WO2018047251A1 - Système d'analyse de données, terminal d'analyse de données et procédé d'analyse de données - Google Patents

Système d'analyse de données, terminal d'analyse de données et procédé d'analyse de données Download PDF

Info

Publication number
WO2018047251A1
WO2018047251A1 PCT/JP2016/076268 JP2016076268W WO2018047251A1 WO 2018047251 A1 WO2018047251 A1 WO 2018047251A1 JP 2016076268 W JP2016076268 W JP 2016076268W WO 2018047251 A1 WO2018047251 A1 WO 2018047251A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
information
recipe
sample
measurement
Prior art date
Application number
PCT/JP2016/076268
Other languages
English (en)
Japanese (ja)
Inventor
恵木　正史
卓成 桂
大輔 福井
美奈 吉村
Original Assignee
株式会社日立ハイテクノロジーズ
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社日立ハイテクノロジーズ filed Critical 株式会社日立ハイテクノロジーズ
Priority to JP2018537917A priority Critical patent/JPWO2018047251A1/ja
Priority to PCT/JP2016/076268 priority patent/WO2018047251A1/fr
Publication of WO2018047251A1 publication Critical patent/WO2018047251A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Definitions

  • the present invention generally relates to data analysis.
  • Patent Document 1 discloses that a co-occurrence relationship between attributes is tested while searching for combinations of attributes (columns) of data, and a statistically significant co-occurrence relationship is automatically extracted.
  • the tool of Non-Patent Document 1 provides a number of statistical techniques for significant difference testing. The user explicitly sets a hypothesis in the tool and tests the statistical significance of the hypothesis.
  • a user creates and tests hypotheses one by one on the measurement data based on knowledge and experience.
  • the number of hypothesis candidates is enormous, many trials and errors are necessary until a reasonable conclusion is obtained.
  • it is necessary to select an appropriate statistical method from a large number of statistical methods according to the hypothesis and sample data.
  • the user's knowledge and experience can cause a large variation in conclusions.
  • an object of the present invention is to assist the user in efficiently creating a hypothesis and selecting an appropriate statistical method.
  • a data analysis system includes a processor and a memory.
  • the memory stores measurement information including measurement data for each item related to the specimen.
  • Processor Recipe generation processing for generating a plurality of recipe information with different combinations of items based on the measurement information items acquired from the memory;
  • Sample data generation processing that applies each recipe information to measurement data and generates sample data corresponding to each recipe information;
  • a test method determining process for determining a statistical test method to be applied to the sample data based on the configuration of the sample data;
  • a test score calculation process that applies a statistical test method determined by the test method determination process to sample data and calculates a test score indicating statistical significance regarding the recipe information used to generate the sample data;
  • An analysis result display process for displaying the test score and recipe information related to the test score in association with each other is executed.
  • the example of the hardware constitutions of a data analysis system is shown.
  • the example of a function structure of a data analysis system is shown.
  • part and channel ID is shown.
  • An example of measurement information is shown.
  • An example of sample information is shown.
  • An example of measurement supplement information is shown.
  • An example of GUI (Graphical User Interface) for setting recipe generation conditions is shown. It is a flowchart which shows the example of a process of a recipe production
  • An example of recipe information is shown.
  • An example of sample data is shown.
  • An example of an analysis result table is shown.
  • An example of measurement by an acceleration sensor is shown.
  • An example of multi-modal measurement is shown.
  • the example of the sample information which concerns on multimodal is shown.
  • the expressions “identification information”, “identifier”, “name”, “name”, “ID” may be used, but these can be replaced with each other. is there.
  • a case where the subject is a sample from the population will be described. Therefore, “sample” and “subject” can be replaced with each other.
  • sample identifier” and “subject ID” can be replaced with each other.
  • the specimen may not be a person such as a device or a store.
  • the process may be described using “program” as a subject.
  • the program is executed by a processor (for example, a CPU (Central Processing Unit)), so that a predetermined process is appropriately performed. Since the processing is performed using at least one of a storage resource (for example, a memory) and a communication interface device, the subject of the processing may be a processor and an apparatus having the processor. Part or all of the processing performed by the processor may be performed by a hardware circuit.
  • the computer program may be installed from a program source.
  • the program source may be a program distribution server or a storage medium (for example, a portable storage medium).
  • FIG. 1 shows an example of the hardware configuration of the data analysis system 10.
  • the data analysis system 10 includes a CPU 12, a memory 14, a storage 22, an input I / F (Interface) 16, an output I / F 18, and a communication I / F 20. These components 12 to 22 are connected to a bus 24 capable of bidirectional communication.
  • the memory 14 stores data and programs. Examples of the memory 14 are DRAM (Dynamic Random Access Memory), MRAM (Magnetic Resistant Random Access Memory), FeRAM (Ferroelectric Random Access Memory), and phase change memory.
  • DRAM Dynamic Random Access Memory
  • MRAM Magnetic Resistant Random Access Memory
  • FeRAM Feroelectric Random Access Memory
  • phase change memory phase change memory
  • the storage 22 may store data and programs. Examples of the storage 22 are HDD (Hard Disk Drive) and SSD (Solid State Drive).
  • the CPU 12 implements various functions of the data analysis system 10 by reading and processing programs and data from the memory 14.
  • the input I / F 16 is an I / F for the user to input information to the data analysis system 10.
  • Examples of the input I / F 16 are a keyboard, a mouse, a microphone, and the like.
  • the output I / F 18 is an I / F for the data analysis system 10 to output information to the user.
  • Examples of the output I / F 18 are a display and a speaker.
  • the data analysis system 10 When the data analysis system 10 receives a request for data analysis from the data analysis terminal 30 via the communication network 4 such as the Internet, the data analysis system 10 performs data analysis based on the request and responds to the data analysis terminal 30 with the analysis result. May be. Similar to the data analysis system 10, the data analysis terminal may include a CPU 31, a memory 34, an input I / F, an output I / F, and a communication I / F 40 connected to the bus 42.
  • FIG. 2 shows an example of the functional configuration of the data analysis system 10.
  • the data analysis system 10 may hold measurement information 100, sample information 200, and measurement supplement information 300 as data. Some or all of these data may be stored in the memory 14 and / or the storage 22.
  • the measurement information 100 will be described later (see FIG. 4).
  • the sample information 200 will be described later (see FIG. 5). Details of the measurement supplement information 300 will be described later (see FIG. 6).
  • the data analysis system 10 includes, as functions, a recipe generation condition acquisition unit 44, a recipe generation unit 42, a sample data generation unit 46, a test method determination unit 48, a test score calculation unit 50, and an analysis result display unit 52. Good. Some or all of these functions may be realized by the CPU 12 executing a program stored in the memory 14. Alternatively, some or all of these functions may be realized by a predetermined logic circuit.
  • the recipe generation condition reception unit 44 receives a condition 400 (referred to as “recipe generation condition”) 400 related to generation of recipe information from the user.
  • the recipe generation condition receiving unit 44 may generate and display a GUI 401 (see FIG. 7) for receiving the recipe generation condition 400 from the user.
  • the recipe generation condition reception unit 44 may acquire the recipe generation condition 400 from the storage 22 or via the communication network 4.
  • the recipe generation unit 42 generates a plurality of recipe information 500.
  • the recipe generation unit 42 may generate recipe information based on the measurement information 100, the sample information 200, and the measurement supplement information 300. In that case, the recipe production
  • the sample data generation unit 46 applies the plurality of recipe information 500 generated by the recipe generation unit 42 to the measurement information 100, and generates a plurality of sample data 600.
  • the sample data generation unit 46 may include a cleansing unit 54 and a feature amount calculation unit 56 as functions.
  • the cleansing unit 54 cleanses the measurement data 110 (see FIG. 4) included in the measurement information 100.
  • the feature amount calculation unit 56 calculates a feature amount using the measurement data 110 cleansed by the cleansing unit 54. Details of the processing of the sample data generation unit 46 will be described later (see FIG. 8).
  • the test method determination unit 48 determines an appropriate statistical test method to be applied to the sample data 600 based on the configuration of the sample data 600 for each of the plurality of sample data 600 generated by the sample data generation unit 46. .
  • the test method determination unit 48 may select an appropriate statistical test method from a plurality of preset statistical test methods.
  • the test score calculation unit 50 applies the statistical test method determined by the test method determination unit 48 to each of the plurality of sample data 600, and relates to the recipe information 500 used to generate the sample data 600. A test score is calculated. The test score is a value indicating the statistical significance of the recipe information 500.
  • the analysis result display unit 52 generates and displays an analysis result in which the test score calculated by the test score calculation unit 50 is associated with the recipe information related to the test score. A display example of the analysis result will be described later (see FIG. 11).
  • FIG. 3 shows the relationship between the measurement site and the channel ID.
  • the cerebral blood flow in each part of the subject's brain is the measurement target.
  • a brain activity measuring device 50 is attached to the subject's head, and the cerebral blood flow in each part of the subject's brain is measured in association with the channel ID.
  • FIG. 4 shows an example of the measurement information 100.
  • the measurement information 100 manages measurement data corresponding to data items.
  • the measurement information 100 in the present embodiment has values measured from each part of the subject's brain when the subject performs the task.
  • the measurement information 100 in FIG. 4 includes a subject ID 102, a task ID 104, a repetition ID 106, a channel ID 108, and measurement data 110 as data items.
  • Subject ID 102 is information for identifying the subject.
  • Task ID 104 is information for identifying a task.
  • the repetition ID 106 is information indicating the number of repetitions of the task.
  • the channel ID 108 is information for identifying each part of the brain that is the measurement target.
  • Measurement data 110 is a value (time-series data) measured at a predetermined interval at a site indicated by the channel ID 108 when the subject ID 102 performs the task ID 104 when the task ID 104 is repeatedly executed.
  • each subject performs two or more contrasted tasks in the same way. This is to perform a significant difference test for the task to be compared.
  • the tasks to be compared include, for example, a case where a drink A and a drink B are drunk, a case where a car A is driven, and a case where a car B is driven.
  • FIG. 5 shows an example of the sample information 200.
  • Specimen information 200 manages information related to specimen attributes. *
  • the sample information 200 shown in FIG. 5 includes “question 01 to question 10” answers regarding the subject ID 202, age 204, gender 206, and preference 208 as data items. Therefore, by referring to one record of the sample information 200, the age 204, sex 206, and preference 208 of the subject with the subject ID 202 can be recognized.
  • FIG. 6 shows an example of the measurement supplement information 300.
  • the measurement supplement information 300 manages information related to the attributes of items other than the sample identifier included in the measurement information.
  • FIG. 6 shows a plurality of attributes that the channel ID 108 of FIG. 4 has.
  • the measurement supplement information 300 includes information regarding a method for grouping a plurality of channel IDs.
  • the group method ⁇ 304 forms three groups from a plurality of channel IDs.
  • the channel IDs 01 to 06 may be grouped into the frontal left group, the channel IDs 07 to 16 into the frontal center group, and the channel IDs 17 through 22 into the frontal right group.
  • the group method ⁇ 306 forms five groups from a plurality of channel IDs.
  • the channel IDs 01 to 04 are in the left BA46 area group
  • the channel IDs 05 to 09 are in the left BA10 field group
  • the channel IDs 10 to 13 are in the frontal center group
  • the channel IDs 14 to 18 are in the right area.
  • the channel IDs 19 to 22 may be grouped into the group near the right BA46 field in the group near the BA10 field.
  • the group method ⁇ 308 forms seven groups from a plurality of channel IDs.
  • channel IDs 01 to 03 are in the group near the lower left frontal gyrus
  • channel IDs 04 to 06 are in the group near the left middle frontal gyrus
  • channel IDs 07 to 09 are in the group near the upper left frontal gyrus
  • channel IDs 10 to 13 May be grouped into the upper frontal gyrus group
  • channel IDs 14-16 may be grouped in the upper right frontal gyrus group
  • channel IDs 17-19 may be grouped in the right middle occipital gyrus group
  • channel IDs 20-22 may be grouped in the lower right frontal gyrus group.
  • FIG. 7 shows an example of a GUI for setting recipe generation conditions.
  • the GUI 401 is generated and displayed by the recipe generation condition receiving unit 44.
  • the user may be able to set via the GUI 401 whether the measurement information item 402 is a candidate for a factor and whether to be a candidate for a stratified item.
  • the measurement information item 402 may include a data item related to the measurement information.
  • the measurement information item 402 may include a task ID 104 of the measurement information 100, a repetition ID 106, a channel ID 108, and group methods ⁇ 304, ⁇ 306, and ⁇ 308 of the measurement supplementary information 300.
  • the user may be able to set through the GUI 401 whether or not the sample information item 404 is a candidate for a factor and whether to be a candidate for a stratified item.
  • the sample information item 404 may include data items related to the sample information 200.
  • the sample information item 404 may include the age 204, the sex 206, and the preference 208 of the sample information 200.
  • the user may be able to set the cleansing method and its strength 406 via the GUI 401.
  • the user may be able to set whether to include each feature quantity 408 as a candidate via the GUI 401.
  • the recipe generation condition reception unit 44 may calculate and display the total number 410 of recipe information generated based on the content set in the GUI 401. Typically, the greater the total number 410, the longer the processing time required for analysis. Thereby, the user can change the setting contents so that the appropriate total number 410 is obtained.
  • FIG. 8 a is a flowchart showing an example of processing of the recipe generation unit 42.
  • the recipe generation unit 42 executes the following processing.
  • the recipe generation unit 42 selects at least one item from the measurement information item 402 and the sample information item 404 as a factor.
  • the items that can be selected as factors may be those set as factor candidates in the GUI 401 in FIG.
  • the recipe generation unit 42 selects whether or not to divide by layer, and executes S14 when stratified and executes S30 when not stratified.
  • the recipe generating unit 42 selects at least one item as a stratified item from the measurement information item 402 and the sample information item 404, excluding the factor selected in (S10).
  • the items that can be selected as the stratified item may be those set as candidates for the stratified item in the GUI 401 of FIG.
  • the recipe generation unit 42 selects at least one value as a layer from the values of the stratified items determined in S14.
  • the recipe generation unit 42 selects a cleansing method and strength.
  • the cleansing method and intensity that can be selected here may be those set as the cleansing method and intensity 406 of the GUI 401 in FIG.
  • the recipe generation unit 42 selects a feature amount.
  • the feature quantities that can be selected here may be those selected as candidates by the feature quantity 408 of the GUI 401 in FIG.
  • the recipe generating unit 42 selects the factor selected in S10, the stratified item selected in S14 if selected by layer in S12, the layer selected in S16, and the selected in S20.
  • Recipe information 500 is generated based on the cleansing method and the feature amount calculation method selected in S22. The recipe information 500 will be described with reference to FIG.
  • the recipe generation unit 42 selects the factor in S10, selects whether or not to classify in S12, selects the stratified item in S14, selects the layer in S16, and selects the cleansing method and strength in S20.
  • a plurality of recipe information may be generated by changing the combination of selections regarding the selection of the feature amount calculation method in S22.
  • the number of candidates for the measurement information item 402 and the sample information item 404 in the GUI 401 of FIG. 7 is increased, the number of recipe information to be generated (the total number of hypotheses) is also increased.
  • FIG. 9 shows an example of recipe information 500.
  • the recipe information 500 includes information indicating how the sample data 600 is created from the measurement data 110. That is, if the contents of the recipe information 500 are different, different specimen data 600 is created.
  • the recipe information 500 includes the factor 502 selected in S10 above, the layer 504 selected in S14 and S16 if selected by layer in S12, the cleansing method 506 selected in S20, and the selection in S22. And the calculated feature value calculation method 508.
  • FIG. 8 b is a flowchart showing an example of processing of the sample data generation unit 46.
  • the sample data generation unit 46 may execute the following process for each of the plurality of recipe information 500 generated by the recipe generation unit 42.
  • the sample data generation unit 46 extracts all the measurement data 110 when the layer 504 is not designated in S12, or the recipe information 500
  • the measurement data 110 matching the conditions of the layer 504 is extracted, and based on the combination of the specimen identifier (subject ID) and the value of the item specified by the factor 502 of the recipe information 500 Then, the extracted measurement data 110 is grouped. Thereby, a plurality of groups having different sample identifiers and factor values are formed.
  • the sample data generation unit 46 cleanses the measurement data 110 belonging to each of the plurality of formed groups by the cleansing method set in the cleansing method 506 of the recipe information 500. Then, the specimen data generation unit 46 calculates the feature amount for each of the plurality of cleansed groups by the method specified by the feature amount calculation method 508 of the recipe information 500.
  • the sample data generation unit 46 generates the sample data 600 using the calculated feature amount of the group.
  • the sample data 600 will be described with reference to FIG.
  • FIG. 10 shows an example of the sample data 600.
  • the sample data 600 is data generated based on the recipe information 500 from the measurement data 110.
  • the sample data 600 in the example of FIG. 10 is generated based on the recipe information 500 in which the task ID 104 and the gender 206 are set in the factor 502 of the recipe information 500.
  • the number of factors is “2” (that is, “task ID” and “gender”), and the number of samples of each factor (the number of male subjects and the number of female subjects of task ID) is available. Absent.
  • the test method determination unit 48 may determine that “two-factor ANOVA (analysis of variance) with different sample numbers” is suitable as the statistical test method.
  • test methods include, for example, “unmatched t-test”, “corresponding t-test”, “one-factor ANOVA with the same number of samples”, “one-factor ANOVA with a different number of samples”, “2 with the same number of samples” "Factor ANOVA”, “two-factor ANOVA with different number of samples”, “two-factor ANOVA with both corresponding factors”, “two-factor ANOVA with one factor but not one factor with the same number of samples”, “There is no correspondence to one factor, two factors ANOVA that corresponds to one factor and the number of samples is different”, “3 factors ANOVA that has no correspondence to three factors and the number of samples is equal”, “No correspondence to three factors and a sample There are three factors ANOVA with different numbers.
  • the test method determination unit 48 may determine which statistical test method is suitable based on the structure of the sample data 600.
  • FIG. 11 shows an example of the analysis result table 700.
  • the result of the statistical test by the statistical test method for the sample data 600 generated by one recipe information 500 may be stored as one record. Each record represents a plurality of different recipe information 500.
  • the analysis result table 700 may include a factor 702, a layer 704, a feature amount 706, a cleansing method 708, and a test score 710 as data items.
  • the factor 702 stores a factor (factor 502 of the recipe information 500) used for calculating the test score 710.
  • the layer 704 stores a layer (layer 504 of the recipe information 500) used for calculating the test score 710.
  • the cleansing method 706 stores the cleansing method (cleaning method 506 of the recipe information 500) used for calculating the test score 710.
  • the feature value calculation method 708 stores the feature value calculation method (the feature value calculation method 508 of the recipe information 500) used for calculating the test score 710.
  • the test score 710 is determined by the test method determination unit 48 with respect to the sample data 600 generated from the recipe information 500 including the values stored in the factor 702, the layer 704, the cleansing method 706, and the feature amount calculation method 708.
  • the result (p value) of the significant difference test calculated based on the statistical test method is stored.
  • the test score 710 may be calculated by the test score calculation unit 50. The smaller the test score 710, the higher the possibility that the difference in the value of the factor 702 will cause a significant difference in the value of the feature quantity 708 when the layer 704 is focused (that is, it is less likely to occur by chance). Show.
  • the difference in task ID (702) is cleansing with the noise type A intensity 3 (706 ) Is likely to cause a significant difference in the average value (708) of cerebral blood flow (ie, the chance of accidental occurrence is “0.0002”) (710).
  • each record of the analysis result table 700 may be interpreted as corresponding to “hypothesis and its test result” in the significant difference test.
  • the analysis result display unit 52 may display the contents of the analysis result table 700 on the display.
  • the analysis result display unit 52 may sort and display the records of the analysis result table 700 in ascending order of the test score 710. By this sorting, the user can easily know a hypothesis having high statistical significance.
  • the user can efficiently create a hypothesis by operating the GUI 401 as shown in FIG. Further, since the test method determination unit 48 determines an appropriate statistical test method based on the structure of the sample data 600, even a user who is not familiar with the statistical method can perform analysis. Further, since the analysis result table 700 can be sorted in ascending order of the test score 710, it is possible to easily know a hypothesis having high statistical significance from a large number of hypotheses.
  • FIG. 12 shows an example of measurement by an acceleration sensor.
  • Measured data is not limited to cerebral blood flow in each part of the brain as in Example 1.
  • the measurement data may be a value measured by an acceleration sensor attached to each part of the body (such as a hand or a foot) as shown in FIG.
  • one acceleration sensor can measure three values of X axis, Y axis, and Z axis.
  • the number of channel IDs when one acceleration sensor 60a is attached to the body is three (measurement values of the X axis, the Y axis, and the Z axis), and the four acceleration sensors 60b, 60c, 60d, and 60e are applied to the body.
  • the number of channel IDs when the is attached is 12.
  • FIG. 13 shows an example of multimodal measurement.
  • ⁇ Multi-modal measurement refers to the simultaneous measurement of data using different types of sensors.
  • the brain activity measuring device 50 and the acceleration sensor 60f are attached to the head of one subject. Then, cerebral blood flow data and acceleration data are measured simultaneously. Thereby, the relationship between the change in the acceleration of the subject's head and the change in the cerebral blood flow in each part of the subject's brain can be analyzed.
  • FIG. 14 shows an example of sample information 220 related to multimodal.
  • a certain kind of data may be converted into a feature amount and added as an item of the sample information 200.
  • the data item of the feature amount 230 of the acceleration sensor is added to the sample information 200 of FIG. 6, and the acceleration data measured from the subject with the subject ID is converted into the feature amount in the feature amount item.
  • a value may be stored.
  • the measurement value by the acceleration sensor is converted into three feature values of small (S), normal (M), and large (L).
  • the measurement information 100 and the sample information 200 are stored in the memory 34 of the data analysis terminal 30, and the data analysis terminal 30 (the CPU 32 thereof) stores the measurement information 100 and the sample information 200 stored in the memory 34 as data. It may be transmitted to the analysis system 10 to request data analysis. Upon receiving the data analysis request, the data analysis system 10 may transmit data for generating the GUI 401 in FIG. 7 to the data analysis terminal 30. The data analysis terminal 30 may generate and display the GUI 401 in response to data for generating the GUI 401 from the data analysis system 10. The user may input the recipe generation condition 400 to the data analysis system 10 through the GUI 401 displayed by the data analysis terminal 30.
  • the data analysis system 10 may generate the recipe information 500, the sample data 600, and the analysis result table 700 based on the input recipe generation condition 400, as in the above-described embodiment. Then, the data analysis system 10 may transmit the analysis result table 700 to the data analysis terminal 30. The data analysis terminal 30 may display the received analysis result table 700.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Automatic Analysis And Handling Materials Therefor (AREA)

Abstract

L'invention concerne un procédé d'analyse de données comprenant les étapes consistant à : générer, sur la base d'éléments d'informations de mesure comprenant des données de mesure pour chaque élément relatif à un échantillon, des informations sur une pluralité de recettes comprenant des combinaisons différentes des éléments ; appliquer chaque composante d'informations de recette aux données de mesure ; générer des données d'échantillon correspondant à chaque composante d'informations de recette ; déterminer, sur la base de la configuration des données d'échantillon, un moyen d'essai statistique devant être appliqué aux données d'échantillon ; appliquer le moyen d'essai statistique déterminé aux données d'échantillon ; calculer une notation d'essai indiquant une signification statistique relative aux informations de recette utilisées dans la génération des données d'échantillon ; et associer la notation d'essai calculée et les informations de recette relatives à la notation d'essai et afficher celles-ci.
PCT/JP2016/076268 2016-09-07 2016-09-07 Système d'analyse de données, terminal d'analyse de données et procédé d'analyse de données WO2018047251A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2018537917A JPWO2018047251A1 (ja) 2016-09-07 2016-09-07 データ分析システム、データ分析端末、及び、データ分析方法
PCT/JP2016/076268 WO2018047251A1 (fr) 2016-09-07 2016-09-07 Système d'analyse de données, terminal d'analyse de données et procédé d'analyse de données

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2016/076268 WO2018047251A1 (fr) 2016-09-07 2016-09-07 Système d'analyse de données, terminal d'analyse de données et procédé d'analyse de données

Publications (1)

Publication Number Publication Date
WO2018047251A1 true WO2018047251A1 (fr) 2018-03-15

Family

ID=61562739

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2016/076268 WO2018047251A1 (fr) 2016-09-07 2016-09-07 Système d'analyse de données, terminal d'analyse de données et procédé d'analyse de données

Country Status (2)

Country Link
JP (1) JPWO2018047251A1 (fr)
WO (1) WO2018047251A1 (fr)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004188026A (ja) * 2002-12-12 2004-07-08 Olympus Corp 情報処理装置
JP2005038256A (ja) * 2003-07-16 2005-02-10 Jgs:Kk 有効因子情報選択装置、有効因子情報選択方法、プログラム、および、記録媒体
JP2010146554A (ja) * 2008-12-17 2010-07-01 Internatl Business Mach Corp <Ibm> データ処理システム、コンピュータ可読の媒体、データ・マイニング結果を分析する方法、データ・モデルを処理する方法（統計技術を使用するデータ・マイニング・モデル解釈、最適化、及びカスタマイゼーション）
JP2011034457A (ja) * 2009-08-04 2011-02-17 Nec Corp データマイニングシステム、データマイニング方法及びデータマイニング用プログラム

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004188026A (ja) * 2002-12-12 2004-07-08 Olympus Corp 情報処理装置
JP2005038256A (ja) * 2003-07-16 2005-02-10 Jgs:Kk 有効因子情報選択装置、有効因子情報選択方法、プログラム、および、記録媒体
JP2010146554A (ja) * 2008-12-17 2010-07-01 Internatl Business Mach Corp <Ibm> データ処理システム、コンピュータ可読の媒体、データ・マイニング結果を分析する方法、データ・モデルを処理する方法（統計技術を使用するデータ・マイニング・モデル解釈、最適化、及びカスタマイゼーション）
JP2011034457A (ja) * 2009-08-04 2011-02-17 Nec Corp データマイニングシステム、データマイニング方法及びデータマイニング用プログラム

Also Published As

Publication number Publication date
JPWO2018047251A1 (ja) 2019-06-24

Similar Documents

Publication Publication Date Title
Weakley et al. The validity and reliability of commercially available resistance training monitoring devices: A systematic review
McAlinden et al. Validation of the Quality of Life Inventory for patients with anxiety and depression
US10478114B2 (en) Brain dysfunction assessment method, brain dysfunction assessment device, and program thereof
KR101855179B1 (ko) 질환 진단을 위한 최적의 진단 요소 셋 결정 장치 및 방법
JP2015062817A5 (ja) 脳活動解析装置、脳活動解析方法、判別器生成装置、判別器生成方法、バイオマーカー装置およびプログラム、健康管理装置およびプログラム、ならびに判別器のプログラム
Brammer The role of neuroimaging in diagnosis and personalized medicine-current position and likely future directions
TW201426620A (zh) 健康檢查路徑評估指標建構系統、方法、運用此方法之裝置及電腦程式產品
JP5854346B2 (ja) トランスクリプトーム解析方法、疾病判定方法、コンピュータプログラム、記憶媒体、及び解析装置
Pirrotta et al. Italian validation of Montreal cognitive assessment
US11955024B2 (en) System and method for normalizing activity results
Bauernhofer et al. Frenetic, underchallenged, and worn-out: Validation of the German “Burnout Clinical Subtypes Questionnaire”—Student survey and exploration of three burnout risk groups in university students
JP6644767B2 (ja) 相互相関に基づかれる階層的なデータ分析に関する方法及び装置
Davison et al. Quantitative data analyses
Thielsch et al. How informative is informative? Benchmarks and optimal cut points for E-Health Websites
Li et al. Variability: Human nature and its impact on measurement and statistical analysis
Walker et al. Fitness center service quality model confirmation SQAS-19.
WO2018047251A1 (fr) Système d&#39;analyse de données, terminal d&#39;analyse de données et procédé d&#39;analyse de données
Umaroh et al. DeLone and McLean Model of Academic Information System Success
JP2017126212A (ja) パスウェイ解析プログラム、パスウェイ解析方法、及び、情報処理装置
JP2012024555A (ja) 疲労のバイオマーカーおよびその利用
Verhagen Bayesian Item Response Theory models for measurement variance
Shih et al. Rating inter-rater reliability of Shih–Hsu Test of attention between an experienced psychiatric occupational therapist and an occupational therapy student: A pilot study
Skoczylas Congruence in research question, design, and analysis: A tutorial on the measurement of change in clinical speech and language research
King Personality traits and user behavior
Ren Dynamic Prediction Of Survival Data Using Single Or Multiple Longitudinal Markers

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16915670

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2018537917

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16915670

Country of ref document: EP

Kind code of ref document: A1