TWI692696B - Text mining support method and device - Google Patents

Abstract

A text mining support method and device, when displaying a scatter diagram representing the result of corresponding analysis, displays a support screen including a scatter diagram and an enlightenment representing the view of the scatter diagram. When displaying a scatter diagram related to words and variables, the basic screen without revelation, the judgment method containing words near the origin as the first support screen for revelation, and the judgment method containing the degree of relevance of words that characterize variables The second support screen as a revelation, the judgment method including the similarity between words as the third support screen as the revelation, and the judgment method as the similarity between the variables as the fourth support screen as the revelation display the instructions indicated by the user Screen. Thereby, it is possible to efficiently perform a process of guiding insights from the graph showing the results of the corresponding analysis.

Description

Text mining support method and device

The present invention relates to a data mining technology, and in particular to a text mining support method and device that supports the execution of text mining.

In recent years, data mining techniques that apply statistics or pattern recognition and other data analysis techniques to a large amount of data, and lead insights (rules, etc. appearing in the data) from a large amount of data are attracting attention. Data mining that takes text data as an object is called text mining. In the following, consider the case of performing correspondence analysis as a kind of data analysis technique for text data.

In the correspondence analysis, the cross tabulation table is rearranged in such a way that the relationship between the header item and the table-side item becomes the largest. The results of the correspondence analysis are usually expressed using scatter diagrams (two-dimensional charts). For example, if the composite table shown in FIG. 2 is subjected to correspondence analysis, the scatter diagram shown in FIG. 3 can be obtained.

Japanese Patent Laid-Open No. 2005-44087 related to the invention of the present application describes a text mining system that presents a user with an analysis flow when using multiple analysis tools. If the system described in the above document is used, even a user with little knowledge or experience related to text mining can use multiple analysis tools for analysis in an appropriate order.

[Problems to be Solved by the Invention] In correspondence analysis, it is more important to examine the obtained scatter diagram than to find the scatter diagram, and to guide insights. However, users with little knowledge or experience related to text mining do not understand the view of the scatter diagram, so even if they view the scatter diagram, they do not understand what to do first. Therefore, a user with little knowledge or experience cannot efficiently perform the process of guiding insights from the scatter diagram.

Although the system described in Patent Document 1 presents the analysis flow to the user, it does not support the process of guiding insights from the analysis results. Therefore, even if the system described in Patent Document 1 is used, the above problem cannot be solved.

Therefore, an object of the present invention is to provide a text mining support method and apparatus for efficiently performing processing that leads insights from a graph showing the results of correspondence analysis.

[Technical Means for Solving the Problems] In order to achieve the above object, the present invention has the following features.

The first aspect of the present invention is a text mining support method, which displays the analysis result obtained by the corresponding analysis, and includes the step of inputting the analysis result; the step of inputting an instruction from the user; The step of displaying the screen data of the screen of the chart; and the step of displaying the screen based on the screen data; and the step of generating the screen data corresponds to the instruction and generates an inspiration including the chart and the view representing the chart ( hint) screen data of the supported screen.

The second aspect of the present invention is the first aspect of the present invention, wherein the step of generating screen data generates from a plurality of support screens and a basic screen including the chart without the revelation, corresponding to the Indicates the screen data of the selected screen.

The third aspect of the present invention is the second aspect of the present invention. In the step of inputting the analysis result, the result of associating the first item with the second item, that is, the first item including the first item is input. The results of the component and the second component and the first component and the second component of the second item are used as the analysis result, and the step of generating the screen data is made by using the first component as the horizontal axis and the first A scatter diagram obtained by plotting the first item and the second item in a plane with two components as the vertical axis is used as the graph.

The fourth aspect of the present invention is the third aspect of the present invention, wherein the plurality of support screens include a first support screen, and the first support screen includes the first item near the origin within the scatter diagram and has no significant The meaning of the characteristics serves as the revelation.

A fifth aspect of the present invention is the fourth aspect of the present invention. The scatter diagram included in the first support screen shows a range around the origin.

A sixth aspect of the present invention is the third aspect of the present invention, wherein the plurality of support screens include a second support screen including the second support screen located in the scatter diagram and departing from the origin to the second item The first item in the direction gives the feature to the second item as the revelation.

A seventh aspect of the present invention is the sixth aspect of the present invention. The scatter diagram included in the second support screen shows a range of directions from the origin to the selected second item.

An eighth aspect of the present invention is the third aspect of the present invention, wherein the plurality of support screens include a third support screen, and the third support screen includes the first items that are close in distance in the scatter diagram and have a high degree of similarity to each other As the revelation.

The ninth aspect of the present invention is the eighth aspect of the present invention, in which the range around the selected first item is shown in the scatter diagram included in the third support screen.

A tenth aspect of the present invention is the third aspect of the present invention, wherein the plurality of support screens include a fourth support screen, and the fourth support screen includes second items that are close in distance in the scatter diagram and have a high degree of similarity to each other As the revelation.

According to an eleventh aspect of the present invention, in the tenth aspect of the present invention, the scatter diagram included in the fourth support screen is shown with a symbol indicating the second item closest to the selected second item. .

A twelfth aspect of the present invention is the third aspect of the present invention. In the step of inputting the analysis result, an input pair includes a word as the first item, and a part of the article as the second item. The frequency of occurrence of each word in each part of is used as a result of corresponding analysis of a composite table of data in the table as the analysis result.

A thirteenth aspect of the present invention is a text mining support device that displays an analysis result obtained by corresponding analysis, and includes an analysis result input section for inputting the analysis result; an instruction input section for inputting instructions from a user; A screen generating unit that generates screen data including a screen representing a graph of the analysis result; and an analysis result display unit that displays a screen based on the screen data; and the screen generating unit corresponding to the instruction generates a screen including the The screen data of the graph and the support screen showing the inspiration of the graph.

A fourteenth aspect of the present invention is the thirteenth aspect of the present invention, wherein the screen generating unit generates from a plurality of support screens and a basic screen including the graph without the revelation, corresponding to the instruction Screen data of the selected screen.

A fifteenth aspect of the present invention is the fourteenth aspect of the present invention. In the analysis result input unit, a result of associating the first item with the second item, that is, the first component including the first item is input And the second component and the results of the first component and the second component of the second item are used as the analysis results, and the screen generating unit is configured to use the first component as the horizontal axis and the second component as A scatter diagram obtained by plotting the first item and the second item in the plane of the vertical axis is used as the graph.

A sixteenth aspect of the present invention is the fifteenth aspect of the present invention. In the analysis result input section, a pair of words is used as the first item, and a part of the article is used as the second item. The appearance frequency of each word in each part is used as the analysis result of the corresponding analysis of the composite table of the data in the table as the analysis result.

[Effect of the Invention] According to the first aspect or the thirteenth aspect, the user can use the support screen including the graph showing the result of the correspondence analysis and the inspiration showing the view of the graph to efficiently perform the process of displaying the result of the correspondence analysis. The chart leads to the processing of insights.

According to the second aspect or the fourteenth aspect, by selectively displaying the support screen including the inspiration and the basic screen without the inspiration, a screen corresponding to the level of the user can be displayed. In addition, by selectively displaying multiple support screens, users can be presented with a variety of chart views.

According to the third aspect or the fifteenth aspect, the user can efficiently perform a process of leading insights from a scatter diagram showing the results of the correspondence analysis related to the first item and the second item.

According to the fourth aspect, the user can use the knowledge that the first item near the origin in the scatter diagram does not have distinctive features, and efficiently perform the process of guiding insights from the graph showing the results of the corresponding analysis.

According to the fifth aspect, the user can view the illustrated range and easily know the first item that does not have a distinctive feature.

According to the sixth aspect, the user can use the knowledge of the first item located in the direction of departure from the origin to the second item in the scatter plot to characterize the second item and efficiently perform the following The graphs corresponding to the results of the analysis lead to the processing of insights.

According to the seventh aspect, the user can view the illustrated range and easily know the first item that characterizes the selected second item.

According to the eighth aspect, the user can use the knowledge that the first items that are close in the scatter diagram have high similarity to each other, and efficiently perform the process of guiding insights from the graph showing the results of the corresponding analysis.

According to the ninth aspect, the user can view the illustrated range and easily know the first item having a high degree of similarity to the selected first item.

According to the tenth aspect, the user can use the knowledge that the second items in the scatter diagram that are close to each other have a high degree of similarity, and efficiently perform a process of guiding insights from the graph showing the results of the corresponding analysis.

According to the eleventh aspect, the user can view the illustrated symbol and easily know the second item with the highest similarity to the selected second item.

According to the twelfth aspect or the eighteenth aspect, the user can efficiently perform a process of guiding insights from a scatter diagram showing the results of correspondence analysis related to words and text parts.

These objects, features, forms, and effects of the present invention and other objects, features, forms, and effects will become clearer with reference to the accompanying drawings and the following detailed description.

Hereinafter, a text mining support method, a text mining support device, and a text mining support program according to an embodiment of the present invention will be described with reference to the drawings. The text mining support method of this embodiment is typically executed using a computer. The text mining support device of this embodiment is typically constructed using a computer. The text mining support program of this embodiment is a program for implementing a text mining support method using a computer. The computer that executes the text mining support program functions as a text mining support device.

FIG. 1 is a block diagram showing the structure of a text mining support device according to an embodiment of the present invention. The text mining support device 10 shown in FIG. 1 includes an analysis result input unit 11, an instruction input unit 12, a screen generation unit 13, and an analysis result display unit 14. The result of the corresponding analysis on the text data is input to the text mining support device 10. The text mining support device 10 displays a scatter diagram showing the input analysis result on the screen.

In FIG. 1, a text analysis device 5 is provided in front of the text mining support device 10. The text data 1 is input into the text analysis device 5. In the following description, the text material 1 is an article material having multiple parts (hereinafter referred to as "chapters"). In addition, in the scenario where correspondence analysis is performed, the “chapter” is also called a “variable”. The text analysis device 5 extracts the words contained in the text data 1, and creates a compound table with the words as the table side items, the chapters as the table head items, and the appearance frequency of each word in each chapter as the data in the table. The text analysis device 5 performs corresponding analysis on the created composite table, and outputs the analysis result 2. By performing the correspondence analysis, two or more components representing the characteristics of the processing target data can be obtained. The analysis result 2 includes at least the first and second components of each word, the first and second components of each variable, the contribution rate of the first component, and the contribution rate of the second component.

FIG. 2 is a diagram showing a composite table that is an object of correspondence analysis. The composite table shown in FIG. 2 is created by inputting the article material of the novel “Human Disqualification” as text material 1 into the text analysis device 5. This novel is a Japanese novel with 5 chapters of "Preface", "First Notes", "Second Notes", "Third Notes" and "Postscript", including "Self", "Human" and "Halibut" , "Mood" and other words. The compound table shown in FIG. 2 contains words such as "self", "human", "flatfish", and "mood" as table-side items, and contains "preface", "first note", "second note", " The five variables (chapters) of the "third note" and "postscript" are used as the header items. The word "human" appears 38 times in the "first note". Corresponding to this, in the compound table shown in FIG. 2, 38 is written in the column (slashed part) in which the front-side item is “human” and the head item is “first note”. In addition, in order to appropriately perform correspondence analysis, the compound table shown in FIG. 2 includes only words having a frequency of occurrence that is more than a predetermined frequency.

FIG. 3 is a diagram showing a scattergram created by the text mining support device 10. As described above, the analysis result 2 input to the text mining support device 10 includes at least the first and second components of each word, the first and second components of each variable, and the contribution rate of the first component, And the contribution rate of the second component. The screen generating unit 13 plots words and variables in a plane with the first component as the horizontal axis and the second component as the vertical axis, thereby creating a scatter diagram. For example, based on the analysis result 2 regarding the composite table shown in FIG. 2, the scatter diagram shown in FIG. 3 is produced. The analysis result display unit 14 displays a screen including the created scattergram.

In FIG. 3, a black circle is written at the position of the word, a hollow square is written at the position of the variable, the word is a standard body, and the variable is written in italics. FIG. 3 shows the contribution rate of the first component and the contribution rate of the second component. Generally, the contribution rate of the first component is greater than the contribution rate of the second component. Considering this point, the distance d between the two points P (p ₁ , p ₂ ) and Q (q ₁ , q ₂ ) in the scatter diagram is the contribution rate k ₁ using the first component and the contribution rate k of the second component _{2. It} is defined as follows (1). d=√[{k ₁ (p ₁ -q ₁ )} ² +{k ₂ (p ₂ -q ₂ )} ² ]…(1) The distance in the following description refers to the definition by formula (1) The distance within the scatter plot. The circle described in the scatter diagram looks like an ellipse whose length in the first component direction is shorter than the length in the second component direction.

FIG. 4 is a block diagram showing the configuration of a computer that functions as the text mining support device 10. The computer 20 shown in FIG. 4 includes a central processing unit (Central Processing Unit, CPU) 21, a main memory 22, a memory unit 23, an input unit 24, a display unit 25, a communication unit 26, and a recording medium reading unit 27. In the main memory 22, for example, dynamic random access memory (Dynamic Random Access Memory, DRAM) is used. In the memory section 23, for example, a hard disk or a solid-state drive is used. The input unit 24 includes, for example, a keyboard 28 or a mouse 29. For the display unit 25, for example, a liquid crystal display is used. The communication section 26 is an interface circuit for wired communication or wireless communication. The recording medium reading unit 27 is an interface circuit for recording media 30 such as a program. In the recording medium 30, for example, non-temporary use of compact disc-read only memory (CD-ROM), digital video disc-read only memory (DVD-ROM), etc. Recording media.

When the computer 20 executes the text mining support program 31, the memory unit 23 stores the text mining support program 31 and the analysis result 2. The text mining support program 31 and the analysis result 2 may be received from a server or another computer using the communication unit 26, or may be read from the recording medium 30 using the recording medium reading unit 27, for example.

When the text mining support program 31 is executed, the text mining support program 31 and the analysis result 2 are copied and transferred to the main memory 22. The CPU 21 uses the main memory 22 as a working memory, and executes the text mining support program 31 stored in the main memory 22, thereby processing the analysis result 2 stored in the main memory 22. At this time, the computer 20 functions as the text mining support device 10. In addition, the configuration of the computer 20 described above is just an example, and any computer can be used to configure the text mining support device 10.

A user who has knowledge or experience related to text mining has the following knowledge about a scatter diagram representing the results of correspondence analysis. Users with knowledge or experience can use this knowledge to guide insights from the scatter diagram. The first knowledge "Words near the origin do not have significant features" The second knowledge "Words located in the direction away from the origin to the variable have a high degree of correlation with the variable and give the variable a feature" The third knowledge "close words have a high similarity to each other" The fourth knowledge "close variables have a high similarity to each other"

On the other hand, users who have little knowledge or experience related to text mining do not have the knowledge described above. Therefore, a user with little knowledge or experience cannot efficiently perform the process of guiding insights from the scatter diagram. In order to solve the above problem, the text mining support device 10 not only displays a screen including a scatter diagram as a basic screen, but also includes a scatter diagram and a hint that represents the scatter diagram in response to an instruction from a user. The screen is displayed as a support screen.

The operation of each part of the text mining support device 10 will be described with reference to FIG. 1. The analysis result 2 output from an external device (for example, the text analysis device 5) is input to the analysis result input unit 11. An instruction from the user is input into the instruction input unit 12. The screen generating unit 13 creates a scatter diagram showing the analysis result 2 and generates screen data of the screen including the scatter diagram. The screen generating unit 13 selectively generates screen data including a scatter diagram and a support screen of the enlightenment and screen data including a scatter diagram and a basic screen without enlightenment in response to the instruction input from the user using the instruction input unit 12 . The analysis result display unit 14 displays the screen based on the screen data generated by the screen generation unit 13. Hereinafter, four types of support screens displayed by the text mining support device 10 will be referred to as the first to fourth support screens.

FIG. 5 is a flowchart showing the operation of the text mining support device 10. First, the CPU 21 transfers the analysis result 2 output from the text analysis device 5 to the main memory 22. Thus, the analysis result 2 is input to the text mining support device 10 (step S101). Then, the CPU 21 creates a scatter diagram based on the analysis result 2 (step S102). The scatter plot is created by plotting words and variables in a plane with the first component as the horizontal axis and the second component as the vertical axis. Then, the CPU 21 generates screen data including the basic screen of the scattergram created in step S102 (step S103). Then, the CPU 21 displays the basic screen on the display unit 25 based on the screen data generated in step S103 (step S104).

6 is a diagram showing a basic screen. The basic screen 100 shown in FIG. 6 includes a screen selection window 101 and a scatter diagram window 102. The scatter diagram shown in FIG. 3 is described in the scatter diagram window 102. The screen selection window 101 has six radio buttons 103. Hereinafter, the six radio buttons 103 are referred to as the first radio button to the sixth radio button. The first radio button to the sixth radio button are respectively associated with the base screen, the first support screen to the fourth support screen, and the end. When the basic screen 100 is displayed, the user operates the keyboard 28 or the mouse 29 and presses any one of the first radio button to the sixth radio button. Thus, an instruction from the user is input.

The CPU 21 accepts an instruction from the user input in the use screen selection window 101 (step S105). Then, in response to an instruction from the user, the CPU 21 proceeds to any one of the following steps (step S106). When the instruction from the user is "basic screen" (when the first radio button is pressed), the CPU 21 proceeds to step S107. In this case, the CPU 21 generates screen data of the basic screen as in step S103 (step S107). When the instruction from the user is "first support screen" (when the second radio button is pressed), the CPU 21 proceeds to step S108. In this case, the CPU 21 generates screen data of the first support screen (step S108). When the instruction from the user is "second support screen" (when the third radio button is pressed), the CPU 21 proceeds to step S109. In this case, the CPU 21 generates screen data of the second support screen (step S109). When the instruction from the user is "third support screen" (when the fourth radio button is pressed), the CPU 21 proceeds to step S110. In this case, the CPU 21 generates screen data of the third support screen (step S110). When the instruction from the user is "the fourth support screen" (when the fifth radio button is pressed), the CPU 21 proceeds to step S111. In this case, the CPU 21 generates screen data of the fourth support screen (step S111). When the instruction from the user is "end" (when the sixth radio button is pressed), the CPU 21 ends the processing.

After the CPU 21 executes any one of steps S107 to S111, the process proceeds to step S112. Then, the CPU 21 displays the screen on the display unit 25 based on the screen data generated in any one of steps S107 to S111 (step S112). Then, the CPU 21 proceeds to step S105. In this way, the text mining support device 10 displays a screen selected from the basic screen and the first to fourth support screens in response to an instruction from the user.

In addition, the constituent elements of the computer 20 shown in FIG. 4 and the steps shown in FIG. 5 correspond to the constituent elements of the text mining support device 10 shown in FIG. 1 as follows. The CPU 21 executing step S101 functions as the analysis result input unit 11. The input unit 24 and the CPU 21 executing step S105 function as the instruction input unit 12. The CPU 21 that executes steps S102 to S103 and steps S106 to S111 functions as the screen generator 13. The display unit 25 and the CPU 21 that executes steps S104 and S112 function as the analysis result display unit 14.

7 is a diagram showing a first support screen. The first support screen 110 shown in FIG. 7 includes a screen selection window 101, a scatter diagram window 112, a word list window 113, and a hint window 114. The first support screen 110 is related to the first knowledge "Words near the origin have no distinctive features". The user can view the first support screen 110 and use the first knowledge to efficiently perform the process of guiding insights from the scatter diagram.

Before displaying the first support screen 110, the user operates the keyboard 28 or the mouse 29 to designate the range determined to be near the origin. The initial value of the range judged to be near the origin may be determined in advance. The scatter diagram shown in FIG. 3 is described in the scatter diagram window 112. In the scatter diagram in the scatter diagram window 112, a circle 115 (the appearance is an ellipse) near the origin is described. The circle 115 is preferably written in a different color (for example, red) from the scatter diagram. In this way, in the scatter diagram included in the first support screen 110, the range around the origin is shown by a circle 115. Therefore, the user can view the illustrated range and easily know words that do not have distinctive features.

In the word list window 113, a word list in which words located near the origin (words in the circle 115) and distances between the words and the origin are arranged in order from shortest to farthest. The upward triangles in the word list window 113 indicate that they are arranged in order from shortest to farthest. In the enlightenment window 114, the title of "points of analysis" is added to record the first knowledge. The inspiration window 114 is arranged at a position overlapping with the scatter diagram window 112.

The size of the circle 115 is determined by any method. For example, the size of the circle 115 can be determined by the user specifying the number of words contained in the circle 115 (for example, 10). Alternatively, the size of the circle 115 may be determined by the user specifying the proportion of words contained in the circle 115 (for example, 10% of the whole). Alternatively, the user may use the mouse 29 to specify the distance from the origin in the first support screen 110 to determine the size of the circle 115.

In the first support screen 110 shown in FIG. 7, words near the origin (words in the circle 115) are displayed in the same form as other words. Alternatively, the words near the origin may be displayed in a form different from other words (for example, in a light color) on the first support screen, or the words near the origin may not be displayed. From the second support screen to the fourth support screen, the words displayed in the form different from other words on the first support screen may be displayed in a form different from other words, or may not be displayed on the first support screen. The word displayed.

8 is a diagram showing a second support screen. The second support screen 120 shown in FIG. 8 includes a screen selection window 101, a scatter diagram window 122, a word list window 123, and an enlightenment window 124. The second support screen 120 is related to the second knowledge "words located in the direction away from the origin to the variable have a high degree of correlation with the variable, and the feature is given to the variable". The user can view the second support screen 120 and use the second knowledge to efficiently perform the process of guiding insights from the scatter diagram.

Before displaying the second support screen 120, the user operates the keyboard 28 or the mouse 29 to select one variable (chapter). Here, the case where the variable "preamble" is selected will be described. The scatter diagram shown in FIG. 3 is described in the scatter diagram window 122. In the scatter diagram in the scatter diagram window 122, an arrow 125 starting from the origin and passing through the selected variable is described, and a predetermined angle (eg 10°) is formed between the arrow 125 and the origin as the starting point Two semi-straight lines 126 and 127 at the angle. In the area enclosed by the semi-straight line 126 and the semi-straight line 127, there is a word located in a direction away from the origin to the selected variable. In this manner, in the scatter diagram included in the second support screen 120, the range from the origin to the direction in which the selected variable deviates is illustrated by the semi-straight line 126 and the semi-straight line 127. Therefore, the user can view the illustrated range and easily know the word that characterizes the selected variable.

In the word list window 123, words (words in the area enclosed by the semi-straight line 126 and the semi-straight line 127) located in the direction away from the origin to the selected variable, and the words and the origin are described The distance is in the order of distance from near to near. The downward triangles in the word list window 123 indicate that they are arranged in order from farthest to closest. In the word list window 123, the second knowledge is associated, and it is described that "the farther the distance from the origin can be determined, the higher the degree of association". In the enlightenment window 124, the title of "analysis points" is added to record the second knowledge. The inspiration window 124 is arranged at a position overlapping the scatter diagram window 122.

As long as the arrow 125 and the half line 126 and the half line 127 are included in the same quadrant, the angle formed by the arrow 125 and the half line 126 and the half line 127 can be determined by any method. When the arrow 125 and the angle are provided to describe the half line 126 and the half line 127, when the half line 126 and the half line 127 are included in a quadrant different from the arrow 125, the half line 126 and the half line 127 are described in the first On the component axis or the second component axis. The arrow 125 is preferably written in a different color (for example, red) from the scatter diagram. The half-line 126 and the half-line 127 are preferably described in a different color (for example, blue) from the scatter diagram and arrow 125.

9 is a diagram showing a third support screen. The third support screen 130 shown in FIG. 9 includes a screen selection window 101, a scatter diagram window 132, a word list window 133, and an inspiration window 134. The third support screen 130 is related to the third knowledge “words that are close to each other have a high degree of similarity”. The user can view the third support screen 130 and use the third knowledge to efficiently perform the process of guiding insights from the scatter diagram.

Before displaying the third support screen 130, the user operates the keyboard 28 or the mouse 29 to select one word, and designates a range around the determined word. Here, the case where the word "eye" is selected will be described. The scatter diagram shown in FIG. 3 is described in the scatter diagram window 132. In the scatter diagram in the scatter diagram window 132, a circle 135 (the appearance is an ellipse) indicating the vicinity of the selected word is described. The circle 135 is preferably written in a different color (for example, red) from the scatter diagram. In this way, in the scatter diagram included in the third support screen 130, the range around the selected word is illustrated by a circle 135. Therefore, the user can view the illustrated range and easily know the word having a high degree of similarity to the selected word.

In the word list window 133, a word list in which words located in the vicinity of the selected word (words in the circle 135) and the distance between the word and the specified word are arranged in order from the shortest to the farthest is described. In the word list window 133, "the closer the distance to the word can be determined, the higher the similarity" is described as the third knowledge. In this example, the word closest to the selected word "eye" is "face". Therefore, the word with the highest similarity to the selected word "eye" is "face". In the enlightenment window 134, the title of "points of analysis" is added to describe the meaning. The inspiration window 134 is arranged at a position overlapping with the scatter diagram window 132.

The size of the circle 135 is determined by an arbitrary method in the same way as the size of the circle 115 in the first support screen 110. For example, the user determines the size of the circle 135 by a method of specifying the number of words contained in the circle 135, a method of specifying the ratio of the words contained in the circle 135, and a method of specifying the distance from the selected word.

10 is a diagram showing a fourth support screen. The fourth support screen 140 shown in FIG. 10 includes a screen selection window 101, a scatter diagram window 142, a variable list window 143, and an inspiration window 144. The fourth support screen 140 is related to the fourth knowledge “the variables with close distances have a high similarity”. The user can view the fourth support screen 140 and use the fourth knowledge to efficiently perform the process of guiding insights from the scatter diagram.

Before displaying the fourth support screen 140, the user operates the keyboard 28 or the mouse 29 to select one variable. Here, the case where the variable "preamble" is selected will be described. The scatter diagram shown in FIG. 3 is described in the scatter diagram window 142. In the scatter diagram in the scatter diagram window 142, an arrow 145 with the selected variable as the starting point and the variable closest to the selected variable as the end point is described. The arrow 145 is preferably written in a different color (for example, red) from the scatter diagram. In this manner, in the scatter diagram included in the fourth support screen 140, an arrow 145 indicating the variable closest to the selected variable is shown. Therefore, the user can view the illustrated arrow 145 and easily know the variable with the highest similarity to the selected variable.

The variable list window 143 describes variables that are closer in distance to the selected variable and a list of variables in which the distances are arranged in order from the closest to the farthest. In the variable list window 143, "the closer the distance to the variable can be determined, the higher the similarity" is described as the fourth knowledge. In this example, the variable closest to the selected variable "Preamble" is the "Postscript". Therefore, the variable with the highest similarity to the selected variable "Preamble" is the "Postscript". In the enlightenment window 144, the title of "points of analysis" is added to describe the meaning. The inspiration window 144 is arranged at a position overlapping with the scatter diagram window 142.

In addition, the text mining support device 10 may display a support screen other than the support screen described above. The support screen may include any content as long as it includes a scatter diagram and a revelation showing the views of the scatter diagram. The revelation can be an observer who expresses the scatter diagram explicitly or an observer who implies the scatter diagram. The revelation can also be included in any part of the support screen. The inspiration can be written in a window that overlaps the scatter diagram window, or in a window that does not overlap the scatter diagram window, or in a message box whose position is fixed.

As described above, the text mining support method of the present embodiment includes the steps of inputting the analysis result 2; the step of inputting the instruction from the user; and the step of generating the screen data including the graph (scatter diagram) representing the analysis result 2 ; And the steps to display the screen based on the screen data. The step of generating screen data corresponds to an instruction, and generates screen data including a graph and a support screen that provides a revelation showing the view of the graph. Therefore, the user can use the support screen including the graph showing the result of the correspondence analysis and the inspiration showing the view of the graph to efficiently perform the process of guiding the insights from the graph showing the result of the correspondence analysis.

The step of generating screen data is generated from a plurality of support screens (first support screen 110, second support screen 120, third support screen 130, and fourth support screen 140) and a basic screen 100 including charts and without revelation, Corresponds to the screen data indicating the selected screen. In this way, the support screen including the inspiration and the basic screen without the inspiration are selectively displayed, whereby the screen corresponding to the level of the user can be displayed. In addition, by selectively displaying multiple support screens, users can be presented with a variety of chart views.

In the step of inputting the analysis result, enter the result of correlating the first item (word) and the second item (variable), that is, the first component and the second component of the first item and the first component of the second item and The result of the second component is the analysis result 2; the step of generating the screen data is to create a scatter plot of the first and second items in a plane with the first component as the horizontal axis and the second component as the vertical axis. As a chart. Therefore, the user can efficiently perform the process of leading insights from the scatter diagram showing the results of the correspondence analysis related to the first item and the second item.

The multiple support screens include: the first support screen 110, which contains the meaning that the first item near the origin in the scatter diagram does not have a distinctive feature as inspiration; the second support screen 120, which contains the scatter diagram located from the origin to the The first item in the direction of the departure of the second item gives the meaning to the second item as inspiration; the third support screen 130 contains the meaning that the first items that are close to each other in the scatter diagram have high similarity as Enlightenment; and the fourth support screen 140 contains the meaning that the similarities between the second items that are close in the scatter diagram are high as inspiration. Therefore, the user can use the enlightenment contained in each support screen to efficiently perform the process of guiding insights from the graph showing the results of the corresponding analysis.

In the scatter diagram included in the first support screen 110, the range around the origin is illustrated with a circle 115. In the scatter diagram included in the second support screen 120, the range from the origin to the direction in which the selected second item departs is shown using half-line 126 and half-line 127. In the scatter diagram included in the third support screen 130, the range around the selected first item is illustrated with a circle 135. In the scatter diagram included in the fourth support screen 140, a symbol (arrow 145) indicating the second item closest to the selected second item is shown. Therefore, the user can view the range or symbol shown in each support screen to easily know the first item that does not have a distinctive feature, the first item that features the selected second item, and the selected first item The first item with a high similarity of item 1 and the second item with a high similarity to the selected second item.

In the step of inputting the analysis result, enter the result of the corresponding analysis of the compound table that takes the word as the first item, the part of the article as the second item, and the appearance frequency of each word in each part of the article as the data in the table As an analysis result. Therefore, the user can efficiently perform a process of leading insights from a scatter diagram showing the results of correspondence analysis related to words and text parts.

The text mining support device 10 of this embodiment and the text mining support program 31 of this embodiment have the same features as the text mining support method of this embodiment, and achieve the same effects.

In addition, in the above description, the text mining support device 10 displays a scatter diagram that two-dimensionally shows the results of the correspondence analysis. The invention is not limited to this, and the present invention can also be applied to a text mining support method and device that displays a graph (for example, a three-dimensional graph) that shows the results of corresponding analysis in multiple dimensions. In addition, similar to the text mining support method and apparatus that displays a scatter diagram showing the results of the correspondence analysis on a composite table related to text data, it can be configured to display and display correspondence to a composite table related to arbitrary data other than text data Data mining support method and device for graphs of analysis results (scatter graphs or three-dimensional graphs, etc.).

According to the text mining support method and apparatus of the present invention, by displaying a support screen including a graph showing the results of the correspondence analysis and a revelation showing the views of the graph, the user can efficiently perform guidance from the graph showing the results of the correspondence analysis Insight handling.

The present invention has been described in detail above, but the above description is illustrative and not restrictive in all respects. It is understood that many other changes or modifications can be conceived without departing from the scope of the present invention.

1‧‧‧Text data 2‧‧‧Analysis result 5‧‧‧Text analysis device 10‧‧‧Text mining support device 11‧‧‧Analysis result input unit 12‧‧‧Instruction input unit 13‧‧‧Screen generation unit 14 ‧‧‧Analysis result display part 20‧‧‧computer 21‧‧‧ CPU 22‧‧‧ main memory 23‧‧‧ memory part 24‧‧‧ input part 25‧‧‧ display part 26‧‧‧ communication part 27‧‧ ‧Recording media reading section 28‧‧‧Keyboard 29‧‧‧Mouse 30‧‧‧Recording media 31‧‧‧Text mining support program 100‧‧‧Basic screen 101 Screen selection windows 102, 112, 122, 132, 142‧‧‧ scatter diagram window 103‧‧‧ radio buttons 110, 120, 130, 140‧‧‧ support screen 113, 123, 133‧‧‧ word list window 143‧‧‧ variable list window 114, 124, 134, 144‧‧‧ Inspiration window 115, 135‧‧‧ circle 125, 145‧‧‧ arrow 126, 127‧‧‧ semi-straight line S101～S112‧‧‧ steps

FIG. 1 is a block diagram showing the structure of a text mining support device according to an embodiment of the present invention. FIG. 2 is a diagram showing a composite table that is an object of correspondence analysis. FIG. 3 is a diagram showing a scattergram created by the text mining support device shown in FIG. 1. 4 is a block diagram showing the configuration of a computer that functions as the text mining support device shown in FIG. 1. 5 is a flowchart showing the operation of the text mining support device shown in FIG. 1. 6 is a diagram showing a basic screen of the text mining support device shown in FIG. 1. 7 is a diagram showing a first support screen of the text mining support device shown in FIG. 1. 8 is a diagram showing a second support screen of the text mining support device shown in FIG. 1. 9 is a diagram showing a third support screen of the text mining support device shown in FIG. 1. FIG. 10 is a diagram showing a fourth support screen of the text mining support device shown in FIG. 1.

101‧‧‧Screen selection window

112‧‧‧ Scatter diagram window

110‧‧‧Support screen

113‧‧‧ Word List Window

114‧‧‧hint window

115‧‧‧ Yuan

Claims (14)

A text mining support method, which is a text mining support method that displays analysis results obtained by correspondence analysis, is characterized by including: a step of inputting the analysis result; a step of inputting an instruction from a user; The step of screen data of the screen of the result graph; and the step of displaying the screen based on the screen data; and in the step of inputting the analysis result, input the result of correlating the first item with the second item, that is, including The results of the first component and the second component of the first item and the first component and the second component of the second item are used as the analysis result. The step of generating screen data includes generating the graph and the indicator A plurality of support screens of the revelation of the view of the graph, and screen data corresponding to the screen selected by the instruction from the basic screens including the graph and not including the revelation are created in the first The component serves as the horizontal axis, and the scatter diagram obtained by plotting the first item and the second item in a plane using the second component as the vertical axis is used as the graph. The text mining support method as described in item 1 of the patent application scope, wherein the plurality of support screens include a first support screen, and the first support screen contains the first item near the origin in the scatter diagram without significant Characteristic meaning Narration. The text mining support method according to item 2 of the patent application scope, wherein the scatter diagram included in the first support screen shows a range around the origin. The text mining support method as described in item 1 of the scope of the patent application, wherein the plurality of support screens include a second support screen that includes a departure from the origin to the second item in the scatter diagram The first item in the direction means that the second item is characterized as the revelation. The text mining support method according to item 4 of the patent application scope, wherein the scatter diagram included in the second support screen shows the range of the direction of departure from the origin to the selected second item. The text mining support method as described in item 1 of the patent application range, wherein the plurality of support screens include a third support screen, and the third support screen contains the first items that are close in distance in the scatter diagram and have a high similarity to each other As the revelation. The text mining support method according to item 6 of the patent application scope, wherein the scatter diagram included in the third support screen shows a range around the selected first item. The text mining support method as described in item 1 of the patent application range, wherein the plurality of support screens include a fourth support screen, and the fourth support screen contains the second items that are close in the scatter diagram and have a high similarity As the revelation. The text mining support method according to item 8 of the patent application scope, wherein the scatter diagram included in the fourth support screen is shown with a symbol representing the second item closest to the selected second item . The text mining support method as described in item 1 of the patent application range, wherein in the step of inputting the analysis result, the input pair uses words as the first item and the part of the article as the second item, the article The frequency of occurrence of each word in each part of is used as a result of corresponding analysis of a composite table of data in the table as the analysis result. The text mining support method according to item 1 of the patent application scope, wherein the screen data of the support screen includes a record related to the knowledge about the view of the chart as the inspiration. A text mining support device, which is a text mining support device displaying analysis results obtained by corresponding analysis, is characterized by comprising: an analysis result input part for inputting the analysis result; an instruction input part for input from a user Instructions; a screen generating unit that generates screen data including a screen representing the analysis results; and an analysis result display unit that displays a screen based on the screen data; and in the analysis result input unit, input the first The result of the correspondence between item 1 and item 2, that is, the result including the first component and the second component of the first item and the first component and the second component of the second item is the analysis result, The screen generating unit generates a plurality of support screens including the chart and a revelation indicating the view of the chart, and a basic screen including the chart and not including the revelation, selected in response to the instruction The screen data of the screen is created by plotting the first item and the second item in a plane with the first component as the horizontal axis and the second component as the vertical axis. The chart. The text mining support device as described in item 12 of the patent application range, wherein in the analysis result input section, a pair is selected with the word as the first item and the part of the article as the second item, and the The appearance frequency of each word in each part is used as a result of corresponding analysis of a composite table of data in the table as the analysis result. The text mining support device according to item 12 of the patent application range, wherein the screen data of the support screen includes a record related to the knowledge about the view of the chart as the inspiration.

Images