JP2019164592A - Text mining method, text mining program, and text mining device - Google Patents

Abstract

To easily recognize a secular change of a hierarchical cluster analysis result.SOLUTION: A text mining method includes: a step for extracting a word from text data constituted of sentences having dates; a step for performing a hierarchical cluster analysis for each analysis period for the extracted word; and a step for displaying a screen including a result of the hierarchical cluster analysis. When an instruction for designating an attention word is input in a first screen including the analysis result, the text mining method displays a second screen indicating a secular change of cluster including the attention word by indicating a cluster name on the basis of the word included in the cluster including the attention word along a time axis.SELECTED DRAWING: Figure 3

Description

  The present invention relates to text mining, and more particularly, to a text mining method, a text mining program, and a text mining apparatus that display a screen including a result of hierarchical cluster analysis.

  In recent years, text mining that analyzes free-written text data and obtains useful information from the analysis results has attracted attention. In text mining, for example, a word is extracted from text data to be analyzed, and information is obtained by analyzing the appearance frequency and appearance tendency of the word.

  Hereinafter, a text mining apparatus that performs hierarchical cluster analysis on words extracted from text data and displays a screen including the results will be considered. In the hierarchical cluster analysis, clusters including words with high similarity are created hierarchically based on the similarity between words. In general, the results of the hierarchical cluster analysis are provided to the analyst using the dendrogram shown in FIG. The analyst can grasp the outline of the text data based on the result of the hierarchical cluster analysis.

  Patent Document 1 describes a text mining device that displays the results of hierarchical cluster analysis in the form shown in FIG. The text mining device described in Patent Document 1 obtains m clusters from the result of the hierarchical cluster analysis when the number of clusters m and the maximum number of display data n in the cluster are given. The clusters are displayed on the screen in a cloud shape, and n or less words are displayed inside each cluster.

Japanese Patent Application Laid-Open No. 2018-18118

  Some text data, such as maintenance work records and call center telephone response records, is composed of sentences with dates and accumulated cumulatively over a long period of time. When hierarchical cluster analysis is performed on such text data, the text data is divided into, for example, months, and hierarchical cluster analysis is performed on the text data of each month. Thereby, the result of a hierarchical cluster analysis can be calculated | required for every month.

  In this case, the analyst selects a notable word (hereinafter referred to as the attention word) from the text data, and each month includes a cluster including the attention word, a time when the cluster including the attention word changes, and the appearance of the attention word. I want to know the change in frequency over time. However, in the conventional text mining device, the user cannot easily recognize the temporal change of the result of the hierarchical cluster analysis.

  Therefore, an object of the present invention is to provide a text mining method, a text mining program, and a text mining apparatus that allow a user to easily recognize a temporal change in the result of hierarchical cluster analysis.

A first aspect of the present invention is a text mining method for displaying a screen including an analysis result of text data,
Extracting words from text data consisting of sentences with dates;
Performing a hierarchical cluster analysis on each word for each analysis period;
Displaying a screen containing the results of the hierarchical cluster analysis,
The step of displaying the screen when an instruction for designating the attention word is input in the first screen including the result displays a second screen showing a change over time of the cluster including the attention word. And

According to a second aspect of the present invention, in the first aspect of the present invention,
The second screen shows a cluster name based on a word included in the cluster along a time axis.

According to a third aspect of the present invention, in the second aspect of the present invention,
The cluster name is formed by concatenating a predetermined number or less of words included in the cluster in descending order of appearance frequency.

According to a fourth aspect of the present invention, in the second aspect of the present invention,
The second screen further includes a mark having an aspect corresponding to the degree of change of the cluster name at a position corresponding to a time when the cluster name changes.

According to a fifth aspect of the present invention, in the fourth aspect of the present invention,
The mark is an arrow having a color corresponding to the degree of change of the cluster name.

According to a sixth aspect of the present invention, in the second aspect of the present invention,
Of the words constituting the cluster name, words changed from the previous cluster name are highlighted in the second screen.

According to a seventh aspect of the present invention, in the second aspect of the present invention,
The second screen further includes a graph showing a temporal change in the appearance frequency of the attention word along the time axis.

According to an eighth aspect of the present invention, in 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.

According to a ninth aspect of the present invention, in the second aspect of the present invention,
When the cluster name often changes greatly, the step of displaying the screen displays a screen including a warning message.

A tenth aspect of the present invention is a text mining program for displaying a screen including an analysis result of text data,
Extracting words from text data consisting of sentences with dates;
Performing a hierarchical cluster analysis on each word for each analysis period;
Causing the computer to execute a step of displaying a screen including the result of the hierarchical cluster analysis using a memory,
The step of displaying the screen when an instruction for designating the attention word is input in the first screen including the result displays a second screen showing a change over time of the cluster including the attention word. And

An eleventh aspect of the present invention is the tenth aspect of the present invention,
The second screen shows 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 formed by concatenating a predetermined number or less of words included in the cluster in descending order of appearance frequency.

A thirteenth aspect of the present invention is the eleventh aspect of the present invention,
The second screen further includes a mark having an aspect corresponding to the degree of change of the cluster name at a position corresponding to a 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 corresponding to the degree of change of the cluster name.

A fifteenth aspect of the present invention is the eleventh aspect of the present invention,
Of the words constituting the cluster name, words changed from the previous cluster name are highlighted in the second screen.

A sixteenth aspect of the present invention is the eleventh aspect of the present invention,
The second screen further includes a graph showing a temporal change in the appearance frequency of the attention word 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.

According to an eighteenth aspect of the present invention, in an eleventh aspect of the present invention,
When the cluster name often changes greatly, the step of displaying the screen displays a screen including a warning message.

A nineteenth aspect of the present invention is a text mining device that displays a screen including an analysis result of text data,
A word extraction unit that extracts words from text data including sentences having dates;
A clustering processing unit that performs hierarchical cluster analysis for each analysis period for the word;
A screen display unit for displaying a screen including the result of the hierarchical cluster analysis,
When an instruction for designating a word of interest is input in the first screen including the result, the screen display unit displays a second screen showing a change over time of a cluster including the word of interest. .

According to a twentieth aspect of the present invention, in a nineteenth aspect of the present invention,
The second screen shows 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 word of interest is input in the first screen including the result of the hierarchical cluster analysis, a change with time of the cluster including the word of interest is detected. By displaying the second screen shown, the user can easily recognize the temporal change in the result of the hierarchical cluster analysis.

  According to the second, eleventh, or twentieth aspect, by indicating the cluster name based on the word included in the cluster including the attention word along the time axis, the user can change the cluster including the attention word over time. It can be easily recognized.

  According to the third or twelfth aspect described above, by indicating along the time axis a cluster name in which words having a high appearance frequency are connected in a cluster including the attention word, the user can change the cluster including the attention word over time. Can be easily recognized.

  According to the fourth, fifth, thirteenth or fourteenth aspect, the mark (an arrow having a color corresponding to the degree of change) having a form corresponding to the degree of change of the name of the cluster including the attention word is included. By displaying the second screen, the user can easily recognize the degree of change of the cluster including the attention word.

  According to the sixth or fifteenth aspect, by highlighting the changed word among the words constituting the name of the cluster including the attention word, the user can recognize a word having a high appearance frequency in the cluster including the attention word. It is possible to easily recognize how it has changed.

  According to the seventh or sixteenth aspect, in addition to the time-dependent change of the cluster including the attention word, the user can display the screen including the graph indicating the time-dependent change of the appearance frequency of the attention word, thereby allowing the user to execute the hierarchical cluster. Changes over time in the results of analysis can be easily recognized.

  According to the eighth or seventeenth aspect, the boundary line is displayed at a position corresponding to the time when the name of the cluster including the attention word changes, and the background mode of the graph is switched for each boundary line. Can easily recognize when the cluster containing the attention word changes.

  According to the ninth or eighteenth aspect, when the name of the cluster including the attention word often changes greatly, the user is in a poor hierarchical cluster analysis by displaying a screen including a warning message. Can be recognized.

It is a block diagram which shows the structure of the text mining device which concerns on embodiment of this invention. It is a block diagram which shows the structure of the computer which operate | moves as a text mining apparatus shown in FIG. It is a flowchart which shows operation | movement of the text mining apparatus shown in FIG. It is a figure which shows the example of the window which shows the result of the hierarchical cluster analysis which the text mining apparatus shown in FIG. 1 displays. It is a figure which shows operation which designates an attention word in the window shown in FIG. It is a figure which shows the example of the window which shows the time-dependent change of the analysis result which the text mining apparatus shown in FIG. 1 displays. It is a figure which shows the example of the display screen of the text mining apparatus shown in FIG. It is a figure which shows the example of a time-dependent change of the result of a hierarchical cluster analysis. It is a continuation figure of FIG. 8A. It is a continuation figure of FIG. 8B. It is a continuation figure of FIG. 8C. It is a figure which shows the window which the text mining apparatus shown in FIG. 1 displays. It is a figure which shows the example of a dendrogram. It is a figure which shows the display mode of the result of the hierarchical cluster analysis in the conventional text mining apparatus.

  Hereinafter, a text mining method, a text mining program, and a text mining apparatus according to an embodiment of the present invention will be described with reference to the drawings. 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 using a computer. A computer that executes the text mining program functions as a text mining device.

  FIG. 1 is a block diagram showing a configuration of a text mining apparatus according to an embodiment of the present invention. A 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. 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 apparatus 10 is as follows. An instruction from the user (text data analyst) is input to the instruction input unit 11. The text data storage unit 12 stores one or more freely described text data. The word extraction unit 13 extracts words from the text data by performing morphological analysis on the text data stored in the text data storage unit 12. The clustering processing unit 14 performs a hierarchical cluster analysis on the words extracted by the word extracting unit 13. The analysis result storage unit 15 stores the analysis result obtained 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 sentences having dates, and stores text data accumulated cumulatively over a long period (for example, several years). The user uses the instruction input unit 11 to input an instruction to specify text data to be analyzed, an analysis period and an analysis interval, an instruction to specify a word of interest, and the like. The word extraction unit 13, the clustering processing unit 14, and the screen display unit 16 perform an operation for displaying a screen including the result of the hierarchical cluster analysis performed on the text data in accordance with an instruction from the user. Further, the screen display unit 16 displays a screen including a temporal change in the result of the hierarchical cluster analysis in accordance with an instruction from the user.

  FIG. 2 is a block diagram illustrating a configuration of a computer that functions as the text mining apparatus 10. The computer 20 illustrated 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. For example, a DRAM is used as the main memory 22. For the storage unit 23, for example, a hard disk or a solid state drive is used. The input unit 24 includes a keyboard 28 and a mouse 29, for example. For example, a liquid crystal display is used for the display unit 25. The communication unit 26 is an interface circuit for wired communication or wireless communication. The recording medium reading unit 27 is an interface circuit of the recording medium 30 that stores programs and the like. As the recording medium 30, for example, 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 text data 32. For example, the text mining program 31 and the text data 32 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.

  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 a text mining program 31 stored in the main memory 22 by using the main memory 22 as a working memory, thereby extracting a word from the text data 32 and hierarchically processing the extracted word. Processing to perform cluster analysis, processing to display a screen containing the analysis results, etc. At this time, the computer 20 functions as the text mining device 10. The configuration of the computer 20 described above is merely an example, and the text mining apparatus 10 can be configured using an arbitrary computer.

  FIG. 3 is a flowchart showing the operation of the text mining apparatus 10. Before performing the operation shown in FIG. 3, the text data storage unit 12 stores one or more text data that are freely described and accumulated. The text data is composed of a sentence having a date (for example, a work date or a reception date), and the text data is divided into a plurality of parts according to the date. The text mining device 10 performs processing on the text data specified by the user among the text data stored in the text data storage unit 12.

  In FIG. 3, the instruction input unit 11 first receives an instruction for specifying text data to be analyzed, an analysis period, and an analysis interval from the user (step S101). The user uses the input unit 24 to input such information in a dialog box (not shown) displayed on the screen. The received instruction is output to each unit of the text mining device 10.

  Next, the word extraction unit 13 reads the designated text data from the text data storage unit 12 (step S102). Next, the word extraction unit 13 extracts words from the read text data by performing morphological analysis on the text data read in step S102 (step S103). At this time, the word extraction unit 13 extracts only words necessary for later analysis from the read text data.

  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 obtains the appearance frequency of the word extracted in step S103 (step S105). Next, the analysis result storage unit 15 stores the hierarchical cluster analysis result obtained in step S104 and the word appearance frequency obtained in step S105 (step S106).

  The clustering processing unit 14 receives from the instruction input unit 11 the analysis period and analysis interval specified by the user. The analysis period indicates a period during which hierarchical cluster analysis is actually performed in the text data to be analyzed. The analysis period is divided into a plurality of periods with the analysis interval as a unit. For example, if the analysis period is a period 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.

  Let p be the number of periods after division. In step S104, the clustering processing unit 14 performs a hierarchical cluster analysis for each of the p periods. More specifically, for each of the p periods, the clustering processing unit 14 uses, for the word extracted in step S103, a sentence having a date and time within the period read out in step S102. Perform hierarchical cluster analysis. For example, the clustering processing unit 14 determines the similarity between two words based on the distance between two words in the text data 32 (how far 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 Ward method, etc.) based on the obtained similarity between words.

  In step S105, the clustering processing unit 14 obtains the appearance frequency of the word for each of the p periods. In step S104, p hierarchical cluster analysis results are obtained, and in step S105, p word application frequencies are obtained. In step S106, the analysis result storage unit 15 stores the result of hierarchical cluster analysis and the appearance frequency of words for each of the p periods.

  Next, the screen display unit 16 displays a screen including the result of the hierarchical cluster analysis stored in the analysis result storage unit 15 (step S107). FIG. 4 is a diagram illustrating an example of a window displayed in step S107. A window 41 shown in FIG. 4 contains the results of the hierarchical cluster analysis. When the number of clusters is set for the result of the hierarchical cluster analysis, words included in each cluster are determined. When the text mining device 10 displays a screen including the result 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 the number of clusters and the maximum number of display data in the clusters as operation parameters. Hereinafter, the former is m and the latter is n. These values are set to predetermined initial values in the initial state. The user may arbitrarily set these values using the instruction input unit 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 included in each cluster are displayed inside each cluster. The number of words displayed in each cluster is limited to n or less. For example, when n = 5 and a cluster includes 10 words, 5 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 device 10 determines whether or not the instruction received in step S111 is an instruction for designating the attention word (step S112). The control of the text mining device 10 proceeds to step S121 if Yes and proceeds to step S113 if No.

  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, or an instruction to close the window. The screen display unit 16 displays the updated screen according to the instruction received in step S111 (step S113). Thereafter, the control of the text mining device 10 proceeds to step S111.

  When step S111 is executed, a screen including the result of the hierarchical cluster analysis is displayed. Hereinafter, it is assumed that a screen including the window 41 shown in FIG. 4 is displayed when step S111 is executed. In addition, clicking the button of the mouse 29 when the mouse cursor 43 is on an element on the display screen is called “clicking an element”, and a cluster including the attention word is called “attention word cluster”. The name given to the attention word cluster is called “attention word cluster name”.

  FIG. 5 is a diagram illustrating an operation for designating the attention word. The user clicks a word (here, “decomposition”) designated as the attention word in the window 41 (first click). At this time, a context menu 42 appears in the display screen. The user clicks the item “to change with time of analysis result” in the context menu 42 (second click). By this operation, the word clicked for the first time is designated as the attention word.

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

  FIG. 6 is a diagram showing a window showing the change over time of the analysis result displayed in step S122. The window 51 shown in FIG. 6 is displayed when “decompose” is designated as the attention word in step S111. For example, as shown in FIG. 7, the window 51 is displayed so as to overlap the window 41 shown in FIG.

  The window 51 includes a line graph 52 showing a temporal change in the appearance frequency of the attention word along the time axis extending in the horizontal direction. As the appearance frequency of the attention word, for example, the ratio of the appearance frequency of the attention word to the total appearance frequency of all the words included in the attention word cluster is used. The appearance frequency of the attention word may be switched to the appearance frequency of the attention word in accordance with an instruction from the user.

  The cluster configuration (elements included in the cluster) obtained by the hierarchical cluster analysis changes with time. Clusters are automatically named to show their aging over time. When the cluster includes only one word, the word is used as it is in the cluster name. When a cluster includes two words, a cluster name in which two words are connected in descending order of application frequency is used. When the cluster includes three or more words, the cluster name is a concatenation of three words having the highest appearance frequency among the words included in the cluster in descending order of appearance frequency. If 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.

  FIG. 8A to FIG. 8D are diagrams illustrating examples of changes over time in the results of hierarchical cluster analysis. 8A to 8D describe the results of hierarchical cluster analysis in different months. In FIGS. 8A to 8D, the cloud shape represents a cluster, and the underlined character string represents a cluster name. The size of the circle represents the appearance frequency of the words described in the circle.

  In the analysis result shown in FIG. 8A, words extracted from the text data are classified into clusters including “drive” and “decomposition”, clusters including “exhaust”, “pressure”, “flow”, and “valve”, and “ It is classified into clusters including “belt”, “rotation”, “check”, “motor” and “tension”. These three clusters are named “drive / disassembly”, “exhaust / pressure / flow”, and “belt / rotation / check”, respectively. Regarding the analysis results shown in FIGS. 8B to 8D, the three clusters are named in the same manner.

  When “decomposition” is designated as the attention word, the attention word cluster name is “decomposition / drive” in the analysis result shown in FIG. 8A, and “drive / belt / rotation” in the analysis result shown in FIG. 8B. The analysis result shown in FIG. 8C is “exhaust / pressure / flow”, and the analysis result shown in FIG. 8D is “exhaust / pressure / decomposition”. Thus, the attention word cluster name changes with time.

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

  An arrow 55 is displayed above the boundary line 54 at a position corresponding to the time when the attention word cluster name 53 changes. The arrow 55 is displayed in a manner corresponding to the degree of change of the attention word cluster name 53. When all the words constituting the attention word cluster name 53 change, a red arrow 55r is displayed. When two words constituting the attention word cluster name 53 change, a blue arrow 55b is displayed. When one word constituting the attention word cluster name 53 changes, a black arrow 55n is displayed. The display mode of the arrow 55 may be arbitrary as long as it differs depending on the degree of change of the attention word cluster name 53. For example, the display size of the arrow 55 may be different depending on the degree of change of the attention word cluster name 53.

  In the example shown in FIG. 6, the attention word cluster name 53 is sequentially changed in the order of “drive / disassembly”, “drive / belt / rotation”, “exhaust / pressure / flow”, and “exhaust / pressure / decomposition”. Change. In the first change, two words constituting the attention word cluster name 53 change, so a blue arrow 55 b is displayed on the first boundary line 54. Since all the words constituting the attention word cluster name 53 change in the second change, a red arrow 55 r is displayed on the second boundary line 54. In the third change, since one word constituting the attention word cluster name 53 changes, a black arrow 55 n is displayed on the third boundary line 54.

  Next, the screen display unit 16 determines 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 or not the change in the attention word cluster name 53 is large based on the number of each type of arrow 55 (step S124). For example, the screen display unit 16 may determine Yes when the number of red arrows 55r exceeds 30% of the total number of arrows 55, and the total of the number of red arrows 55r and the number of blue arrows 55b is the arrow 55. You may judge Yes when it exceeds 60% of the total number of. The control of the text mining device 10 proceeds to step S125 if Yes, and proceeds to step S111 if No.

  In the former case, the screen display unit 16 displays a screen including a warning message (step S125). FIG. 9 is a diagram showing the window displayed in step S125. The window 61 shown in FIG. 9 includes a warning message indicating that it is recommended to review the setting of the hierarchical cluster analysis (for example, the number of clusters and the number of target words) because the configuration of the attention word cluster often changes greatly. Yes. 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 includes a step of extracting words from text data including sentences having dates (steps S102 and S103), and a hierarchical analysis for each extracted analysis word for each analysis period. A step of performing cluster analysis (step S104) and a step of displaying a screen including the analysis result by hierarchical cluster analysis (step S105) are provided. When an instruction for designating the attention word is input in the first screen (the screen including the window 41) including the analysis result (FIG. 5), the step of displaying the screen (step S122) is performed for the cluster including the attention word. A second screen (screen including window 51) showing the change over time is displayed. According to the text mining method according to the present embodiment, when an instruction for designating a word of interest is input in the first screen including the result of the hierarchical cluster analysis, the second change indicating the temporal change of the cluster including the word of interest. By displaying the screen, the user can easily recognize the temporal change in the result of the hierarchical cluster analysis.

  Further, the second screen shows the cluster name (the attention word cluster name 53) based on the words included in the cluster including the attention word along the time axis. In addition, this cluster name is obtained by concatenating a predetermined number or less (three or less) of words included in a cluster including the attention word in descending order of appearance frequency. Therefore, the user can easily recognize the temporal change of the cluster including the attention word.

  Further, 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 attention 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 in the name of the cluster including the attention word. Further, of the words constituting the cluster name, words that have changed from the previous cluster name (such as “belt” and “rotation” shown in FIG. 6) are highlighted in the second screen. Therefore, the user can easily recognize how a word having a high appearance frequency has changed in the cluster including the attention word.

  Further, the second screen includes a graph (line graph 52) showing a change with time of appearance frequency of the attention word along the time axis. In addition to the change over time of the cluster containing the attention word, the user can easily recognize the change over time in the results of the hierarchical cluster analysis by displaying a screen containing a graph showing the change over time in the appearance frequency of the attention word. Can do. Further, the second screen includes a boundary line 54 at a position corresponding to the time when the name of the cluster including the attention word changes, and the background of the graph has a different aspect for each boundary line. Therefore, the user can easily recognize when the cluster including the attention word changes. When the name of the cluster including the attention word often changes greatly, the step of displaying the screen displays a screen including the warning message (screen including the window 61). Therefore, the user can recognize that the hierarchical cluster analysis is malfunctioning.

  The text mining apparatus 10 and the text mining program 31 according to the present embodiment have the same characteristics as the above text mining method and have 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 temporal change of the result of the hierarchical cluster analysis.

DESCRIPTION OF SYMBOLS 10 ... Text mining apparatus 11 ... Instruction input part 12 ... Text data storage part 13 ... Word extraction part 14 ... Clustering process part 15 ... Analysis result storage part 16 ... Screen display part 20 ... Computer 21 ... CPU
DESCRIPTION OF SYMBOLS 22 ... Main memory 29 ... Mouse 30 ... Recording medium 31 ... Text mining program 32 ... Text data 41, 51, 61 ... Window 42 ... Context menu 43 ... Mouse cursor 52 ... Line graph 53 ... Attention word cluster name 54 ... Boundary line 55 ... arrow

Claims (20)

A text mining method for displaying a screen including an analysis result of text data,
Extracting words from text data consisting of sentences with dates;
Performing a hierarchical cluster analysis on each word for each analysis period;
Displaying a screen containing the results of the hierarchical cluster analysis,
The step of displaying the screen when an instruction for designating the attention word is input in the first screen including the result displays a second screen showing a change over time of the cluster including the attention word. And text mining method.   The text mining method according to claim 1, wherein the second screen shows a cluster name based on a word included in the cluster along a time axis.   The text mining method according to claim 2, wherein the cluster name is a concatenation of words included in the cluster by a predetermined number or less in descending order of appearance frequency.   The text mining according to claim 2, wherein 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. Method.   The text mining method according to claim 4, wherein the mark is an arrow having a color corresponding to a degree of change of the cluster name.   The text mining method according to claim 2, wherein a word that has changed from a previous cluster name among words constituting the cluster name is highlighted in the second screen.   The text mining method according to claim 2, wherein the second screen further includes a graph showing a temporal change in the appearance frequency of the attention word along the time axis.   The second screen according to 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. Text mining method.   3. The text mining method according to claim 2, wherein when the cluster name often changes greatly, the step of displaying the screen displays a screen including a warning message. 4. A text mining program for displaying a screen including the analysis result of text data,
Extracting words from text data consisting of sentences with dates;
Performing a hierarchical cluster analysis on each word for each analysis period;
Causing the computer to execute a step of displaying a screen including the result of the hierarchical cluster analysis using a memory,
The step of displaying the screen when an instruction for designating the attention word is input in the first screen including the result displays a second screen showing a change over time of the cluster including the attention word. A text mining program.   The text mining program according to claim 10, wherein the second screen shows a cluster name based on a word included in the cluster along a time axis.   12. The text mining program according to claim 11, wherein the cluster name is obtained by concatenating a predetermined number or less of words included in the cluster in descending order of appearance frequency.   The text mining according to claim 11, wherein 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. program.   The text mining program according to claim 13, wherein the mark is an arrow having a color corresponding to a degree of change of the cluster name.   12. The text mining program according to claim 11, wherein a word that has changed from a previous cluster name among words constituting the cluster name is highlighted in the second screen.   12. The text mining program according to claim 11, wherein the second screen further includes a graph showing a change with time of appearance frequency of the attention word along the time axis.   17. The second screen according to claim 16, 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. Text mining program.   12. The text mining program according to claim 11, wherein when the cluster name often changes greatly, the step of displaying the screen displays a screen including a warning message. A text mining device that displays a screen including the analysis result of text data,
A word extraction unit that extracts words from text data including sentences having dates;
A clustering processing unit that performs hierarchical cluster analysis for each analysis period for the word;
A screen display unit for displaying a screen including the result of the hierarchical cluster analysis,
When an instruction for designating a word of interest is input in the first screen including the result, the screen display unit displays a second screen showing a change over time of a cluster including the word of interest. , Text mining equipment.   The text mining device according to claim 19, wherein the second screen shows a cluster name based on a word included in the cluster along a time axis.

Images