KR20190110435A - Text mining method, text mining program and text mining apparatus - Google Patents

Abstract

Provided is a text mining method, which comprises: a step of extracting a word from text data consisting of a statement having a date; performing hierarchical cluster analysis on the extracted word for each analysis period; and displaying a screen including the result of the hierarchical cluster analysis. When an instruction for designating a key word in the first screen including the analysis result is inputted, a second screen including a change over time of the cluster including the key word is displayed by indicating a name of the cluster based on the word included in the cluster including the key word. Accordingly, the change over time of the hierarchical cluster analysis result may be easily recognized.

Description

Text Mining Methods, Text Mining Programs, and Text Mining Devices {TEXT MINING METHOD, TEXT MINING PROGRAM AND TEXT MINING APPARATUS}

The present invention relates to text mining, and more particularly, to a text mining method, a text mining program, and a text mining device for displaying a screen including a result of a hierarchical cluster analysis.

Recently, text mining has been attracting attention for analyzing freely described text data and obtaining useful information from the analysis results. In text mining, for example, information is obtained by extracting a word from text data to be analyzed and analyzing the frequency of occurrence of the word, the tendency of the word, and the like.

Hereinafter, a text mining apparatus for performing a hierarchical cluster analysis on words extracted from text data and displaying a screen including the results will be considered. In hierarchical cluster analysis, clusters containing words with high similarity are hierarchically created based on the similarity between words. In general, the results of the hierarchical cluster analysis are provided to the analyst using the tree diagram (dendogram) shown in FIG. 10. The analyst can grasp the outline of the text data based on the results of the hierarchical cluster analysis.

Japanese Unexamined Patent Application Publication No. 2018-18118 describes a text mining apparatus that displays the results of hierarchical cluster analysis in an embodiment shown in FIG. 11. Given that the number of clusters m and the maximum number of display data n within a cluster are given, the text mining apparatus described in this document obtains m clusters from the results of the hierarchical cluster analysis, and displays the m clusters on the screen in a cloud shape. N or fewer words are displayed inside each cluster.

Among the text data, there is a text having a date, such as a maintenance work record or a call reception record of a call center, and accumulates cumulatively over a long period of time. When performing the hierarchical cluster analysis on such text data, the text data is divided by month, for example, and the hierarchical cluster analysis is performed on the text data of each month. Thus, the results of the hierarchical cluster analysis can be obtained monthly.

In this case, the analyst selects a word to be noted from the text data (hereinafter referred to as a keyword), the cluster containing the keyword in the month, the time when the cluster containing the keyword changes, and the appearance of the keyword. I want to know how the frequency changes over time. However, in the conventional text mining apparatus, the user cannot easily recognize the change over time of the hierarchical cluster analysis result.

Therefore, an object of the present invention is to provide a text mining method, a text mining program, and a text mining device in which a user can easily recognize a change over time of a hierarchical cluster analysis result.

According to a first aspect of the present invention, there is provided a text mining method for displaying a screen including analysis results of text data.

Extracting a word from text data consisting of a statement having a date;

Performing a hierarchical cluster analysis on the word for each analysis period;

And displaying a screen including a result of the hierarchical cluster analysis.

When an instruction for designating a key word in the first screen including the result is input, the step of displaying the screen may include displaying a second screen indicating a change over time of a cluster including the key word. It is characterized by.

The second aspect of the present invention is, in the first aspect of the present invention,

The second screen is characterized by displaying a cluster name based on a word included in the cluster along a time axis.

The third aspect of the present invention is, in the second aspect of the present invention,

The cluster name is characterized by concatenating the words included in the cluster by a predetermined number or less in the order of appearance frequency.

A fourth aspect of the present invention, in the second aspect of the present invention,

The second screen may further include a mark having an aspect corresponding to the degree of change of the cluster name at a position corresponding to the time when the cluster name changes.

A fifth aspect of the present invention is the fourth aspect of the present invention,

The mark is an arrow having a color in accordance with the degree of change of the cluster name.

A sixth aspect of the present invention is the second aspect of the present invention,

The word changed from the previous cluster name among the words constituting the cluster name is highlighted in the second screen.

A seventh aspect of the present invention, in the second aspect of the present invention,

The second screen may further include a graph indicating a change over time of the frequency of appearance of the main fish along the time axis.

The eighth aspect of the present invention is the seventh aspect of the present invention,

The second screen further includes a boundary line at a position corresponding to a time when the cluster name changes, and the background of the graph has a different aspect for each boundary line.

A ninth aspect of the present invention is the second aspect of the present invention,

In the case where the cluster name is often greatly changed, the step of displaying the screen is characterized by displaying a screen including a warning message.

A tenth aspect of the present invention is a text mining program for displaying a screen including a result of analysis of text data,

Extracting a word from text data consisting of a statement having a date;

Performing a hierarchical cluster analysis on the word for each analysis period;

The computer executes the step of displaying a screen including the result of the hierarchical cluster analysis in a computer using a memory,

When an instruction for designating a key word in the first screen including the result is input, the step of displaying the screen may include displaying a second screen indicating a change over time of a cluster including the key word. It is characterized by.

An eleventh aspect of the present invention is the tenth aspect of the present invention,

The second screen is characterized by displaying a cluster name based on a word included in the cluster along a time axis.

A twelfth aspect of the present invention is the eleventh aspect of the present invention,

The cluster name is characterized by concatenating the words included in the cluster by a predetermined number or less in the order of appearance frequency.

A thirteenth aspect of the present invention is the eleventh aspect of the present invention,

The second screen may further include a mark having an aspect corresponding to the degree of change of the cluster name at a position corresponding to the time when the cluster name changes.

A fourteenth aspect of the present invention is the thirteenth aspect of the present invention,

The mark is an arrow having a color in accordance with the degree of change of the cluster name.

A fifteenth aspect of the present invention is the eleventh aspect of the present invention,

The word changed from the previous cluster name among the words constituting the cluster name is highlighted in the second screen.

A sixteenth aspect of the present invention is the eleventh aspect of the present invention,

The second screen may further include a graph indicating a change over time of the frequency of appearance of the main fish along the time axis.

A seventeenth aspect of the present invention is the sixteenth aspect of the present invention,

The second screen further includes a boundary line at a position corresponding to a time when the cluster name changes, and the background of the graph has a different aspect for each boundary line.

An eighteenth aspect of the present invention is the eleventh aspect of the present invention,

In the case where the cluster name is often greatly changed, the step of displaying the screen is characterized by displaying a screen including a warning message.

A nineteenth aspect of the present invention is a text mining device that displays a screen including a result of analysis of text data,

A word extracting unit for extracting a word from text data consisting of a sentence having a date;

A clustering processor configured to perform hierarchical cluster analysis on the word for each analysis period;

A screen display unit which displays a screen including a result of the hierarchical cluster analysis,

When an instruction for designating a key word in the first screen including the result is input, the screen display unit displays a second screen indicating a change over time of a cluster including the key word. do.

A twentieth aspect of the present invention is the nineteenth aspect of the present invention,

The second screen is characterized by displaying a cluster name based on a word included in the cluster along a time axis.

According to the first, tenth, or nineteenth aspect, when an instruction for designating a target word is input in a first screen including a result of a hierarchical cluster analysis, a change over time of a cluster including the target word is performed. By displaying a second screen indicating, the user can easily recognize a change over time of the hierarchical cluster analysis result.

According to the second, eleventh, or twentieth aspect, a cluster name based on a word included in a cluster containing a key word is represented along a time axis, whereby the user can change the time course of the cluster including the key word. It can be easily recognized.

According to the third or twelfth aspect, a cluster name in which a word with a high frequency of occurrence is displayed along the time axis according to the third or twelfth aspect, so that the user can change the time course of the cluster including the keyword. It can be easily recognized.

According to the fourth, fifth, thirteenth or fourteenth aspect, a second screen including a mark (arrow having a color according to the degree of change) having an aspect according to the degree of change of the cluster name including a note By displaying, the user can easily recognize the degree of change of the cluster including the keyword.

According to the sixth or fifteenth aspect, by highlighting the changed words among the words constituting the name of the cluster containing the key word, the user can see how the word with high frequency in the cluster containing the key word has changed. It can be easily recognized.

According to the seventh or sixteenth aspect, in addition to the change over time of the cluster containing the main word, the user may display a screen including a graph showing a change over time of the frequency of occurrence of the main word. The change over time of the hierarchical cluster analysis result can be easily recognized.

According to the eighth or seventeenth aspect, a boundary line is displayed at a position corresponding to a time when a name of a cluster including a core change, and the aspect of the graph background is changed for each boundary so that the user includes the core. You can easily recognize when the cluster is changing.

According to the ninth or eighteenth aspect, when a name of a cluster including a notice word is largely changed, a screen including a warning message is displayed, whereby the user can recognize that the hierarchical cluster analysis is not well performed. Can be.

1 is a block diagram showing a configuration of a text mining apparatus according to an embodiment of the present invention.
FIG. 2 is a block diagram showing the configuration of a computer operating as a text mining apparatus shown in FIG. 1.
FIG. 3 is a flowchart showing the operation of the text mining apparatus shown in FIG. 1.
FIG. 4 is a diagram illustrating an example of a window indicating a result of hierarchical cluster analysis displayed by the text mining apparatus illustrated in FIG. 1.
FIG. 5 is a diagram illustrating an operation for specifying a target word in the window shown in FIG. 4.
FIG. 6 is a diagram illustrating an example of a window showing a change over time of an analysis result displayed by the text mining device illustrated in FIG. 1.
FIG. 7 is a diagram illustrating an example of a display screen of the text mining apparatus illustrated in FIG. 1.
8A illustrates an example of change over time of a hierarchical cluster analysis result.
FIG. 8B is a view following FIG. 8A.
8C is a view following FIG. 8B.
FIG. 8D is a view following FIG. 8C.
FIG. 9 is a diagram illustrating a window displayed by the text mining device illustrated in FIG. 1.
10 is a diagram illustrating an example of a tree.
It is a figure which shows the display aspect of the hierarchical cluster analysis result in the conventional text mining apparatus.

EMBODIMENT OF THE INVENTION Hereinafter, with reference to drawings, the text mining method, the text mining program, and the text mining apparatus which concern on embodiment of this invention are demonstrated. The text mining method according to the present embodiment is typically executed using a computer. The text mining program according to the present embodiment is a program for implementing a text mining method using a computer. The text mining device according to the present embodiment is typically configured by using a computer. The computer executing the text mining program functions as a text mining device.

1 is a block diagram showing the configuration of a text mining apparatus according to an embodiment of the present invention. The text mining device 10 shown in FIG. 1 includes an instruction input unit 11, a text data storage unit 12, a word extraction unit 13, a clustering processing unit 14, an analysis result storage unit 15, and a screen display unit. (16) is provided. The text mining device 10 performs a hierarchical cluster analysis on the text data stored in the text data storage unit 12, and displays a screen including the analysis result.

The outline of the operation of the text mining device 10 is as follows. Instruction input unit 11 receives an instruction from a user (analyzer of text data). The text data storage unit 12 stores one or more freely described text data. The word extraction unit 13 extracts a word from the text data by performing a morpheme analysis on the text data stored in the text data storage unit 12. The clustering processing unit 14 performs hierarchical cluster analysis on the words extracted by the word extraction unit 13. The analysis result storage unit 15 stores the analysis result by the clustering processing unit 14. The screen display unit 16 displays screen data based on the analysis result stored in the analysis result storage unit 15.

The text data storage unit 12 is composed of a door having a date, and stores text data accumulated cumulatively over a long period of time (for example, several years). The user inputs, using the instruction | indication input part 11, the instruction | indication which designates the text data which is an analysis object, an analysis period, an analysis interval, an instruction to designate a keyword, etc. The word extracting unit 13, the clustering processing unit 14, and the screen display unit 16 perform an operation for displaying a screen including a result of performing hierarchical cluster analysis on the text data according to the instruction from the user. do. In addition, the screen display unit 16 displays a screen including a change over time of the hierarchical cluster analysis result in accordance with an instruction from the user.

2 is a block diagram showing the configuration of a computer functioning as the text mining device 10. The computer 20 shown in FIG. 2 includes a CPU 21, a main memory 22, a storage unit 23, an input unit 24, a display unit 25, a communication unit 26, and a recording medium reading unit 27. Equipped with. For example, a DRAM is used for the main memory 22. For example, a hard disk or a solid state drive is used for the storage unit 23. The input unit 24 includes, for example, a keyboard 28 and a mouse 29. The liquid crystal display is used for the display part 25, for example. The communication unit 26 is an interface circuit of wired communication or wireless communication. The recording medium reading unit 27 is an interface circuit of the recording medium 30 that stores a program or the like. As the recording medium 30, a non-transitory recording medium such as a CD-ROM, a DVD-ROM, or a USB memory is used.

When the computer 20 executes the text mining program 31, the storage unit 23 stores the text mining program 31 and the text data 32. The text mining program 31 and the text data 32 may be received using, for example, the communication unit 26 from a server or another computer, and the recording medium reading unit 27 is read from the recording medium 30. It may be used as read out.

When the text mining program 31 is executed, the text mining program 31 and the text data 32 are copied and transferred to the main memory 22. The CPU 21 executes the text mining program 31 stored in the main memory 22 using the main memory 22 as the working memory, thereby processing to extract a word from the text data 32, and the extracted word. Processing for performing hierarchical cluster analysis, displaying a screen including the analysis result, and the like. At this time, the computer 20 functions as the text mining device 10. In addition, the structure of the computer 20 mentioned above is only an example, and the text mining apparatus 10 can be comprised using arbitrary computers.

3 is a flowchart showing the operation of the text mining device 10. Before performing the operation shown in FIG. 3, the text data storage unit 12 freely describes and stores one or more text data accumulated cumulatively. The text data consists of a statement having a date (for example, a work day or a reception date, etc.), and the text data is divided into a plurality of parts by the date. The text mining device 10 processes the text data specified by the user among the text data stored in the text data storage unit 12.

In FIG. 3, the instruction | indication input part 11 receives the instruction which designates the text data, an analysis period, and an analysis interval which are analysis objects from a user first (step S101). The user inputs these information into the dialog box (not shown) displayed on the screen using the input part 24. The received instruction is output to each part of the text mining apparatus 10.

Next, the word extraction unit 13 reads out the specified text data from the text data storage unit 12 (step S102). Next, the word extraction part 13 extracts a word from the text data read-out by performing morphological analysis with respect to the text data read-out in step S102 (step S103). At this time, the word extracting unit 13 extracts only words necessary for subsequent analysis from the text data read out.

Next, the clustering processing unit 14 performs hierarchical cluster analysis on the words extracted in step S103 (step S104). Next, the clustering processing unit 14 calculates the frequency of appearance of the words extracted in step S103 (step S105). Next, the analysis result storage part 15 memorize | stores the result of the hierarchical cluster analysis calculated | required in step S104, and the frequency of appearance of the word calculated | required in step S105 (step S106).

The clustering processing unit 14 receives the analysis period and analysis interval specified by the user from the instruction input unit 11. The analysis period represents a period during which hierarchical cluster analysis is actually performed among the text data to be analyzed. The analysis period is divided into a plurality of periods based on the analysis interval. For example, if the analysis period is from June 1, 2005 to May 31, 2015, and the analysis interval is one month, the analysis period of 11 years is divided into 132 periods.

The number of periods after the division is set to p. In step S104, the clustering processing unit 14 performs hierarchical cluster analysis for each of the p periods. More specifically, the clustering processing unit 14 analyzes the hierarchical cluster using a statement having a date and time within the period of the text data read out and output in step S102 for the words extracted in step S103 for each of the p periods. Is carried out. The clustering processing unit 14 calculates the similarity between the two words based on, for example, the distance between the two words in the text data 32 (how far apart the two words appear). The clustering processing unit 14 performs hierarchical cluster analysis using a predetermined method (for example, the shortest distance method, the longest distance method, the group average method, the decimal method, the word method, and the like) based on the similarity between the obtained words. .

In step S105, the clustering processing unit 14 calculates the frequency of appearance of words for each of the p periods. In step S104, p hierarchical cluster analysis results are obtained, and in step S105, the frequency of occurrence of words is determined by p. In step S106, the analysis result storage unit 15 stores the results of the hierarchical cluster analysis and the frequency of occurrence of words for each of the p periods.

Next, the screen display unit 16 displays a screen including the results of the hierarchical cluster analysis stored in the analysis result storage unit 15 (step S107). 4 is a diagram illustrating an example of a window displayed in step S107. The window 41 shown in FIG. 4 contains the results of the hierarchical cluster analysis. When the number of clusters is set for the results of the hierarchical cluster analysis, words included in each cluster are determined. When the text mining device 10 displays a screen including the results of the hierarchical cluster analysis, the text mining device 10 displays a plurality of clusters in the form shown in FIG. 4 instead of the tree diagram.

The text mining device 10 has, as an operation parameter, the number of clusters and the maximum number of display data in the cluster. Hereinafter, the former is m and the latter is n. These values are set to predetermined initial values in an initial state. The user may set these values arbitrarily using the instruction | indication input part 11. In the text mining device 10, the words extracted in step S103 are classified into m clusters. Each cluster contains one or more words. In the window 41, m clusters are displayed in a cloud shape, and words contained in each cluster are displayed inside each cluster. The number of words displayed inside each cluster is limited to n or less. For example, when a cluster includes ten words when n = 5, five words are displayed inside the cluster displayed on the screen.

Next, the instruction input unit 11 receives an instruction from the user (step S111). Next, the text mining apparatus 10 determines whether or not the instruction received in step S111 is an instruction for designating a key word (step S112). If yes, control of the text mining device 10 proceeds to step S121, and if no, the control proceeds to step S113.

In the latter case, the instruction received in step S111 is, for example, an instruction to move the window, an instruction to hide the window, an instruction to close the window, or the like. The screen display unit 16 displays the screen after the update in accordance with the instruction received in step S111 (step S113). Thereafter, the control of the text mining device 10 proceeds to step S111.

When executing step S111, the screen containing the result of a hierarchical cluster analysis is displayed. Hereinafter, when performing step S111, it is assumed that the screen including the window 41 shown in FIG. 4 is displayed. Clicking on the button of the mouse 29 when the mouse cursor 43 is on an element in the display screen is referred to as "clicking on the element", and the cluster containing the main word is called "major key cluster", The name given to the main cluster is called the main cluster name.

5 is a diagram illustrating an operation for designating a target word. The user clicks on the word (here, "disassembly") that is designated as the key word in the window 41 (first click). At this time, the context menu 42 appears in the display screen. The user clicks on the item "with change over time of the analysis result" in the context menu 42 (second click). By this operation, the word clicked on the first time is designated as the key word.

In the case of Yes in step S112, the screen display unit 16 reads out the result of the hierarchical cluster analysis from the analysis result storage unit 15 and the frequency of appearance of the key word (step S121). Next, the screen display unit 16 displays a screen including a change over time of the hierarchical cluster analysis result based on the read-out data (step S122).

FIG. 6 is a diagram showing a window showing a change over time of the analysis result displayed in step S122. FIG. The window 51 shown in FIG. 6 is displayed when "decomposition" is designated as a key word in step S111. For example, as shown in FIG. 7, the window 51 overlaps the window 41 shown in FIG. 4.

The window 51 includes a line graph 52 representing a change over time of the appearance frequency of the main fish along the time axis extending in the horizontal direction. As the frequency of occurrences of the main word, for example, a ratio of the number of occurrences of the main word to the total number of occurrences of all the words included in the main word cluster is used. The frequency of appearance of the major word may be changed to the number of occurrences of the major word in accordance with an instruction from the user.

The configuration (element included in the cluster) of the cluster obtained by the hierarchical cluster analysis changes over time. Clusters are automatically named to indicate changes over time in the cluster. If the cluster contains only one word, that word is used as it is in the cluster name. In the case where the cluster includes two words, the cluster name is a combination of two words in order of appearance. In the case where the cluster includes three or more words, the cluster name is used by concatenating three words having the highest frequency of appearance among the words included in the cluster in the order of appearance frequency. In addition, when the set of words constituting the cluster name is the same, even if the order of the words is different, they are treated as the same cluster name.

8A to 8D are diagrams showing examples of changes over time of the hierarchical cluster analysis results. 8A-8D, the result of the hierarchical cluster analysis in another month is described. In Figs. 8A to 8D, the cloud figures represent clusters, and the underlined strings represent cluster names. The size of the circle indicates the frequency of appearance of the words described in the circle.

In the analysis result shown in FIG. 8A, the words extracted from the text data include a cluster including "drive" and "decomposition", a cluster including "exhaust", "pressure", "flow" and "valve", and " The belt is classified into clusters including "rotation", "check", "motor" and "tension". These three clusters are named "driving, disassembling," "exhaust, pressure, flow," and "belt, rotation, and check." Also about the analysis result shown to FIG. 8B-FIG. 8D, three clusters are named by the same method.

When "decomposition" is designated as the key word, the key word cluster name is "decomposition and drive" in the analysis result shown in FIG. 8A, "drive belt rotation" in the analysis result shown in FIG. 8B, and shown in FIG. 8C. In the analysis result, it is "exhaust pressure / flow", and in the analysis result shown in FIG. 8D, it is "exhaust pressure / decomposition". In this way, the key cluster name changes over time.

The window 51 shown in FIG. 6 includes the main cluster name 53, the boundary line 54, and the arrow 55 in addition to the line graph 52. The key cluster name 53 is displayed above the broken line graph 52 along the time axis extending in the horizontal direction. The boundary line 54 is displayed in the line graph 52 at the position corresponding to the time when the main cluster name 53 changes. The key cluster name 53 is displayed for each period divided by the boundary line 54. The background of the broken line graph 52 has a different aspect (for example, a different color or a different pattern) for each boundary line 54. Among the words constituting the main cluster name 53, a word changed from the previous cluster name (a word not included in the old main cluster name and included in the new main cluster name) is Is highlighted. In window 51, such words are shown in Gothic and in italics.

An arrow 55 is displayed at the position corresponding to the timing at which the main cluster cluster name 53 changes at the upper portion of the boundary line 54. An arrow 55 is displayed in the form according to the degree of change of the main cluster name 53. When all of the words constituting the main cluster name 53 change, a red arrow 55r is displayed. In the case where two words constituting the main cluster name 53 change, a blue arrow 55b is displayed. In the case where one word constituting the main cluster name 53 changes, a black arrow 55n is displayed. The display mode of the arrow 55 may be arbitrary as long as it varies depending on the degree of change in the target cluster name 53. For example, the display size of the arrow 55 may differ depending on the degree of change of the main cluster name 53.

In the example shown in FIG. 6, the main cluster name 53 is time in order of "drive and decomposition", "drive belt rotation", "exhaust pressure and flow", and "exhaust pressure and decomposition". Changes over time In the first change, two words constituting the main cluster name 53 change, and a blue arrow 55b is displayed on the first boundary line 54. In the second change, all of the words constituting the target cluster name 53 change, so that the red arrow 55r is displayed on the second boundary line 54. In the third change, one word constituting the main cluster name 53 changes, and therefore, a black arrow 55n is displayed on the third boundary line 54.

Next, the screen display unit 16 obtains the number of arrows 55 included in the screen displayed in step S122 for each type (step S123). Next, the screen display unit 16 determines whether the change in the target cluster name 53 is large based on the number of arrows 55 of each type (step S124). The screen display unit 16 may determine, for example, when the number of the red arrows 55r exceeds 30% of the total number of the arrows 55, and the number of the red arrows 55r and the blue arrows ( When the sum total of the number of 55b) exceeds 60% of the total number of the arrow 55, you may judge yes. If yes, control of the text mining device 10 proceeds to step S125, and if no, the control proceeds to step S111.

In the former case, the screen display unit 16 displays a screen including a warning message (step S125). 9 is a diagram illustrating a window displayed in step S125. Since the window 61 shown in FIG. 9 often changes the structure of a target cluster largely, the warning message which recommends to readjust the setting of hierarchical cluster analysis (for example, number of clusters or number of target words). It includes. Thereafter, the control of the text mining device 10 proceeds to step S111.

As described above, the text mining method according to the present embodiment performs steps (steps S102 and S103) of extracting words from text data consisting of a sentence having a date, and performs hierarchical cluster analysis on the extracted words for each analysis period. Steps (step S104) to perform and steps (step S107, S113, S121 to S125) which display the screen containing the analysis result by hierarchical cluster analysis are provided. When an instruction for designating a key word is input in the first screen (screen including window 41) containing the analysis result (FIG. 5), the step of displaying the screen (step S122) includes the key word. A second screen (screen including window 51) indicating a change over time of the cluster is displayed. According to the text mining method according to the present embodiment, when an instruction for designating a key word in the first screen including the result of the hierarchical cluster analysis is input, it indicates a change over time of the cluster including the key word. By displaying the second screen, the user can easily recognize the change over time of the hierarchical cluster analysis result.

In addition, the second screen displays a cluster name (the main cluster name 53) based on the words included in the cluster including the main word along the time axis. In addition, this cluster name connects the words contained in the cluster containing a key word by the predetermined number or less (three or less) in the order of high frequency of appearance. Thus, the user can easily recognize the change over time of the cluster containing the key word.

In addition, the second screen includes a mark having an aspect corresponding to the degree of change of the cluster name at a position corresponding to the time when the name of the cluster including the main word changes. This mark may be an arrow 55 having a color corresponding to the degree of change of the cluster name. By displaying the second screen including such a mark (arrow 55), the user can easily recognize the degree of change of the cluster name including the keyword. Moreover, the word ("belt", "rotation", etc. which are shown in FIG. 6) which changed from the previous cluster name among the words which comprise a cluster name is highlighted and displayed in a 2nd screen. Therefore, the user can easily recognize how the word with high appearance frequency has changed in the cluster including the main word.

The second screen also includes a graph (line graph 52) showing a change over time of the frequency of appearance of the main fish along the time axis. In addition to the change over time of the cluster containing the main fish, by displaying a screen including a graph showing a change over time of the frequency of occurrence of the main fish, the user can change the time course of the hierarchical cluster analysis results. Changes can be easily recognized. In addition, the second screen includes a boundary line 54 at a position corresponding to the time when the name of the cluster including the main word changes, and the background of the graph has a different aspect for each boundary line. Thus, the user can easily recognize when the cluster containing the key word changes. In addition, when the name of the cluster including the main word is largely changed, the step of displaying the screen displays a screen (screen including window 61) including a warning message. Thus, the user can recognize that hierarchical cluster analysis is not doing well.

The text mining device 10 and the text mining program 31 according to the present embodiment have the same characteristics as the above-described text mining method and exhibit the same effects. According to the text mining method, the text mining apparatus 10, and the text mining program 31 according to the present embodiment, the user can easily recognize the change over time of the hierarchical cluster analysis result.

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

10: text mining device
11: instruction input unit
12: text data storage unit
13: word extraction unit
14: clustering processing unit
15: analysis result storage unit
16: screen display unit
20: computer
21: CPU
22: main memory
29: mouse
30: recording medium
31: Text Mining Program
32: text data
41, 51, 61: windows
42: context menu
43: mouse cursor
52: line graph
53: main cluster name
54: boundary line
55: arrow

Claims (20)

A text mining method for displaying a screen containing analysis results of text data, the method comprising:
Extracting a word from text data consisting of a sentence having a date;
Performing a hierarchical cluster analysis on the word for each analysis period;
And displaying a screen including a result of the hierarchical cluster analysis.
When an instruction for designating a key word in the first screen including the result is input, the step of displaying the screen may include displaying a second screen indicating a change over time of a cluster including the key word. Text mining method, characterized in that. The method of claim 1,
And the second screen displays a cluster name based on a word included in the cluster along the time axis. The method of claim 2,
The cluster name is a text mining method characterized by concatenating a word included in the cluster by a predetermined number or less in the order of appearance frequency. The method of claim 2,
And the second screen further includes a mark having an aspect corresponding to a degree of change of the cluster name at a position corresponding to a time when the cluster name changes. The method of claim 4, wherein
And the mark is an arrow having a color in accordance with the degree of change of the cluster name. The method of claim 2,
The word changed from the previous cluster name among the words constituting the cluster name is highlighted in the second screen. The method of claim 2,
And the second screen further includes a graph indicating a change over time of the frequency of appearance of the main fish along the time axis. The method of claim 7, wherein
The second screen further includes a boundary line at a position corresponding to a time when the cluster name changes, and the background of the graph has a different aspect for each boundary line. The method of claim 2,
The text mining method is characterized in that, in the case where the cluster name is greatly changed, the step of displaying the screen displays a screen including a warning message. A text mining program stored in a medium for displaying a screen including a result of analysis of text data, the text mining program comprising:
Extracting a word from text data consisting of a sentence having a date;
Performing a hierarchical cluster analysis on the word for each analysis period;
The computer executes the step of displaying a screen including the result of the hierarchical cluster analysis in a computer using a memory,
When an instruction for designating a key word in the first screen including the result is input, the step of displaying the screen may include displaying a second screen indicating a change over time of a cluster including the key word. And a text mining program stored in the medium. The method of claim 10,
And the second screen displays a cluster name based on a word included in the cluster along a time axis. The method of claim 11,
The cluster name is a text mining program stored in a medium, characterized in that the number of words included in the cluster are linked in the order of high frequency. The method of claim 11,
And the second screen further comprises a mark having an aspect corresponding to the degree of change of the cluster name at a position corresponding to the time when the cluster name changes. The method of claim 13,
And the mark is an arrow having a color in accordance with the degree of change of the cluster name. The method of claim 11,
The word changed from the previous cluster name among the words constituting the cluster name is highlighted in the second screen. The text mining program stored in the medium. The method of claim 11,
And the second screen further comprises a graph indicating a change over time of the frequency of appearance of the main fish along the time axis. The method of claim 16,
The second screen further includes a boundary line at a position corresponding to a time when the cluster name changes, and the background of the graph has a different aspect for each boundary line. The method of claim 11,
And when the cluster name is greatly changed, the step of displaying the screen displays a screen including a warning message. A text mining device for displaying a screen containing analysis results of text data,
A word extracting unit for extracting a word from text data consisting of a sentence having a date;
A clustering processor configured to perform hierarchical cluster analysis on the word for each analysis period;
A screen display unit which displays a screen including a result of the hierarchical cluster analysis,
When an instruction for designating a key word in the first screen including the result is input, the screen display unit displays a second screen indicating a change over time of a cluster including the key word. , Text mining device. The method of claim 19,
And the second screen displays a cluster name based on a word included in the cluster along the time axis.

Images