JP2010205218A - Data analysis support device, data analysis support system, data analysis support method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a data analysis support device or the like capable of easily making an analysis processor such as a computer perform targeted analysis even without having analytic technical knowledge or experience. <P>SOLUTION: An analysis provider registers analysis setting information (data item information indicating an attribute of each data item itself of analysis object data and analysis setting information that defines an analysis method, or the like) as analysis support information in a support information DB 6. When the analysis provider performs analysis setting about a new analysis, a control part 10 of an analysis support server 2 extracts similar data from the analysis support information registered in the support information DB 6 and presents the similar data as analysis support information. When an analysis user inputs analysis object data and its purpose, the control part 10 of the analysis support server 2 extracts the most similar one from the analysis support information registered in the support information DB 6, and executes analysis processing using an analysis method associated with the data of the most similar one. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a data analysis support apparatus that supports data analysis.

In recent years, with the development of information technology and the spread of the Internet, information sources have been diversified and a large amount of information has been accumulated. In addition, various analysis methods using a computer or the like have been proposed to analyze such information.
In general, the flow of data analysis includes (1) collection and selection of data to be analyzed, (2) selection of analysis method, (3) analysis based on the selected method, (4) browsing of analysis results, action It is done in the procedure of examination.

However, since the analysis target data includes a lot of information, there is a problem that it is difficult to select data to be used for the analysis (the above procedure (1)).
In addition, analysis is a highly specialized field, and if there is no specialized knowledge or experience, it is difficult to select an analysis method suitable for the purpose (the above procedure (2)).

On the other hand, with conventional analysis tools, tools with excellent operability such as visualization by iconifying each operation are provided, or a plurality of analyzes are combined based on past analysis history as in Patent Document 1. Data analysis systems that perform complex analysis have been provided.
In the data analysis system of Patent Document 1, history information of combination patterns of analysis methods (rank analysis, transition analysis, distribution analysis, correlation analysis, etc.) and information on recommended analysis patterns are accumulated, By displaying a guide screen and prompting the user to make various settings, the analysis is carried out by inheriting the past complex analysis pattern.

JP 2005-157896 A

  However, the data analysis system of Patent Document 1 is useful when performing a composite analysis combining a plurality of types of analysis techniques, but does not support an analysis that is not a composite analysis.

  The present invention has been made in view of such problems, and the object of the present invention is to easily cause an analysis processing apparatus such as a computer to perform an intended analysis even without analysis expertise or experience. It is to provide a data analysis support device or the like that can be used.

  In order to solve the above-described problem, a first invention is a data analysis support apparatus including a support information database for accumulating data used for data analysis support, and for each data item included in analysis target data, a data item Item information setting means for receiving setting of item information that defines attributes of itself, analysis setting means for receiving setting of analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data, the item information, and the item information Analysis support information registration means for registering analysis setting information as analysis support information in the support information database; item information or analysis setting information set for newly input analysis target data; Analysis support information having analysis setting information is extracted from the support information database and presented. And support information presenting means, the data analysis support apparatus comprising: a.

  A second invention is a data analysis support apparatus comprising a support information database for accumulating data used for data analysis support, wherein the support information database includes each analysis target data included in analysis performed in the past. Item information that defines the attributes of the data item itself and analysis setting information that defines the analysis method according to the analysis purpose of the analysis target data are registered in advance as analysis support information, and analysis for newly performed analysis is performed. Input means for inputting target data and analysis purpose, extraction means for extracting analysis support data and analysis support information that is input by the input means from the support information database, and analysis extracted by the extraction means Analysis target data input by the input means based on the analysis setting information set in the support information Perform an analysis process with a data analysis support apparatus characterized by comprising display means visible for displaying the analysis result.

  A third invention is a data analysis support apparatus including a support information database for storing data used for support of data analysis, and an item defining an attribute of the data item itself for each data item included in analysis target data Item information setting means for receiving setting of information, analysis setting means for receiving setting of analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data, and analysis support for the item information and the analysis setting information Analysis support information registration means for registering in the support information database as information, item information or analysis setting information set for newly input analysis target data, and analysis support having item information or analysis setting information that matches or is similar Analysis support information presentation means for extracting and presenting information from the support information database; and Display that performs analysis processing on the newly input analysis target data based on the analysis setting information set in the analysis support information presented by the analysis support information presenting means, and displays the analysis result in a viewable manner Means for supporting data analysis.

  In the first or third invention described above, the analysis support information presenting means preferably presents analysis support information having a high degree of similarity preferentially when a plurality of extraction results are obtained for the analysis support information. .

  In the first or third invention, the analysis support information presenting means desirably displays the extracted analysis support information side by side in the analysis setting information setting screen.

  In the first, second, or third invention, the item information preferably defines a data format, a data type, and data content as attributes of each data item.

  In the first, second, or third invention, the analysis setting information preferably defines at least an analysis purpose, a target category, an analysis method, and analysis target data.

  A fourth invention is a data analysis support system including a support information database for accumulating data used for data analysis support in cooperation with an analysis management system used by an analysis provider, the analysis management system including an analysis For each data item included in the target data, item information setting means for accepting the setting of item information that defines the attributes of the data item itself, and setting of analysis setting information that defines the analysis method according to the analysis purpose of the analysis target data An analysis setting means for receiving the analysis information, an analysis support information registration means for registering the item information and the analysis setting information in the support information database as analysis support information, and presenting analysis support information for the newly input analysis target data Presentation request means for requesting to perform the analysis, and the data analysis support system includes the analysis management system. Analysis support information having item information or analysis setting information that matches or similar to the set item information or analysis setting information for the newly input analysis target data in response to a request from the system. A data analysis support system comprising first analysis support information presenting means that extracts from a database and presents it to the analysis management system.

  A fifth invention is a data analysis support system comprising a support information database for accumulating data used for data analysis support in cooperation with an analysis processing system used by an analysis user, wherein the support information database includes: Analysis support information includes item information that defines attributes of each data item itself included in analysis target data in analysis performed in the past, and analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data. The analysis processing system is registered in advance, and the analysis processing system inputs the analysis target data and the analysis purpose, and the analysis target data input by the input unit and the analysis support information suitable for the analysis purpose are extracted. Extraction request means for requesting the analysis support system, and the data analysis support system includes the analysis process In response to a request from the system, the second analysis support information that is extracted from the support information database and is provided to the analysis processing system by analyzing the analysis target data input by the input means and the analysis support information suitable for the analysis purpose The analysis processing system further includes an analysis target input by the input unit based on the analysis setting information set in the analysis support information presented by the second analysis support information presentation unit. A data analysis support system comprising display means for executing an analysis process on data and displaying the analysis result in a viewable manner.

  According to a sixth aspect of the present invention, there is provided a data analysis including a support information database for accumulating data used for data analysis support by linking an analysis management system used by an analysis provider and an analysis processing system used by an analysis user. The analysis management system includes an item information setting unit that receives setting of item information that defines an attribute of the data item itself for each data item included in the analysis target data, and an analysis purpose of the analysis target data Analysis setting means for accepting setting of analysis setting information that defines an analysis method according to the method, analysis support information registration means for registering the item information and the analysis setting information in the support information database as analysis support information, and newly Presentation request means for requesting to present analysis support information about the input analysis target data, and The analysis processing system is configured to input an input unit for inputting analysis target data and an analysis purpose for analysis to be newly performed, and to analyze the analysis target data input by the input unit and analysis setting information suitable for the analysis purpose. Extraction request means for requesting the support system, and the data analysis support system includes item information or analysis settings set for the newly input analysis target data in response to a request from the analysis management system. Analysis support information having item information or analysis setting information that matches or similar to information is extracted from the support information database and presented to the analysis management system; and from the analysis processing system According to the requirements of the analysis, the analysis object data input by the input means and the analysis purpose Second analysis support information presenting means for extracting analysis support information from the support information database and presenting it to the analysis processing system, wherein the analysis processing system further comprises the second analysis support information presenting means. Display means for executing analysis processing on the analysis target data input by the input means based on the analysis setting information set in the analysis support information presented by, and displaying the analysis result in a viewable manner Is a data analysis support system characterized by

  According to a seventh aspect of the present invention, there is provided an item information setting step for receiving setting of item information that defines an attribute of a data item for each data item included in the analysis target data, and an analysis method according to the analysis purpose of the analysis target data An analysis setting step for receiving settings of analysis setting information to be defined, an analysis support information registration step for registering the set item information and analysis setting information as analysis support information in the support information database, and newly input analysis target data An analysis support information presenting step for extracting and presenting analysis support information having the item information or analysis setting information that matches or similar to the item information or the analysis setting information from the support information database. This is a characteristic data analysis support method.

  8th invention defines the item information which defines the attribute of each data item itself contained in analysis object data about the analysis object data in the analysis performed in the past, and the analysis method according to the analysis purpose of the analysis object data The analysis setting information is a data analysis support method using a support information database registered in advance as analysis support information, and an input step for inputting analysis target data and an analysis purpose for a newly performed analysis is input. Extraction step for extracting analysis support data and analysis support information suitable for the analysis purpose from the support information database, and analysis input information set based on the analysis setting information set in the extracted analysis support information And a display step for displaying the analysis result in a viewable manner. It is over data analysis support method

  According to a ninth aspect of the present invention, there is provided an item information setting step for accepting setting of item information that defines an attribute of the data item itself for each data item included in the analysis target data, and an analysis method according to the analysis purpose of the analysis target data An analysis setting step for receiving settings of analysis setting information to be defined, an analysis support information registration step for registering the set item information and analysis setting information as analysis support information in the support information database, and newly input analysis target data The analysis support information presenting step for extracting and presenting the analysis support information having the item information or the analysis setting information set with respect to the item information or the analysis setting information that matches or similar to the item information or the analysis setting information, and the presented analysis Based on the analysis setting information set in the support information, the newly input analysis Perform an analysis process for elephants data, a display step visible for displaying the analysis results, a data analysis support method, which comprises a.

A tenth invention is a program for causing a computer to function as the data analysis support apparatus described in the first invention.
An eleventh invention is a program for causing a computer to function as the data analysis support device described in the second invention.
A twelfth invention is a program for causing a computer to function as the data analysis support device described in the third invention.

  ADVANTAGE OF THE INVENTION According to this invention, even if there is no analysis expertise or experience, the data analysis assistance apparatus etc. which can make an analysis processing apparatus, such as a computer, easily perform target analysis can be provided.

System configuration example of data analysis support system 1 Hardware configuration example of the analysis support server 2 The figure explaining analysis data information DB61 The figure explaining analysis setting information DB62 The figure explaining analysis support information DB63 The flowchart which shows the flow of the analysis management process which the analysis support server 2 performs Display example of data import screen 91 used on the analysis management system 20 side Display example of analysis target data setting screen 92 Display example of analysis setting screen 93 The flowchart which shows the flow of the data analysis process which the analysis support server 2 performs Display example of analysis processing initial screen 95 used on the analysis processing system 30 side Display example of analysis result display screen 96 Example of analysis setting using data analysis support system 1 (Example 1) Example of analysis setting using data analysis support system 1 (Example 2)

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram showing a system configuration of a data analysis support system 1 according to the present invention.

  As shown in FIG. 1, the data analysis support system 1 includes an analysis support server 2, analysis provider terminals 7, 7,... Used by an analysis provider, and analysis user terminals 8, used by an analysis user. .., And each device is connected for communication with each other via a network 9.

  The analysis support server 2 links the analysis management unit 3 that functions as a server of the analysis management system 20, the analysis processing unit 4 that functions as a server of the analysis processing system 30, and the analysis management system 20 and the analysis processing system 30. And an analysis support unit 5 that performs analysis support, and a support information database (DB) 6 that stores information related to analysis support.

  As shown in FIG. 1, the analysis management system 20 includes an analysis management unit 3 of the analysis support server 2, analysis provider terminals 7, 7,... And a network 9. It is used to set methods, analysis points, action points, etc. Here, the analysis provider refers to a person who provides an analysis service, an analysis system, and the like to a user or the like.

  As shown in FIG. 1, the analysis processing system 30 includes an analysis processing unit 4, an analysis user terminal 8, 8,..., And a network 9 of the analysis support server 2, and wants to perform a desired analysis. An analysis user imports data (log, attribute data, etc.) used for analysis and uses it for browsing analysis results. Here, the analysis user refers to a person who uses a system related to analysis in order to obtain an analysis result.

  The analysis support unit 5 of the analysis support server 2 cooperates with the analysis management unit 3 and the analysis processing unit 4 to provide useful support for the analysis provider and the analysis user.

  Specifically, the analysis support unit 5 executes an analysis management process (see FIG. 6) described later. In the analysis management process, the analysis support unit 5 accumulates item information, analysis setting information, and the like of the analysis target data set by the analysis management unit 3 in the support information database 6 as analysis support information. The analysis support unit 5 extracts appropriate analysis support information from the support information database 6 according to the item information of the analysis target data set by the analysis management unit 3 and presents it as reference data.

  Moreover, the analysis support part 5 performs the data analysis process (refer FIG. 7) mentioned later. In the data analysis processing, the analysis support unit 5 selects the analysis target data input to the analysis processing unit 4 and the analysis method suitable for the analysis purpose from among the analysis support information accumulated in the support information database 6. And the extracted analysis method is presented to the analysis processing unit 4. The analysis processing unit 4 performs data analysis based on the presented analysis technique, and displays the analysis result in a viewable manner.

  The support information database 6 includes an analysis data information DB (database) 61, an analysis setting information DB 62, and an analysis support information DB 63. Specific contents of each data stored in the analysis data information DB 61, the analysis setting information DB 62, and the analysis support information DB 63 will be described later (see FIGS. 3, 4, and 5).

  In the following description, “database” may be abbreviated as “DB”, and “table” may be abbreviated as “TB”.

Next, the hardware configuration of the analysis support server 2 will be described with reference to FIG.
As shown in FIG. 2, the analysis support server 2 includes a control unit 10, a storage unit 11, a media input / output unit 13, a communication unit 15, an input unit 17, a display unit 19, a peripheral device I / F unit 21, and the like on a bus 23. It is connected and configured.

The control unit 10 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like.
The CPU calls and executes a program stored in the storage unit 11, ROM, recording medium, or the like to a work memory area on the RAM, and drives and controls each unit connected via the bus 23. The CPU of the control unit 10 executes analysis management processing (see FIG. 6) and data analysis processing (see FIG. 10) described later.
Further, the control unit 10 functions as the analysis management unit 3, the analysis processing unit 4, and the analysis support unit 5 of FIG.

  The ROM permanently holds a computer boot program, a program such as BIOS, data, and the like. The RAM temporarily holds the loaded program and data, and includes a work area used by the control unit 10 for performing various processes.

The storage unit 11 is an HDD (hard disk drive), and stores a program executed by the control unit 10, data necessary for program execution, an OS (operating system), and the like. These program codes are read by the control unit 10 as necessary, transferred to the RAM, and read and executed by the CPU.
The storage unit 11 also functions as the support information database 6 shown in FIG. 1, and stores an analysis data information DB 61, an analysis setting information DB 62, an analysis support information DB 63, and the like.

The media input / output unit 13 (drive device) is a media input / output device such as a floppy (registered trademark) disk drive, PD drive, CD drive, DVD drive, or MO drive, and performs data input / output.
The communication unit 15 includes a communication control device, a communication port, and the like, and is a communication interface that mediates communication with the network 9 and performs communication control.

The input unit 17 is an input device such as a keyboard, a pointing device such as a mouse, or a numeric keypad, and outputs input data to the control unit 10.
The display unit 19 includes, for example, a display device such as a liquid crystal panel or a CRT monitor, and a logic circuit (video adapter or the like) for executing display processing in cooperation with the display device. The display unit 19 is input under the control of the control unit 10. Display information is displayed on a display device.

  Next, the configuration of data handled by the data analysis support system 1 will be described with reference to FIGS.

  In the analysis data information DB 61 shown in FIG. 3A, information (hereinafter referred to as data item information) defining attributes for each data item included in data to be analyzed (hereinafter referred to as analysis target data). A set data item information table 71 and an actual data table 72 of analysis target data are stored.

  Here, as described above, the data item information table 71 is information defining attributes for each data item included in the analysis target data, and is defined by “data format”, “data type”, and “data content”. expressed. The data item information table 71 is set by the analysis provider in the analysis management unit 3 (or the analysis management unit 3 from the analysis provider terminal 7 via the network 9).

  The “data format” is information regarding the format of the data, and the value is “quantity” or “category” as shown in Table 1 below.

  The “data type” represents the type of data, and the values are “ID”, “object variable”, “explanatory variable”, “time series”, “attribute” as shown in Table 2 below. Etc.

  The “data content” represents the meaning of the data, and the values thereof are “key information”, “purpose associated information”, “position information”, “time information”, “time information”, “table information”, as shown in Table 3 below. Attribute ", etc.

  The actual data table 72 stores column names (data item names) of imported data and the actual data (data).

  For example, when log data 73 and attribute data 74 shown in FIG. 3B are imported as analysis target data, a data item information table 76 as shown in FIG. 3C is set by the analysis provider. To do.

The log data in FIG. 3B includes data items such as “user ID”, “purchased product ID”, “store ID”, “price”, “number”, and the attribute data is “user ID”, “sex cd”. (Code) ”,“ Regional cd (Code) ”and other data items are included.
In this case, as shown in FIG. 3C, the data item information corresponding to the data item “user ID” is set such that the data format is “quantity” and the data type is “ID”. Similarly, the data item information corresponding to the data item “purchased product ID” is set such that the data format is “category”, the data type is “object variable”, and the data content is “key information”.

An analysis setting information table 77 is stored in the analysis setting information DB 62 shown in FIG. The analysis setting information table 77 is information that defines an analysis method according to the analysis purpose of the analysis target data, and “analysis purpose”, “target category”, and “analysis method” are set in association with “analysis target data”. The
The analysis setting information table 77 is set by the analysis provider in the analysis management unit 3 (or the analysis management unit 3 from the analysis provider terminal 7 via the network 9).

  Here, the “target category” in the analysis setting information table 77 is classified into categories for the purpose of analysis, and the values are “correlation”, “prediction”, “ “Classification” and the like are stored.

The “analysis method” stores a method used for analysis. For example, names representing various analysis methods such as “association analysis”, “clustering”, “decision tree”, and the like are stored.
The “analysis target data” stores the data name (column name) of the analysis target data (actual data). For example, “product ID” is stored as a key, “price”, and “sex” are stored as targets.

  FIG. 4B is an example of analysis setting information 78 for analysis targeting the log data 73 and attribute data 74 of FIG. As shown in FIG. 4B, “Analysis purpose” is “Find what is bought at the same time”, “Target category” is “Correlation”, “Analysis method” is “Association analysis”, “Analysis target data” Is set to analysis setting information 78 such as “key: product ID, target: price, gender”.

  The setting of the analysis setting information 78 requires analysis expertise and experience. Therefore, in the data analysis support system 1 of the present embodiment, reference data (analysis support information) is presented when setting the analysis setting information 78. Details will be described later.

The analysis support information DB 63 shown in FIG. 5 stores analysis support information 79, 80,... For each analysis provided in the past. The analysis support information 79, 80, ... includes reference data item information 791, 801, ..., reference analysis setting information 792, 802, ....
The reference data item information 791, 801,... Is the same as that stored in the analysis data information DB 61 as the data item information table 71 in FIG. .. Is the same as that stored in the analysis setting information DB 62 as the analysis setting information table 77 in FIG.

  The analysis support information (the reference data item information table 71 and the reference analysis setting information 77) is registered in the analysis support information DB 63 by the analysis support unit 5 whenever analysis setting is performed in the analysis management unit 4 or periodically. Is done.

  Next, the flow of processing related to analysis support performed in the data analysis support system 1 will be described with reference to FIGS.

6 to 9 are flowcharts showing the flow of analysis management processing executed using the analysis support server 2 and the analysis provider terminal 7, and screen display examples at each stage.
10 to 14 are a flowchart showing a flow of data analysis processing executed using the analysis support server 2 and the analysis user terminal 8, and a screen display example at each stage.

First, the flow of the analysis management process in FIG. 6 will be described.
In the analysis management process, the control unit 10 of the analysis support server 2 first receives input (import) of log data and attribute data (step S101). The data input here is data used as a sample for actual analysis.

  In step S <b> 101, the control unit 10 of the analysis support server 2 reads data related to the data import screen 91 illustrated in FIG. 7 from the storage unit 11 and displays the data on the display unit 19. Alternatively, data related to the data import screen 91 is transmitted to the analysis provider terminal 7 via the network 9 and displayed on the display unit of the analysis user terminal 7.

  In the data import screen 91, when the data name “Log1.csv” and “Attribute.csv” are specified by the analysis provider and the “Import” button is pressed, the data is input to the analysis support server 2. The input actual data such as log data and attribute data is stored in the analysis data information DB 61 of the support information database 6.

  Next, the control unit 10 receives the setting of data item information (data format, data type, data content) for each data item of the input log data and attribute data in the analysis support server 2 (step S102).

  In step S <b> 102, the control unit 10 of the analysis support server 2 generates display data related to the analysis target data setting screen 92 and displays the analysis target data setting screen 92 illustrated in FIG. 8 on the display unit 19. Alternatively, display data regarding the analysis target data setting screen 92 is transmitted to the analysis provider terminal 7 via the network 9 and displayed on the display unit of the analysis provider terminal 7.

  The analysis target data setting screen 92 is provided with setting fields for setting the data format, data type, and data content for each data item included in the log data and attribute data imported in step S101. . In each setting column, settable values may be displayed so as to be selectable in the form of pull-down menus 925, 926, and 927. In this case, the pull-down menu 925 in the data format setting column includes each item (quantity and category) shown in Table 1 above, and the pull-down menu 926 in the data type setting column shows in Table 2 above. Each item (ID, purpose variable, explanatory variable, time series, attribute, etc.) is included, and in the pull-down menu 927 in the data content setting column, each item (key information, purpose associated information, Location information, time information, attributes, etc.).

In step S102, when the data item information for the log data and the attribute data is set, the control unit 10 stores the set data item information in the analysis data information DB 61 of the support information database 6.
For example, as shown in FIG. 8, the data item “user ID” of the log data is set such that the data format is “quantity” and the data type is “ID”. The data item “product ID” is set such that the data format is “category”, the data type is “object variable”, and the data content is “key information”. Similarly, data item information is also set for attribute data.

  When the setting of the data item information is completed, the control unit 10 of the analysis support server 2 accepts the setting of the analysis method (analysis purpose, purpose category, analysis method, target data selection). At this time, when an instruction to refer to another analysis is input from the input unit 17 (step S103; Yes), the control unit 10 reads display data related to the analysis setting screen 93 illustrated in FIG. Displayed on the unit 19. Alternatively, display data related to the analysis setting screen 93 is transmitted to the analysis provider terminal 7 via the network 9 and displayed on the display unit of the analysis provider terminal 7.

  As shown in FIG. 9, the analysis setting screen 93 includes an analysis setting area 931, a reference area 932, and an analysis target data display area 933.

  In the analysis setting area 931, input fields 931a, 931b, 931c, and 931d for each setting item (“analysis purpose”, “object category”, “analysis method”, “analysis target data”) are provided. The “analysis purpose” input field 931a can be input in text format. The “object category” input field 931b, the “analysis method” input field 931c, and the “analysis target data” input field 931d include: Each value can be selected and input in a pull-down format.

  In the analysis target data display area 933, the data item information set in step S102 for the input log data and attribute data (analysis target data) is displayed.

  In the reference area 932, analysis support information is displayed. The analysis support information displayed here is extracted and presented in the processing after the next step S104.

  In step S104, the control unit 10 extracts analysis support information suitable for the current analysis setting from the support information DB 6. That is, it matches the data item information or analysis setting information of the current analysis target data from the analysis support information (reference data item information or reference analysis setting information) already registered in the analysis support information DB 63 of the support information database 6. Or a similar thing is extracted and shown (step S104, step S105).

Here, the similar determination will be described.
When extracting the analysis support information, the control unit 10 determines the similarity between the registered analysis support information and the current analysis, for example, in the following order of <A> (1) to (5): Assume that the analysis support information to be referenced is extracted.

<A> When only analysis target data (data item information) is input (1) Data format, data type, and data content of log data and attribute data are the same (2) Data format, data type, and data of log data Same content (3) Data type of log data, same data content (4) Same data content of log data (5) Same data type of log data

Further, when extracting the analysis support information, the control unit 10 registers the data item information and a part of the analysis setting information (analysis purpose or purpose category) by referring to the information. Similar information may be extracted from the analysis support information.
In this case, for example, analysis support information referred to in the following order of priority <B> (1) to (5) is extracted.

<B> When the analysis purpose or purpose category of analysis target data (data item information) and analysis setting information is input (1) The data format, data type, and data content of log data and attribute data are the same, and analysis settings Information analysis purpose and target category are the same (2) Log data format, data type, and data content are the same, and analysis setting information target category is the same (3) Log data data type and data content are the same, and The same target category of analysis setting information (4) The same data content of log data and the same target category of analysis setting information (5) The same data type of log data and the same target category of analysis setting information

  In the example of the analysis setting screen 93 shown in FIG. 9, analysis support information is extracted for the analysis target data displayed in the analysis target data display area 933. In the analysis setting area 931, only the analysis purpose 931a is input, and it is assumed that the setting contents are not fixed for the purpose category 931b, the analysis method 931c, and the analysis target data 931d.

  The control unit 10 uses the contents of the data item information (data format, data type, data content) in the analysis target data display area 933 and the analysis purpose “find what is bought at the same time” in the analysis setting area 931 as search keys. The analysis support information to be referred to is extracted from the analysis support information DB 63 in accordance with either the above-mentioned priority order <A> or <B>. Then, the extraction result is displayed in the reference area 932.

  Although not shown in the figure, the data item information displayed in the analysis data display area 933 and the data item information displayed in the reference area 932 have the same data format, data type, data content, etc., respectively. You may make it display with the selected color and a corresponding display format.

  When a plurality of search results are obtained, only analysis support information having a high degree of similarity is preferentially displayed, or a search result selection number 932a is displayed in the reference area 932 as shown in FIG. Search results (reference analysis data) corresponding to the number are displayed.

  Here, as the search result, the reference data item information 932b including the data items “user ID”, “purchased product ID”, and “price ID” in the log data and the data item “gender cd” in the attribute data is extracted. (Upper reference area 932). The reference data item information 932b is associated with reference analysis setting information 932c in the lower part of the reference area 932.

  The analysis provider refers to the contents presented in the reference area 932 (particularly, “target category”, “analysis method”, “analysis target data” of the reference analysis setting information 932c), while the analysis setting area 931 is not set It is possible to set a purpose category 931b, an analysis method 931c, and analysis target data 931d.

  When the analysis setting is performed as described above (step S106), the control unit 10 associates the analysis setting information set in the analysis setting area 931 with the data item information of the analysis target data, and the analysis data information DB 61 The information is stored in the analysis setting information DB 62 and registered as analysis support information in the analysis support information DB 63 (step S107).

  In step S103, when an instruction to refer to another analysis is not input (step S103; No), the process of steps S104 to S105 is skipped, and the process proceeds to step S106. In this case, nothing is displayed in the reference area 932 of the analysis setting screen 93, and “analysis purpose” 931a, “object category” 931b, “analysis method” 931c, and “analysis target data” 931d in the analysis setting area 931 are displayed. The analysis provider himself / herself judges and inputs a desired value.

  As described above, a series of processing relating to analysis setting by the analysis provider or accumulation of analysis support information is completed.

Next, the flow of data analysis processing will be described with reference to FIGS.
In the data analysis process, the control unit 10 of the analysis support server 2 first receives input (import) of log data and attribute data, and input of an analysis purpose (steps S201 and S202). The data input here is the data that is actually subject to analysis.

  In steps S <b> 201 and S <b> 202, the control unit 10 of the analysis support server 2 reads display data related to the input screen 95 as illustrated in FIG. 11 from the storage unit 11 and displays the display data on the display unit 19. Alternatively, display data relating to the input screen 95 is transmitted to the analysis user terminal 8 via the network 9 and displayed on the display unit of the analysis user terminal 8.

  The input screen 95 includes an analysis target data input field 951 and an analysis purpose input field 952. In the input screen 95, for example, the data name “Log1.csv” or “Attribute.csv” is designated by the analysis user, the purpose of analysis is selected from the pull-down menu 953, and the “execute analysis” button 954 is pressed. Then, those data are input to the analysis support server 2.

Next, the control unit 10 of the analysis support server 2 determines an analysis method (step S203).
In step S <b> 203, the control unit 10 extracts analysis target data input in step S <b> 201 from the support information DB 6 and analysis support information suitable for the purpose of analysis. At this time, the control unit 10 determines data item information (data format, data type, data content) for each data item included in the data input in step S201, and uses the determined data item information and analysis purpose. Extract suitable analysis support information.

  The extraction of the analysis support information is performed in the same manner as the processing in step S104 in the flowchart of FIG. However, in this data analysis process, in order to determine the analysis method, only one piece of analysis support information needs to be extracted under a higher priority condition.

  When the appropriate analysis support information is extracted, the control unit 10 sets the analysis method associated with the analysis support information and the target data to be selected as the analysis method and target data for the input analysis target data.

Thereafter, the control unit 10 performs an analysis process based on the set analysis method (step S204), and displays the analysis result on the display unit 19 or transmits it to the analysis user terminal 8 for display (step S205). ).
FIG. 12 shows an example of the analysis result display screen 96.

  Through the series of data analysis processes described above, the analysis user can input the log data and attribute data to be analyzed and perform an operation of selecting an analysis purpose. Since the analysis is performed by the method and the analysis result is displayed, the analysis result can be easily browsed.

  As described above, in the data analysis support system 1 of the present embodiment, the analysis support server 2 includes data item information (data format, data type, data content, etc.) of analysis target data input by the analysis provider. Item definition) and analysis setting information (analysis purpose, purpose category, analysis method, analysis method such as target data) are registered in the support information DB 6 as analysis support information. When a new analysis setting is made for new analysis target data, the control unit 10 of the analysis support server 2 is similar from the past setting information (analysis support information DB 63) registered in the support information database 6. Setting information (reference data item information or reference analysis setting information) is extracted and presented as analysis support information.

  Therefore, the analysis provider can set an analysis method or the like while referring to the analysis support information. Therefore, even those who have no analysis expertise can easily set the analysis, and even beginners can easily set the analysis.

Also, when multiple pieces of analysis support information are extracted, they are presented preferentially from those with high similarity, or some of them are referred to in a selectable manner. A practical data analysis support system that is easy to use can be provided.
Further, when the analysis support server 2 presents the analysis support information, the analysis support server 2 generates a display screen having an analysis setting area (analysis setting area 931) and a reference area 932 in the same display screen. It is possible to provide a data analysis support system and the like that can be easily set up while referring to the analysis support information, and that is easier to use.

  Furthermore, in the data analysis support system 1, when analysis target data and its purpose are input from an analysis user, the control unit 10 of the analysis support server 2 stores past setting information (analysis data) registered in the support information DB 6. ) To extract the most similar analysis data (reference item information or reference analysis setting information), execute the analysis process using the analysis method associated with this analysis data, and present the analysis results to be viewable To do.

  Therefore, the analysis user can obtain the analysis result by a very simple operation.

  Furthermore, in the data analysis support system 1, when the analysis support information is extracted, the search target is data item information (data format, data type, data content) or analysis setting information (analysis purpose, purpose category). It is not necessary to use any information (personal information) about the analysis provider and analysis user. Therefore, the data analysis support system 1 has the secondary effect that personal information of other analysis providers and analysis users is not leaked in spite of referring to analysis performed by others in the past. Obtainable.

  Hereinafter, a specific example of data analysis support using the data analysis support system 1 of the present embodiment will be described.

FIG. 13 is an example of analysis settings performed by the analysis support server 2.
(1) The analysis to be performed this time is “analysis using information on the status of services used by finance (banks, etc.)”.
(2) “Analysis data of POS information system” is registered as similar analysis support information in the analysis support information DB.

  13A shows analysis target data, FIG. 13B shows analysis support information extracted by analysis management processing, and FIG. 13C shows analysis setting information set this time.

  That is, in step S101 to S102 of the analysis management process shown in FIG. 6, data item information of analysis target data is set as shown in FIG. When “refer to another analysis” is selected in step S103, the control unit of the analysis support server 2 extracts analysis support information having data item information similar to the set data item information, It is presented as shown in 13 (B).

  Comparing the data item information in FIGS. 13A and 13B, the “user ID”, “use service”, and “deposit amount” of the log data in FIG. Corresponding to “user ID”, “purchased product ID”, and “price”, the data format, data type, and data content are the same. In addition, “sex cd” of the attribute data in FIG. 13A corresponds to “sex cd” in FIG. 13B, and the data format, data type, and data content match. Therefore, it can be said that analysis support information having similar data item information has been extracted in accordance with the above-described priority order <A> (1) “the data format, data type, and data content of log data and attribute data are the same”.

  Further, in step S105 of FIG. 6, the reference analysis setting information associated with the data item information is extracted from the analysis support information DB 63 and presented as shown in FIG. 13B. The analysis setting information shown in FIG. 13B is that “analysis purpose” is “find what is bought at the same time”, “object category” is “correlation”, and “analysis method” is “association analysis”. The “analysis target data” is “key: product ID, target: price, gender”.

  The analysis provider can set the current analysis setting information while referring to the analysis support information (analysis setting information) presented in FIG. Specifically, as shown in FIG. 13C, “analysis purpose” is set to “find services used simultaneously”, “target category” is set to “correlation”, and “analysis method” is set to “association analysis”. The “analysis target data” is set as “key: use service, target: deposit amount, age”, and the content that suits the purpose of this time is set almost following the content of the analysis support information in FIG. it can.

  The point to be noted in the first embodiment described above is that even if the data item (column name) included in the log data or attribute data is different between the current analysis target data and the past analysis target data, the data item The information (data format, data type, data content) that matches (similar) information is preferentially extracted. That is, it is possible to refer to the setting contents of data analysis in different fields (such as finance and purchased product POS data) without depending on the column name included in the log data or the like.

Another specific example using the data analysis support system 1 of the present embodiment will be described.
FIG. 14 is another example of analysis setting performed in the analysis support server 2.
(1) The analysis to be carried out this time is “analysis using information on diseases and symptoms that tend to occur concurrently with medical care”.
(2) “Analysis data of POS information system” is registered as similar analysis support information in the analysis support information DB. This is the same as that referred to in the first embodiment.

  14, (A) shows analysis target data, (B) shows analysis support information extracted by the analysis management process, and (C) shows analysis setting information set this time.

  That is, in step S101 to S102 of the analysis management process shown in FIG. 6, data item information of analysis target data is set as shown in FIG. When “refer to other analysis” is selected in step S103, the control unit 10 of the analysis support server 2 extracts analysis support information having data item information similar to the set data item information, Presented as shown in FIG.

  Comparing the data item information in FIGS. 14A and 14B, the “user ID” and “disease / symptom” of the log data in FIG. 14A are “user ID” in FIG. Corresponding to “purchased product ID”, the data format, the data type, and the data content are the same. In addition, regarding the “region” of the log data in FIG. 14A, the “price” in FIG. 14B matches the data type and the data content. In addition, “sex cd” of the attribute data in FIG. 14A corresponds to “sex cd” in FIG. 14B, and the data format, data type, and data content are the same. Therefore, it can be said that the analysis support information having similar data item information is extracted according to the above-described priority order <A> (3) “the data type and the data content of the log data are the same”.

  Further, in step S105 in FIG. 6, the reference analysis setting information associated with the data item information is extracted from the analysis support information DB 63 and presented as shown in FIG. The analysis setting information shown in FIG. 14B is that “Analysis purpose” is “Find what is bought at the same time”, “Target category” is “Correlation”, and “Analysis method” is “Association analysis”. The “analysis target data” is “key: product ID, target: price, gender”.

  The analysis provider can set the current analysis setting information while referring to the analysis support information (analysis setting information) presented in FIG. Specifically, as shown in FIG. 14C, the “analysis purpose” is set to “search for easily occurring diseases and symptoms”, the “target category” is set to “correlation”, and the “analysis method” is set to “association analysis”. The “analysis target data” is “key: illness / symptom, target: region, age, gender”, and the content that fits the purpose of this time, almost following the content of the analysis support information in FIG. Can be set.

  In the second embodiment, as in the first embodiment, even if the data items (column names) included in the log data and the attribute data are different between the current analysis target data and the past analysis target data, Data items with similar data item information (data format, data type, data content) are preferentially extracted.

In the above-described embodiment, as an example of a system configuration, as shown in FIG. 1, (a) a data analysis support system 1 in which both an analysis management system 20 and a data analysis system 30 are linked by an analysis support server 2. However, the present invention is not limited to this.
For example, as a system aspect, (b) a data analysis support system in which the analysis management system 20 and the analysis support server 2 are linked, or (c) a data analysis support system in which the data analysis system 30 and the analysis support server 2 are linked. Good.

Further, the analysis support server 2 does not perform communication connection via the network 9 with the analysis provider terminal 7 of the analysis management system 20 or the analysis user terminal 8 of the data analysis system 30, and performs local processing as an analysis support apparatus. May be executed.
Also in this case, (d) the analysis support apparatus (analysis support server 2) performs only analysis management processing, (e) the analysis support apparatus (analysis support server 2) performs only data analysis processing, (f ) Any of an analysis support apparatus (analysis support server 2) that performs both analysis management processing and data analysis processing may be employed.

  In addition, it is obvious that those skilled in the art can come up with various changes and modifications within the scope of the technical idea disclosed in the present application, and these naturally belong to the technical scope of the present invention. It is understood.

DESCRIPTION OF SYMBOLS 1 ... Data analysis support system 2 ... Analysis support server 3 ... Analysis management part 4 ... Analysis processing part 5 ... Analysis support part 6 ... Support information database 61 ... Analysis data information DB
62 ... Analysis setting information DB
63 ... Analysis support information DB
7 ... Analysis provider terminal 8 ... Analysis user terminal 9 ... Network

Claims (16)

A data analysis support apparatus having a support information database for accumulating data used for data analysis support,
For each data item included in the analysis target data, an item information setting unit that accepts setting of item information that defines the attribute of the data item itself,
Analysis setting means for receiving setting of analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data;
Analysis support information registration means for registering the item information and the analysis setting information in the support information database as analysis support information;
Analysis support information that is extracted from the support information database and presented with analysis support information having item information or analysis setting information that matches or similar to item information or analysis setting information set for newly input analysis target data Presentation means;
A data analysis support apparatus comprising: A data analysis support apparatus having a support information database for accumulating data used for data analysis support,
The support information database includes item information that defines attributes of each data item itself included in analysis target data in analysis performed in the past, and analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data Is pre-registered as analysis support information,
For new analysis to be performed, input means for inputting analysis target data and analysis purpose;
Extraction means for extracting analysis support information suitable for the analysis object data and analysis purpose input by the input means from the support information database;
Display means for executing analysis processing on the analysis target data input by the input means based on the analysis setting information set in the analysis support information extracted by the extraction means, and displaying the analysis result in a viewable manner; ,
A data analysis support apparatus comprising: A data analysis support apparatus having a support information database for accumulating data used for data analysis support,
For each data item included in the analysis target data, an item information setting unit that accepts setting of item information that defines the attribute of the data item itself,
Analysis setting means for receiving setting of analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data;
Analysis support information registration means for registering the item information and the analysis setting information in the support information database as analysis support information;
Analysis support information that is extracted from the support information database and presented with analysis support information having item information or analysis setting information that matches or similar to item information or analysis setting information set for newly input analysis target data Presentation means;
Based on the analysis setting information set in the analysis support information presented by the analysis support information presenting means, an analysis process is performed on the newly input analysis target data, and the analysis result is displayed in a viewable manner. Display means;
A data analysis support apparatus comprising:   4. The analysis support information presenting means preferentially presents analysis support information having a high degree of similarity when a plurality of extraction results are obtained for the analysis support information. Data analysis support device.   The data analysis support apparatus according to claim 1 or 3, wherein the analysis support information presentation unit displays the extracted analysis support information side by side in a setting screen of analysis setting information.   The data analysis support apparatus according to any one of claims 1 to 3, wherein the item information defines a data format, a data type, and data contents as attributes of each data item.   The data analysis support apparatus according to any one of claims 1 to 3, wherein the analysis setting information defines at least an analysis purpose, a purpose category, an analysis method, and analysis target data. A data analysis support system comprising a support information database for accumulating data used for data analysis support in cooperation with an analysis management system used by an analysis provider,
The analysis management system includes:
For each data item included in the analysis target data, an item information setting unit that accepts setting of item information that defines the attribute of the data item itself,
Analysis setting means for receiving setting of analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data;
Analysis support information registration means for registering the item information and the analysis setting information in the support information database as analysis support information;
A presentation requesting means for requesting to present analysis support information for newly input analysis target data,
The data analysis support system
In response to a request from the analysis management system, analysis support information having item information or analysis setting information that matches or is similar to item information or analysis setting information set for the newly input analysis target data, A data analysis support system comprising first analysis support information presentation means that is extracted from the support information database and presented to the analysis management system. A data analysis support system comprising a support information database for accumulating data used for data analysis support in cooperation with an analysis processing system used by an analysis user,
The support information database includes item information that defines attributes of each data item itself included in analysis target data in analysis performed in the past, and analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data Is pre-registered as analysis support information,
The analysis processing system includes:
Input means for inputting analysis target data and analysis purpose;
Extraction request means for requesting the data analysis support system to extract analysis target data and analysis support information suitable for the analysis purpose input by the input means,
The data analysis support system
In response to a request from the analysis processing system, second analysis data extracted by the input means and analysis support information suitable for an analysis purpose are extracted from the support information database and presented to the analysis processing system. Provided with analysis support information presentation means,
The analysis processing system further performs an analysis process on the analysis target data input by the input unit based on the analysis setting information set in the analysis support information presented by the second analysis support information presentation unit. A data analysis support system comprising display means for executing the above and displaying the analysis result in a viewable manner. A data analysis support system comprising a support information database for accumulating data used for data analysis support by linking an analysis management system used by an analysis provider and an analysis processing system used by an analysis user,
The analysis management system includes:
For each data item included in the analysis target data, an item information setting unit that accepts setting of item information that defines the attribute of the data item itself,
Analysis setting means for receiving setting of analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data;
Analysis support information registration means for registering the item information and the analysis setting information in the support information database as analysis support information;
A presentation requesting means for requesting to present analysis support information for newly input analysis target data,
The analysis processing system includes:
For new analysis to be performed, input means for inputting analysis target data and analysis purpose;
Extraction request means for requesting the data analysis support system to extract the analysis target data input by the input means and analysis setting information suitable for the analysis purpose, and
The data analysis support system
In response to a request from the analysis management system, the analysis support information having item information or analysis setting information that matches or similar to item information or analysis setting information set for newly input analysis target data First analysis support information presenting means for extracting from the information database and presenting it to the analysis management system;
In response to a request from the analysis processing system, second analysis data extracted by the input means and analysis support information suitable for an analysis purpose are extracted from the support information database and presented to the analysis processing system. Analysis support information presentation means;
With
The analysis processing system further performs an analysis process on the analysis target data input by the input unit based on the analysis setting information set in the analysis support information presented by the second analysis support information presentation unit. A data analysis support system comprising display means for executing the above and displaying the analysis result in a viewable manner. For each data item included in the analysis target data, an item information setting step for accepting setting of item information that defines the attribute of the data item itself,
An analysis setting step for receiving setting of analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data;
An analysis support information registration step for registering the set item information and analysis setting information in the support information database as analysis support information;
Analysis support information that is extracted from the support information database and presents analysis support information having item information or analysis setting information that matches or is similar to the item information or the analysis setting information with respect to newly input analysis target data A presentation step;
A data analysis support method characterized by comprising: For the analysis target data in the analysis performed in the past, item information that defines the attribute of each data item itself included in the analysis target data, and analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data A data analysis support method using a support information database registered in advance as analysis support information,
For the new analysis to be performed, an input step for inputting analysis target data and analysis purpose,
An extraction step of extracting from the support information database analysis support information suitable for the input analysis object data and analysis purpose;
Based on the analysis setting information set in the extracted analysis support information, a display step for performing analysis processing on the input analysis target data and displaying the analysis result in a viewable manner;
A data analysis support method characterized by comprising: For each data item included in the analysis target data, an item information setting step for accepting setting of item information that defines the attribute of the data item itself,
An analysis setting step for receiving setting of analysis setting information that defines an analysis method according to the analysis purpose of the analysis target data;
An analysis support information registration step for registering the set item information and analysis setting information in the support information database as analysis support information;
Analysis support information that is extracted from the support information database and presented with analysis support information having item information or analysis setting information that matches or similar to item information or analysis setting information set for newly input analysis target data A presentation step;
Based on the analysis setting information set in the presented analysis support information, a display step for performing analysis processing on the newly input analysis target data and displaying the analysis result in a viewable manner;
A data analysis support method characterized by comprising:   A program for causing a computer to function as the data analysis support apparatus according to claim 1.   A program for causing a computer to function as the data analysis support apparatus according to claim 2.   A program for causing a computer to function as the data analysis support apparatus according to claim 3.

Images