thereof. Moreover, the group generation unit 113 may group the samples based on the results of the clustering processing to generate a first sample group and a second sample group.

[0087] Clustering can be performed using known analytical methods such as group averaging, Ward's method, shortest distance method, longest distance method, and k-means method. Additionally, the group generation unit 113 can perform UMAP (Uniform Manifold Approximation and Projection) as preprocessing or analytical processing on the data to be analyzed while performing clustering.

[0088] Thus, the group generation unit 113 can perform grouping based on instructions from the user, or it can perform grouping based on analysis processing that does not depend on instructions from the user. In other words, the group generation unit 113 can use any method to perform grouping, as long as it is a method that can appropriately group samples according to their characteristics.

[0089] The text acquisition unit 114 acquires text data corresponding to each of the multiple samples. More specifically, the text acquisition unit 114 acquires text data that records the experimental content of the samples. The text acquisition unit 114 outputs the acquired text data to the extraction unit 115.

[0090] The text acquisition unit 114 can store the acquired text data in the storage device 130 whenever it acquires text data. Furthermore, the text acquisition unit 114 can read text data from the storage device 130 as needed and output it to the extraction unit 115.

[0091] Furthermore, in this case, the text acquisition unit 114 does not need to read all the text data stored in the storage device 130, but can only read the text data that the user has accepted and output to the extraction unit 115.

[0092] In addition, the text acquisition unit 114 can read out the text data corresponding to the samples grouped by the group generation unit 113 and output it to the extraction unit 115.

[0093] In other words, the text acquisition unit 114 can create a database storing text data. Moreover, the text acquisition unit 114 according to this embodiment can be configured in such a way that the user can appropriately select data as the object of analysis from the created database.

[0094] Furthermore, in the database mentioned here, text data can be stored in a corresponding manner with the aforementioned analysis object data. In other words, the server 100 involved in this embodiment can store the acquired analysis object data and text data in the same database on a sample-by-sample basis.

[0095] Here, the text data involved in this embodiment refers to the data recorded in text form as documents created by experimenters who performed experiments related to the sample in order to record the experimental content.

[0096] For example, text data can be created by an experimenter inputting an article containing experimental content into the user terminal 200 using an input unit such as a keyboard. In this case, the text acquisition unit 114 acquires the text data containing the experimental content of the sample by receiving the input text data from the user terminal 200. In this case, the text data can be referred to as electronic experimental notes.

[0097] Alternatively, the document recording the experimental content can also be content written on paper by the experimenter using writing instruments. In this case, the text acquisition unit 114 can acquire image data of the paper on which the user has recorded the experimental content from the user terminal 200, and perform image processing on the acquired image data to convert the text written on the paper into text data. Furthermore, the text acquisition unit 114 can acquire the converted text data as text data recording the experimental content of the sample. In this case, the document recording the experimental content can be a so-called experimental notebook.

[0098] As described above, the text acquisition unit 114 of this embodiment can acquire text data that has been recorded as electronic data from the beginning, and can also acquire text data generated by converting text information originally recorded on paper.

[0099] In other words, the text acquisition unit 114 can acquire data other than text data, and obtain text data by converting the acquired data or extracting it from the acquired data.

[0100] For example, the text acquisition unit 114 can extract text data from electronic data containing both text data and image data. The image referred to here could be, for example, a photograph showing the appearance of a sample, a sketch of experimental equipment, or an image representing the measured spectral data of the sample. In other words, the image referred to here can be any image recorded in experimental notes or electronic experimental notes.

[0101] Although details will be described later, the analysis system according to this embodiment evaluates the differences in information extracted from text data acquired by the text acquisition unit. In other words, the analysis system according to this embodiment extracts differences between samples from documents created by the experimenter to record experimental content and evaluates the extracted differences.

[0102] Based on this configuration, the analysis system can, for example, extract differences that the user may not have noticed or paid attention to. As a result, the analysis system 1 according to this embodiment can assist in analysis that makes full use of the experimenter's insights.

[0103] The extraction unit 115 acquires text data from the text acquisition unit 114. The extraction unit 115 extracts extraction information related to the experimental content from the text data. The extraction unit 115 outputs the extracted information to the difference evaluation unit 116.

[0104] The extracted information involved in this embodiment is used to assist in the analysis of the sample. Therefore, it is preferable that the extracted information involved in this embodiment includes, for example, information related to the experimental conditions of the sample and information related to the physical properties of the sample.

[0105] The extraction information extracted by the extraction unit 115 is information obtained by extracting information related to the experimental content contained in the text data in a categorizable or evaluable manner. Therefore, the extraction information involved in this embodiment may include, for example, numerical data recording multiple variables, or identification data recording identification information about various elements related to each sample.

[0106] In the above case, the extraction unit 115 may first extract at least one of the items related to the experimental conditions of the sample and the physical properties of the sample.

[0107] Furthermore, the extraction unit 115 can extract numerical data by quantifying the degree of the extracted items for each sample. Additionally, the extraction unit 115 can also extract identification data by extracting identification information related to the extracted items from the text data.

[0108] In the above-described case, the extraction unit 115 can, for example, refer to a database containing vocabulary related to experimental conditions and vocabulary related to physical properties. Furthermore, the extraction unit 115 can extract items by retrieving vocabulary recorded in the database from text data.

[0109] Alternatively, the extraction unit 115 can extract items, for example, by using artificial intelligence (AI) trained to take text data as input data and output items.

[0110] When extracting numerical data, the extraction unit 115 can quantify the degree of an item by extracting the value representing the degree of the item recorded in the text data.

[0111] For example, if the text data states "the reaction solution was heated at a reaction temperature of 60 degrees Celsius for 1 hour," the extraction unit 115 first extracts "reaction temperature" and "reaction time" as items related to the experimental conditions. Furthermore, for the "reaction temperature" item, the degree of the item is quantified by extracting the value "60 degrees Celsius." Similarly, for the "reaction time" item, the degree of the item is quantified by extracting the value "1 hour."

[0112] In addition, the extraction unit 115 can extract items related to the physical properties of the sample, assign a first predetermined value to samples that exhibit the physical property, and assign a second predetermined value to samples that do not exhibit the physical property, thereby quantifying the degree of the physical property.

[0113] For example, if text data is obtained stating "a product with foaming properties is obtained" and text data is obtained stating "a product is obtained" without mentioning foaming properties, the extraction unit 115 first extracts "foaming properties" as an item related to the physical properties of the sample.

[0114] Furthermore, the extraction unit 115 assigns a first predetermined value (e.g., "1") to the "foaming property" item of the sample corresponding to the text data that states "a foaming product is obtained". Additionally, the extraction unit 115 assigns a second predetermined value (e.g., "0") to the "foaming property" item of the sample corresponding to the text data that states "a product is obtained" but does not mention foaming property.

[0115] However, when the extraction unit 115 extracts items related to the physical properties of the sample, the extraction unit 115 may extract only the identification information related to whether or not the physical property is displayed without quantifying the degree of the physical property.

[0116] In addition, the extraction unit 115 can also extract words that indicate the degree of an item from the text data and quantify the degree of the item based on the extracted words.

[0117] For example, when there is text data stating "a slightly foamed product is obtained", text data stating "a foamed product is obtained", and text data stating "a vigorously foamed product is obtained", the extraction unit 115 first extracts "foaming property" as an item related to the physical properties of the sample. In addition, the extraction unit 115 extracts "slightly" and "vigorously" as words indicating the degree of the item.

[0118] Furthermore, the extraction unit 115 assigns a first predetermined value (e.g., "1") to the "foaming property" item of the sample corresponding to the text data described as "obtaining a slightly foamed product". Additionally, the extraction unit 115 assigns a second predetermined value (e.g., "2") larger than the first predetermined value to the "foaming property" item of the sample corresponding to the text data described as "obtaining a foamed product". Furthermore, the extraction unit 115 assigns a third predetermined value (e.g., "3") larger than the second predetermined value to the "foaming property" item of the sample corresponding to the text data described as "obtaining a violently foamed product".

[0119] In addition, the extraction unit 115 can extract the identification information related to each extracted item as extraction information.

[0120] In this case, the extraction unit 115 may, for example, extract the "production batch number of the compound" as an item and extract the corresponding "production batch number" as identification information.

[0121] In addition, the extraction unit 115 can extract, for example, the "date and time of sample creation" as an item, and extract the corresponding "date and time" as identification information.

[0122] Additionally, the extraction unit 115 can extract, for example, the "sample creation method" as an item. In this case, the extraction unit 115 can pre-assign an identifier to the type of "sample creation method". Moreover, the corresponding sample creation method can be identified based on information recorded in text data, and the identifier corresponding to the identified creation method is recorded as extraction information.

[0123] Thus, the extraction unit 115 of this embodiment extracts information related to the experimental content from text data, but its extraction method is not particularly limited. In other words, as long as a method is capable of extracting information from text data that can classify or evaluate samples, the extraction unit 115 of this embodiment can use any method to extract the information.

[0124] In this embodiment, the extraction unit 115 may, for example, use artificial intelligence (AI) to extract information from text data.

[0125] In this case, the extraction unit 115 extracts information by using artificial intelligence (AI) trained in a manner that takes text data as input data and outputs extracted data in a form that can classify or evaluate samples in the difference evaluation unit 116 described later.

[0126] The difference evaluation unit 116 obtains the identification information of samples belonging to the first and second groups from the group generation unit 113. Furthermore, the difference evaluation unit 116 obtains the extraction information corresponding to the samples belonging to the first and second groups from the extraction unit 115.

[0127] The difference evaluation unit 116 evaluates the differences between multiple extracted information pieces. More specifically, the difference evaluation unit 116 according to this embodiment evaluates the differences between the extracted information corresponding to the first sample group and the extracted information corresponding to the second sample group. The difference evaluation unit 116 outputs the evaluation result to the notification unit 117.

[0128] For example, the difference evaluation unit 116 according to this embodiment can evaluate the difference in extracted information by comparing the first sample group and the second sample group for each item extracted by the extraction unit 115. Moreover, the difference evaluation unit 116 can output identification information of items evaluated as having differences between the extracted information to the notification unit 117.

[0129] For example, when comparing the first sample group and the second sample group for items with recorded numerical data, the difference evaluation unit 116 can compare the distribution range of the values ​​or the average of the values. In other words, the difference evaluation unit 116 can use any method to compare the first sample group and the second sample group, as long as it is a method that can compare and evaluate the values ​​or the distribution of values ​​in each sample group.

[0130] Furthermore, the difference evaluation unit 116 can, for example, determine that there is a difference for the item if the difference between the compared numerical data is above a predetermined threshold, and output the identification information of the item to the notification unit 117.

[0131] In addition, for example, when comparing the first sample group and the second sample group with respect to items with recorded identification information, the difference evaluation unit 116 can compare the first sample group and the second sample group based on the identification information assigned to each group.

[0132] Furthermore, in this case, the difference evaluation unit 116 can also compare the first sample group and the second sample group based on the proportion of the assigned identification information in each group. Moreover, the difference evaluation unit 116 can, for example, determine that a difference exists for this item if the difference in the proportions of the compared identification information is above a predetermined threshold.

[0133] The notification unit 117 obtains the identification information of items evaluated as having differences from the difference evaluation unit 116. The notification unit 117 then notifies the user of items evaluated as having differences among multiple extracted information.

[0134] For example, the notification unit 117 can evaluate the user notification as having differences by having the display device on the user terminal 200 display messages such as "the date and time of the experiment performed in the first sample group and the second sample group are different", "the batch number of the material used in the selected sample and the unselected sample is different", and "the distribution of the reaction temperature is different in the first sample group and the second sample group".

[0135] In other words, the notification unit 117 can notify users by embedding items that are rated as having differences into the message.

[0136] Additionally, for example, notification unit 117 can provide users with a list of items that are rated as having discrepancies.

[0137] Furthermore, for example, the notification unit 117 can provide the user with a list of all items displayed by the extraction unit 115. Moreover, items rated as having discrepancies can be highlighted. This highlighting may include, for example, displaying items rated as having discrepancies in a different color than other items, or displaying a prescribed marker near items rated as having discrepancies.

[0138] With this configuration, the notification unit 117 of this embodiment can also evaluate user notifications as items with discrepancies.

[0139] (Analyze the system's actions)

[0140] Next, the operation of the analysis system, namely the analysis method involved in the first embodiment, will be described in detail. Figure 5 This is a flowchart illustrating the operation of the analysis system according to the first embodiment. Reference will be made appropriately in the following description. Figures 1-4 .

[0141] exist Figure 5 In the processing steps, the processor 110 of the server 100 reads the program stored in the storage device 130 into the memory 120 and executes it to function as the data acquisition unit 111, display control unit 112, group generation unit 113, text acquisition unit 114, extraction unit 115, difference evaluation unit 116, and notification unit 117.

[0142] In the analysis method according to this embodiment, firstly, the processor 110 acquires the analysis target data (step ST1). More specifically, in step ST1, the processor 110 acquires the analysis target data as at least one of measurement data obtained by measuring a sample and numerical data related to the sample. In other words, in step ST1, the processor 110 functions as an analysis target data acquisition unit 111.

[0143] For example, in step ST1, the processor 110 receives a selection of analysis object data from the user terminal 200 to perform the analysis. Furthermore, the processor 110 retrieves the selected analysis object data from the storage device 130.

[0144] Next, the processor 110 displays the analysis object data (step ST2). More specifically, in step ST2, the processor 110 displays the analysis object data in a manner selectable by the user. In other words, in step ST2, the processor 110 functions as a display control unit 112.

[0145] Next, the processor 110 generates the first sample group and the second sample group by having the user select the displayed analysis object data (step ST3). In other words, in step ST3, the processor 110 functions as the group generation unit 113.

[0146] Next, processor 110 acquires text data (step ST4). More specifically, in step ST4, processor 110 acquires text data related to the sample. In other words, in step ST4, processor 110 functions as a text acquisition unit 114.

[0147] Next, processor 110 extracts information from the text data (step ST5). More specifically, in step ST5, processor 110 extracts information related to the experimental content from the text data. In other words, in step ST5, processor 110 functions as extraction unit 115.

[0148] Next, the processor 110 evaluates the difference between the extracted information corresponding to the first group and the extracted information corresponding to the second group (step ST6). In other words, in step ST6, the processor 110 functions as a difference evaluation unit 116.

[0149] Finally, the processor 110 evaluates the user notifications as having discrepancies (step ST7), and the analysis system according to this embodiment concludes a series of actions. In other words, in step ST7, the processor 110 functions as the notification unit 117.

[0150] As explained above, the analysis system according to this embodiment evaluates the differences in information extracted from the text data obtained by the text acquisition unit.

[0151] Based on this configuration, the analysis system can, for example, extract differences between samples whose importance the user was unaware of. As a result, the analysis system 1 according to this embodiment can assist in analyses that flexibly utilize the experimenter's insights.

[0152] <Other Implementation Methods>

[0153] The analysis system described in the first and second embodiments is implemented as a server 100, but the configuration of the analysis system disclosed herein is not limited thereto. For example, the analysis system disclosed herein may also be implemented using two or more computer devices. Furthermore, for example, the analysis system disclosed herein may also be implemented as a user terminal 200, with some or all of its components configured as such.

[0154] The present invention has been described above according to the above embodiments, but the present invention is not limited to the above embodiments. Of course, it includes various modifications, alterations and combinations that can be made by those skilled in the art within the scope of the invention.

[0155] Based on the disclosure described herein, it will be apparent that embodiments of this disclosure can be modified in various ways. Such modifications should not be considered as departing from the spirit and scope of this disclosure, and as will be apparent to those skilled in the art, all such modifications are intended to be included within the scope of the technical solutions.

Claims

1. An analysis system, characterized in that, have: The text acquisition unit acquires text data corresponding to multiple samples respectively; The extraction unit extracts extraction information related to the sample from the text data; and The difference evaluation department evaluates the differences between the extracted information.

2. The analysis system according to claim 1, characterized in that, The extracted information includes at least one of information related to the experimental conditions of the sample and information related to the physical properties of the sample.

3. The analysis system according to claim 2, characterized in that, The extraction unit extracts at least one of the items related to the experimental conditions of the sample and the physical properties of the sample.

4. The analysis system according to claim 3, characterized in that, The analysis system also includes a notification unit, which notifies the user of items evaluated by the difference evaluation unit as having differences among multiple extracted information items.

5. The analysis system according to any one of claims 1 to 4, characterized in that, The analysis system also includes a group generation unit, which generates a first sample group and a second sample group based on the user's selection of the samples. The difference evaluation unit evaluates the difference between the extracted information corresponding to the first sample group and the extracted information corresponding to the second sample group.

6. The analysis system according to claim 5, characterized in that, It also has: The analysis object data acquisition unit acquires analysis object data, which includes at least one of measurement data obtained by measuring a sample and numerical data related to the sample; and The display control unit enables the display of multiple analyzed objects. The group generation unit accepts the sample selection by having the user select the displayed analysis object data.

7. The analysis system according to any one of claims 1 to 4, characterized in that, It also has: The analysis object data acquisition unit acquires analysis object data, which includes at least one of measurement data obtained by measuring a sample and numerical data related to the sample; and The group generation unit performs clustering processing on at least one of the analysis object data and the analysis results of the analysis object data, and generates a first sample group and a second sample group based on the results of the clustering processing. The difference evaluation unit evaluates the difference between the extracted information corresponding to the first sample group and the extracted information corresponding to the second sample group.

8. An analytical method, characterized in that, Obtain the text data corresponding to multiple samples respectively. Extract information related to the sample from the text data. The differences between the extracted information are evaluated.

9. A program product, characterized in that, To make the computer perform the following actions: Obtain the text data corresponding to multiple samples respectively. Extract information related to the sample from the text data. The differences between the extracted information are evaluated.

Images