JPH07311798A - Information display method for data and data analysis supporting system - Google Patents

Abstract

PURPOSE:To display the data in a form capable of letting a human efficiently perform analysis without omitting information as much as possible. CONSTITUTION:The correspondence relation of the array of picture elements and the array of the data described in a picture generation information file 112 corresponding to an inputted analysis object (101) and the correspondence relation of the color of the picture elements and the value of the data of the array of the data in a data file 111 are obtained (102) and pictures are generated (103) and displayed (104) corresponding to them. The respective pieces of the correspondence relation described in the picture generation information file 112 are determined so as to generate characteristic patterns useful for the analysis of the corresponding analysis object in the generated pictures. The color of the displayed picture element indicates the value of the data. Thus, even a large volume of the data can be displayed with a high information density as the pictures and the feature of the data array is recognized from the feature of the pictures.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data analysis support technology for supporting efficient analysis of a large amount of data, and in particular, it can quickly grasp the situation of many sites such as factories, distribution bases, and sales offices. The present invention relates to a data analysis support technique suitable for supporting data analysis of a logistics operation manager who must make a decision based on the above.

[0002]

2. Description of the Related Art It is difficult for a general person to analyze a large amount of data directly from the raw data itself.

Therefore, in the past, when analyzing a large amount of data, it has been widely practiced to divide a large amount of data into groups for each small amount of data and analyze the data sequentially for each of the divided groups.

Further, it has been widely practiced that a large amount of data is processed / collected into a small amount of information in advance and the small amount of information is analyzed to analyze the large amount of data.

[0005]

However, according to the method of analyzing the data for each of the divided groups, the timing of the analysis shifts between the groups, so that the data over a plurality of groups cannot be properly analyzed in real time. There was a problem. There was also a problem that the whole analysis took too much time.

Further, according to the method of processing / aggregating a large amount of data into a small amount of information in advance, a part of the information originally possessed by the data is lost due to the aggregation, so that the data is not always properly collected. Sometimes it cannot be analyzed.

Therefore, an object of the present invention is to provide a data analysis system capable of displaying data in a form that enables humans to efficiently analyze the data without losing data information as much as possible.

It is another object of the present invention to analyze data in a data analysis system by a method suitable for such a data analysis system.

[0009]

To achieve the above object, the present invention provides a method of displaying information of data included in an array of data in which each series of data is not associated with the coordinates of space. Then, according to the purpose of analysis, according to a predetermined correspondence relationship, there is a step of generating an image of a pixel color array from the data array, and a step of displaying the generated image, the correspondence relationship Is
An image pattern representing the feature of interest in the analysis according to the analysis purpose of the data array is determined to appear in the image to be displayed, the position of the data in the array of the data and the array of the pixels. The correspondence with the pixel position,
Provided is a data information display method characterized in that there is a correspondence between a data value and a pixel color.

Here, the array of data in which each series of data is not associated with the coordinates of the space means image data,
It is an array of data in which a series of data is characterized by a value having a logical meaning, not an array of measured data of each point in a three-dimensional space. For example, an array of monthly sales data for each product is applicable. In this example, the data array has two data series, a data series for a product and a data series for a month.

[0011]

According to the data information display method of the present invention,
An image pattern representing the feature of interest in the analysis according to the analysis purpose of the data array is determined to appear in the image to be displayed, the position of the data in the array of the data and the array of the pixels. The correspondence with the pixel position,
An image of a pixel color array is generated from the data array and displayed according to the correspondence between the data value and the pixel color.

Here, for example, if the correspondence relationship between the position of the data in the array of data and the position of the pixel in the array of pixels is determined so that one pixel is generated from one data, the image An image is generated that includes information about the data for the pixels included in. If the correspondence between the data value and the pixel color is determined so that each value that the data can have and the pixel color correspond to each other, the information of the color of each pixel is Matches the information in.

Next, regarding the display of an image, a display generally used for displaying an image is composed of an extremely large number of pixels. For example, a high-definition display of 1000 pixels × 1000 pixels has 1 million pixels. Therefore, if the pixels of the generated image and the pixels of such a display are displayed in a one-to-one correspondence, it is possible to display information on all one million data at once. Further, for example, in the case of a high-performance display capable of full-color display, the pixel color is 1670.
All colors can be displayed. Therefore, if it is up to 24 bits, it is possible to display each possible value of the data and the color of the pixel in a one-to-one correspondence, and in this case, the information of the individual data is completely lost. Absent.

Further, in the present invention, the above-mentioned two correspondences are defined so that an image pattern representing a feature of interest in the analysis according to the analysis purpose appears in the displayed image.

Therefore, the user can directly grasp the characteristics of the data array from the image pattern appearing in the displayed image and analyze the data array.

As described above, according to the present invention, the user is
Due to the high image and drawing pattern recognition ability of human beings, even large amounts of data can be efficiently analyzed.

[0017]

Embodiments of the data analysis support system according to the present invention will be described below.

First, a first embodiment of the present invention will be described.

FIG. 1 shows the hardware configuration of the data analysis support system according to the first embodiment.

As shown in the figure, the data analysis support system according to the first embodiment includes a computer 201 and an output device 20.
2, the input device 203, 204, and the storage device 205 are connected.

The output device 202 is a display device. As the display device 202, it is preferable to use a color display device, particularly a high-definition color graphic display. In FIG. 1, a keyboard and a mouse are shown as the input devices 203 and 204, respectively. However, another input device may be used. For example, as the input device 204, another pointing device such as a trackball may be used. If the storage device 205 has a sufficient capacity to execute programs and store a large amount of data files, a magnetic storage device, an optical disk device, or a semiconductor memory can be used as the storage device 205.

FIG. 1 shows an example of the hardware configuration of the data analysis system according to the present invention.
The data analysis support system according to the present invention can be realized by using various devices generally called a computer or a computer system. Specifically, it can be realized using a large-scale computer, a workstation, a personal computer or the like. In addition, a device dedicated to support data analysis can be used.

The storage device 205 has a data file 111 and an analysis-purpose-specific image generation information file 112.
And the data analysis support program is stored in advance. The data file 111 stores data to be analyzed. This data is a set of data handled as an array. The analysis-purpose-specific image generation information file 112 exists for each purpose of data analysis. In each analysis purpose-specific image generation information file 112, the correspondence relationship between the above-described data array and the pixel array in the image, that is, the position in the data array and the pixel in the image corresponding to the data in the image Information about the correspondence with the position and the correspondence between the data value and the pixel value is described. Hereinafter, information on the correspondence relationship between the data array and the pixel array in the image and the correspondence relationship between the data value and the pixel value will be referred to as “image generation information”. The image generation information is created so that the characteristic pattern useful for the analysis of the analysis target corresponding to the analysis target file 112 is generated in the image described above. Further, the image is created so that the image is generated by laying it on a two-dimensional plane without any gap. Here, the pixel value is a value that defines the color of the pixel (lightness, hue, saturation), and in the first embodiment, each of the values that the data can take and the color of the corresponding pixel are , The image generation information is determined so that there is a one-to-one correspondence as much as possible. If the color of a pixel is represented by 24 bits, and if the number of bits of data is 24 bits or less, each of the possible values of the data and the color of the corresponding pixel are made to correspond one-to-one. be able to.

The data analysis support program calculates a data analysis support process of generating an image from the data in the data file 111 and displaying it on the output device 202 according to the image generation information described in the analysis purpose-specific image generation information file 112. It is a program realized on 201.

Below, according to the data analysis support program,
The operation of the data analysis support process realized on the computer 201 will be described.

FIG. 2 shows a processing procedure performed by the data analysis support process.

As shown in the figure, in this processing, the purpose of analysis is input from the input device 203 or 204 (step 101). Then, using the input analysis purpose as a key, a data file 1 that stores the data to be analyzed
11 and the analysis-purpose-specific image generation information file 112 corresponding to this analysis purpose are read from the storage device 205 (step 102). However, the purpose of analysis and the data file 11
The correspondence between 1 and the analysis-purpose-specific image generation information file 112 is stored in the storage device 205 in advance.

Then, an image is generated from the data in the data file 111 according to the image generation information described in the analysis-purpose-specific image generation information file 112 (step 103). The image is generated by sequentially obtaining the two-dimensional coordinates that specify the position in the image generated by the pixel corresponding to each data and the value of the corresponding pixel.

Finally, the generated image is output to the output device 202 (step 104).

Since one pixel of the image displayed in this way corresponds to one piece of data, according to the present first embodiment, it is much more difficult than when data (for example, numerical values or text) is directly displayed. This means that information about each data could be displayed with high information density. Also, the value (color) of each pixel
Since the image generation information is created so that the value that can be taken by the data and the value that the data can have are as one-to-one as possible, the information displayed is not much lost. In addition, since human image recognition ability is extremely high, such an image enables intuitive analysis of a large amount of data, which was difficult with a raw data (text, numerical value) array. You can
Furthermore, since the lung case display information is created so that a characteristic pattern useful for the purpose of analysis occurs in the image, the user can intuitively analyze the data from this characteristic pattern. You will be able to do it.

In the first embodiment, the data file 111 is included in the analysis-specific image generation information file 112.
Although the correspondence between the data value and the pixel value is directly specified, the correspondence between the data processed and edited in the data file 111 and the pixel value may be specified. For example, the image generation information file for each analysis purpose 1
12, the correspondence relationship between the deviation value with respect to the data array of the data file 111 and the pixel value may be defined, and an image corresponding to the deviation value data of the data of the data file 111 may be generated and displayed. This is to obtain the deviation value of each data from the data of the data file 111, and to use this deviation value in the data file 111 in the above-mentioned processing.
It may be used instead of the data. Similarly, regarding the correspondence relationship between the data array and the pixel array in the image, similarly, the correspondence relationship between the array in which the data of the data file 111 is processed and edited and the pixel array is analyzed-specific image generation information file 112. The data corresponding to the data in the data file 111 may be processed and edited, and an image corresponding to the array may be generated and displayed.

Further, the image generation information file for each analysis purpose 1
The image generation information stored in 12 is the data file 111
It may be the program itself for generating an image from the data of. In this case, run this program on computer 201
The image is generated by executing the above.

The second embodiment of the present invention will be described below.

The second embodiment is a case in which a part of the above-mentioned first embodiment is modified and the life management of seasonal products is applied for the purpose of analysis. Therefore, the configuration and operation are almost the same as those in the first embodiment, and only different parts will be described.

Data file 11 used in the second embodiment
The format of No. 1 is shown in FIG.

As shown in the figure, the data file 111 according to the third embodiment stores the actual sales value for each product (single product) in time series (by week). The actual sales value in the figure is data associated with the pixels in the image in the second embodiment. It is said that large-scale retail stores, etc., have managed product scores ranging from tens of thousands to hundreds of thousands at the single item level.
The data amount of the data file 111 becomes enormous.

Next, in the second embodiment, instead of the analysis-purpose-specific image generation information file 112, the analysis-purpose-specific image generation / recognition information file 312 is used. In the second embodiment, an analysis purpose-specific image generation / recognition information file 312 corresponding to the analysis purpose “life management of (seasonal product)” will be described as an example. FIG. 4 shows the contents of the analysis purpose-specific image generation / recognition information file 312.

As shown in the figure, the analysis purpose-specific image generation / recognition information file 312 includes an identifier 5011 of a data file 111 to be analyzed using the analysis purpose-specific image generation / recognition information file 312 and predetermined data. An object code level program for generating an image so as to generate a characteristic pattern useful for lifespan management according to the correspondence between the array of pixels and the array of pixels in the image and the correspondence between the data value and the pixel value. 5012, predetermined normal pattern information 5013, abnormal pattern information 5014 (abnormal pattern 50141 and comment information 5014 indicating the cause thereof)
2) is stored. Here, the normal pattern indicates that the data in the data file 111 is analyzed as normal as a result of the analysis by the corresponding analysis purpose (life management) of the analysis-specific image generation / recognition information file 312. ,
If the characteristic pattern is the same as the pattern that should be formed, and the abnormal file is one that is analyzed as unhealthy, then the pattern that the characteristic pattern may form is likely to form. The pattern is equal to the pattern. Also,
In the comment information 50142, the data in the data file 111 forming the above-mentioned characteristic pattern forming a pattern equal to the corresponding abnormal pattern is used as the corresponding analysis purpose (life management) in the analysis-specific image generation / recognition information file 312. ) Shows the problems that should be obtained when analyzed.

The operation of the data analysis support process implemented on the computer 201 in accordance with the data analysis support program according to the second embodiment will be described below.

FIG. 5 shows the procedure of processing performed by the data analysis support process.

First, for the purpose of analysis (for seasonal products)
When "life management" is input from the keyboard 203 (step 101), the corresponding analysis purpose-specific image generation / recognition information file 312 is searched using the input analysis purpose (life management) as a key, and the analysis purpose is further analyzed. The data file 111 specified by the data identifier 5011 in the separate image generation / recognition information file 312 is searched (step 1
02). In addition, to accept the input of analysis purpose, prepare the available analysis purpose in advance as a menu and display it.
It may be accepted according to the designation of the menu by clicking the mouse 204.

Next, the program 5012 in the retrieved image-by-analysis purpose / recognition information file 312 is executed, and the image 6 from the data in the retrieved data file 111 is executed.
01 is generated.

Image generation by the program 5012 will be described.

First, as shown in FIG. 6, the vertical axis of the generated image is the product identification code, the horizontal axis is the time (week), and data (the number of sales of a specific product in a specific week) is assigned to the corresponding pixel. . Next, the value of each pixel is determined from the value of the assigned data.

Now, assuming that the number of bits of data is 8 bits, if the display 202 can express 256 gradations of light and shade (brightness and darkness), the pixel value can be determined in a one-to-one relationship with the number of sold products. . That is, for example, the number of sold items is set to be darkest (at level 0) and the number of sold items is set to be brightest (at level 255).

If the number of bits of data is 9 bits or more and the display 202 can express only gray scale (brightness / darkness) 256 gradations, the upper limit of the pixel value is limited to the level 255, or Has a maximum value of 25
Try to normalize so that it does not exceed 5. When the number of sold products is extremely different between products, it is desirable to add a normalization process to set the number of sold products in the best-selling week to level 255 for each product.

In this way, when the products are simply arranged in the order of the predetermined identification codes (the order of registration or the order of the size of the identification code), the brightness (luminance) of each line (pixel row) Since the distribution is extremely discontinuous and the displayed image is likely to flicker, when generating the image, the lines adjacent to the top and bottom should be rearranged so that they have almost the same brightness distribution. Good. This can be easily realized by obtaining the correlation value between the lines and collecting close ones.

Now, as described above, the data file 1 for a series of products which are considered to have essentially the same life characteristics.
In the image generated for 11, the characteristic patterns equal to the pattern 5013 can be expected to emerge when the life characteristics of the products are the same.

When the image is generated in this way, next, the analysis-purpose-specific image generation / recognition information file 31 is generated.
2 that the normal pattern information 5013 and the abnormal pattern information 5 that may appear when the above array display method is performed
014 is searched (step 304).

Next, the image 60 generated in step 103.
1 is compared with the normal pattern 5013 searched in step 304 to determine the similarity. Then, if the similarity is higher than a predetermined level, the data of the data file 111 is determined to be normal for analysis purpose “life management, and the result of the determination of normal is used as analysis result information.

If it is not judged to be normal,
Further, the image 601 generated in step 103 is compared with each abnormal pattern 5014 searched in step 304. And, for example, the abnormal pattern with the highest similarity
The comment information 50142 corresponding to the comment is the analysis result information.

The image 60 generated in step 103
1 and the normal pattern 5013 searched in step 304,
The similarity judgment with the abnormal pattern 5014 is performed by, for example,
This can be performed by applying a widely known pattern matching technique between images. Or normal pattern 5
013, normal pattern 5 instead of abnormal pattern 5014
013, power spectrum pattern of abnormal pattern 5014
Are stored in the analysis purpose image generation / recognition information file 312, and the similarity between this and the power spectrum pattern of the generated image obtained by the FFT processing is determined using a neural network or the like that has been learned in advance. ,
From this, the generated image 601 and the normal pattern 501
3. Similarity determination with the abnormal pattern 5014 may be performed. The similarity determination can be further facilitated by sampling the power spectrum pattern, reconstructing it by the mathematical formula of chaos theory, and then drawing it as a chaotic attractor to expand the difference.

When step 305 ends, finally, the image 601 and the analysis result information obtained in step 305 are output to the display 202. At this time, image 6
Since the feature pattern of 01 may be difficult to see, it is preferable to be able to interactively change the display level and display magnification of the pixel.

By the way, fashion products, which are representative of seasonal products, have the characteristics that they have a short lifespan and have been manufactured / stocked several months before the sales period. However, it will lead to a huge loss of sales opportunities and the end of the period. These life management problems may include seasonal trends (temperature and period) of the entire product group and misreading of the life characteristics of some products (early summer products → midsummer products).
Sales plans and inventory policies based on such false life recognition need to be discovered / corrected immediately.

However, according to the second embodiment, it is possible to immediately and intuitively grasp the problem caused by the sales plan or inventory policy based on the erroneous life recognition from the displayed image. Further, since the problem or the like is displayed as the analysis result information, the user can directly recognize the problem or the like from this and can determine the next action to be taken by using this as a guide.

The third embodiment will be described below.

The data analysis support system according to the third embodiment includes steps 305 and 306 (see FIG. 5) for performing processing for displaying analysis result information in the data analysis support system according to the second embodiment. It is deleted and 705 shown in FIG. 7 is added instead.

In step 705, the image 601 generated in step 103 and either or both of the normal pattern 5013 and the abnormal pattern 50141 retrieved in step 304 are directly displayed.

That is, in the third embodiment, the generated image 601, the normal pattern 5013 and the abnormal pattern 501 are included.
The similarity determination of 41 is left to the human image recognition ability.

In the first embodiment and the third embodiment, of the data in the data file 111, the data for which an image is to be generated is increased in range for a specific series. If repeated continuously for a short time, the image is displayed like a moving image in step 705.
For the data of FIG. 3, for example, if the process of generating an image for each of the weeks to which the data for which the image is to be generated belongs is sequentially increased from the past to the present and is repeated in a short time, It is possible to display a moving image that changes along with the transition of the number of sales of each product.

The fourth embodiment of the present invention will be described below.

In the fourth embodiment, the processing performed by the data analysis support process in the second embodiment is realized by using a dedicated circuit.

That is, as shown in FIG. 8, with the input analysis purpose as a key, corresponding analysis purpose-specific image generation /
A file search unit 802 that searches the recognition information file 312 and further searches the data file 111 specified by the identifier 5011 of the data in the analysis-specific image generation / recognition information file 312, and the analysis purpose searched by the file search unit 802. An image generation unit 804 that executes the program 5012 in the separate image generation / recognition information file 312 and generates an image from the data of the searched data file 111,
Image generation by analysis purpose searched by the file search unit 802 /
A pattern search unit 804 for searching a normal pattern 5013 and an abnormal pattern 5014 in the recognition information file 312,
The image generated by the image generation unit 804 and the pattern search unit 80
4 performs the similarity judgment of the normal pattern 5013 and the abnormal pattern 5014 searched by 4 to generate the analysis result information according to the judgment result. The comparison unit 805, the image generated by the image generation unit 804 and the comparison unit 805 are The dedicated device 2010 having a display unit 806 for displaying the generated analysis result information on the output device 202 is provided in place of the computer 201 shown in FIG.

According to such a dedicated device 2010, higher speed processing can be expected.

In each of the above embodiments, as the normal pattern 5013 and the abnormal pattern 5014, the images of the same size as the image generated from the data of the data file 111 are used. 5013, abnormal pattern 5
As 014, an image smaller than the image generated from the data of the data file 111 including only the characteristic part may be used. The similarity determination in this case is performed by, for example, normal pattern 50 in the generated image.
13. Perform between the abnormal pattern 5014 and the most matching portion.

Further, the pixels of the image generated in each of the above embodiments and the pixels on the screen of the output device 202 are not always 1: 1.
Need not correspond to.

Further, in each of the above embodiments, a two-dimensional image is generated and displayed from a two-dimensional data array, but a three-dimensional image is generated from a three-dimensional data array or three-dimensional data is generated. Each embodiment may be expanded so that an image is generated from a data array of an arbitrary dimension, such as by consolidating or expanding the array to generate a two-dimensional image.

The embodiments of the present invention described above are
In addition to the life management of seasonal products described above, the present invention can be applied to many fields in which a large amount of data needs to be analyzed.

For example, as shown in FIG. 9, the configuration shown in FIG. 8 is provided with a database management unit 900 that collects data online and stores it in the data file 111 of the storage device 205 and manages it. It can be applied to data analysis support in data management systems.

As described above, the configuration shown in FIG. 8 can be replaced with a process realized on the computer 201 by a program, and the database management unit 90
0 can also be replaced with a process realized on the computer 201 by a program.

[0071]

As described above, according to the present invention, there is provided a data analysis system capable of displaying data in a form in which a person can efficiently analyze the data, without omitting the information of the data as much as possible. be able to.

Further, it is possible to analyze data by a method suitable for such a data analysis system.

[Brief description of drawings]

FIG. 1 is a diagram showing a hardware configuration of a data analysis support system according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing a procedure of processing performed by the data analysis support system according to the first embodiment of the present invention.

FIG. 3 is a diagram showing the format of a data file used in the second embodiment of the present invention.

FIG. 4 is a diagram showing the contents of an image generation / recognition information file according to the second embodiment of the present invention.

FIG. 5 is a flowchart showing a procedure of processing performed by the data analysis support system according to the second embodiment of the present invention.

FIG. 6 is a diagram showing correspondence between pixels in an image and data in the second embodiment of the present invention.

FIG. 7 is a flowchart showing a procedure of processing performed by the data analysis support system according to the third embodiment of the present invention.

FIG. 8 is a diagram showing a configuration of a data analysis support system according to the first embodiment of the present invention.

FIG. 9 is a block diagram showing a configuration of a data management system according to an embodiment of the present invention.

[Explanation of symbols]

 201 ... Calculator 202 ... Output device (display) 203 ... Input device (keyboard) 204 ... Input device (mouse) 205 ... Storage device

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Takahiro Tajima 4-6 Kanda Surugadai, Chiyoda-ku, Tokyo Inside Hitachi Systems Business Division (72) Inventor Mitsutaka Kosaka 1099 Ozenji, Aso-ku, Kawasaki-shi, Kanagawa Hitachi, Ltd. Manufacturing Systems Development Laboratory

Claims (12)

[Claims] 1. A method of displaying information of data included in an array of data in which each series of data is not associated with a coordinate of space, according to a predetermined correspondence relationship according to an analysis purpose,
A step of generating an image that is an array of pixel colors from an array of data, and a step of displaying the generated image, wherein the correspondence is in the analysis of the data array according to the analysis purpose. The image pattern representing the feature of interest is
Correspondence between the position of the data in the array of data and the position of the pixel in the array of pixels, which is determined to appear in the image to be displayed, and the correspondence between the data value and the pixel color. Information display method of data characterized by. 2. The data information display method according to claim 1, wherein the data array is a data array that can obtain a specific analysis result when an analysis is performed according to the analysis purpose. A step of preparing in advance a feature that should appear in the image generated from the array of the data, and a step of determining the consistency between the image generated from the array of the data and the feature, A step of displaying the specific analysis result when the consistency is higher than a predetermined level, the method for displaying information of data. 3. The data information display method according to claim 2, wherein the feature is that the data array has a specific analysis result when an analysis according to the analysis purpose is performed. If it is, it is an image pattern that should appear in the image generated from the array of the data, and the information display method of the data. 4. The data information display method according to claim 1, wherein the data array is a data array that can obtain a specific analysis result when an analysis according to the analysis purpose is performed. Information of the data, including the step of preparing in advance an image pattern that should appear in the image generated from the array of the data, and the step of displaying the prepared image pattern together with the generated image. Display method. 5. The data information display method according to claim 1, 2, or 3, further comprising the step of rearranging the series of pixel color arrays for each series according to the color distribution of each series. How to display the characteristic data. 6. A data information display method according to claim 1, wherein the correspondence between the data value and the pixel color is the data value and the pixel brightness. A method of displaying data, characterized in that there is a correspondence relationship in which a hue, a saturation, or a combination thereof is associated with a data value. 7. A data file storing an array of data in which each series of data is not associated with coordinates in space,
Each is an information file associated with a specific analysis purpose, the image pattern representing the feature of interest in the analysis according to the corresponding analysis purpose of the data array,
The correspondence relationship between the position of the data in the array of data and the position of the pixel in the array of pixels, which is determined to appear in the image to be displayed, and the correspondence relationship between the value of the data and the pixel color are defined. Recording means for recording the information file storing the information to be recorded, input means for accepting the designation of the analysis purpose, and the information file recorded in the recording means and corresponding to the analysis purpose received by the input means. Image generating means for generating an image of an array of pixel colors from the array of data stored in the data file recorded in the recording means according to the two correspondences defined by the information A data analysis support system comprising: a display unit that displays the generated image. 8. The data analysis support system according to claim 7, wherein the information file obtains an analysis result that the data sequence is normal when the analysis is performed according to the corresponding analysis purpose. In the case of a data array that is generated, the normal pattern, which is an image pattern that should appear in the image generated from the data array, and the data array are
An image pattern that should appear in an image generated from the array of the data when the array of the data gives an analysis result that is abnormal when the analysis according to the corresponding analysis purpose is performed. The abnormal pattern of at least one is further stored as a reference pattern, and the display unit stores the reference pattern stored in the information file corresponding to the analysis purpose received by the input unit together with the image generated by the image generation unit. A data analysis support system characterized by displaying. 9. The data analysis support system according to claim 7, wherein the information file obtains an analysis result that the data array is abnormal when the analysis is performed according to the corresponding analysis purpose. If the data is an array of data to be stored, an abnormal pattern that is an image pattern that should appear in an image generated from the array of data, and a problem associated with the abnormal pattern that indicates the cause of the abnormality The data analysis support system further stores point information, and the data analysis support system matches the image generated by the image generation unit with the abnormal pattern stored in the information file corresponding to the analysis purpose received by the input unit. Further comprising a comparing means for determining, the display means is associated with an abnormal pattern whose consistency determined by the comparing means is equal to or higher than a predetermined level. The reason indicated by the problem information is, data analysis support system and displaying together images by the image generating means has generated. 10. The data analysis support system according to claim 7, wherein the information file obtains an analysis result that the data array is normal when the analysis is performed according to the corresponding analysis purpose. And a normal pattern, which is an image pattern that should appear in an image generated from the data array, the data analysis support system generates the image generation means. Further comprising a comparing means for determining the consistency between the image and the normal pattern stored in the information file corresponding to the analysis purpose received by the input means, wherein the display means determines the comparing means. When the consistency is equal to or higher than a predetermined level, a display indicating that the data arrangement is normal is displayed together with the display of the image generated by the image generating means. Data analysis support system which is characterized. 11. The data analysis support system according to claim 7, 8, 9 or 10, wherein the correspondence between the value of the data and the pixel color is the value of the data, the brightness, the hue and the color of the pixel. A data analysis support system characterized by a correspondence relationship that associates a degree or a combination thereof with a data value. 12. A database including a data file in which each data series stores an array of data not associated with spatial coordinates, and an information file each associated with a specific analysis purpose, wherein: The image pattern that represents the feature of interest in the analysis according to the corresponding analysis purpose of the data array is
The correspondence relationship between the position of the data in the array of data and the position of the pixel in the array of pixels, which is determined to appear in the image to be displayed, and the correspondence relationship between the value of the data and the pixel color are defined. Recording means for recording an information file storing information to be stored, a database management means for managing a database on the recording means, an input means for receiving designation of an analysis purpose, and an input recorded in the recording means. A pixel color array from the array of data stored in the data file of the database according to the two correspondences defined by the information stored in the information file corresponding to the analysis purpose received by the means. A data management system having an image generating means for generating the image and a display means for displaying the generated image.