JP2005182133A - Document classification device and document classification method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To facilitate discrimination of a category of a classification destination and change of a category of documents erroneously classified. <P>SOLUTION: A document classification device includes a document storage section, a category setting section, a document classification section, a classification result storage section, a classification list acquisition section, a candidate list acquisition section, a list display section, a change category selection section, and a change result storage section. The classification list acquisition section acquires a catalogue list of categories of electronic documents classified in the document classification section, the candidate list acquisition section acquires a catalogue list of categories of electronic documents which is not classified in the document classification section, and the list display section displays contents of a classification list and a candidate list. The change category selection section selects a change category for the electronic document in a candidate list CDL of a category. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a document classification apparatus and a document classification method for classifying electronic documents created electronically and digitized documents including character-recognized image documents into predetermined categories.

  Along with the development of computers and networks, the accumulation of a large amount of digitized documents in a database is progressing, and along with this, the expectation for electronic processing relating to a large amount of digitized document databases, for example, document classification, has increased.

Japanese Patent Application Laid-Open No. 2004-228561 discloses a document that automatically classifies documents, displays candidates in the order of similar categories, and then allows a user to determine a category to be classified.

JP-A-9-22414

  However, in Patent Document 1, since documents are displayed one by one and a user determines a category to be classified, when a plurality of documents are classified such as how other documents are classified. It was difficult to grasp the whole picture.

  The present invention has been made to solve such a conventional problem, and an object thereof is to facilitate classification of a classification destination category and category change of a misclassified document.

  Another object of the present invention is to make it possible to classify electronic documents into a plurality of categories.

  According to the present invention, it is possible to easily display in which category a plurality of documents are classified, and it is also possible to easily correct a category of a misclassified document.

  The document classification apparatus according to the present invention includes a classification unit that classifies an input electronic document into a category, and a candidate category list acquisition unit that acquires, as a candidate list, categories that are candidates when the electronic document is classified by the classification unit. Display means for displaying a list of the plurality of electronic documents classified by the classification means so that the category into which each electronic document is classified can be determined, and a user-desired one of the plurality of electronic documents displayed in the list. Display control means for controlling to display a candidate category list of the instructed electronic document while displaying a list of a plurality of electronic documents displayed on the display means when the electronic document is instructed to be corrected A correction for correcting the classified category of the designated electronic document based on the category selected from the candidate category list And a stage.

  The document classification method of the present invention includes a classification step of classifying an input electronic document into a category, and a candidate category list acquisition step of acquiring a candidate category as a candidate list when the electronic document is classified by the classification step. A display step of displaying a list of the plurality of electronic documents classified in the classification step so that a category into which each electronic document is classified can be determined, and a user-desired one of the plurality of electronic documents displayed in the list A display control step for controlling to display a candidate category list of the instructed electronic document while displaying a list of the plurality of electronic documents displayed in the display step when the electronic document is instructed to be corrected The designated electronic document is classified based on a category selected from the candidate category list. And a correction step of modifying the category you are.

  Next, Embodiment 1 of the document classification apparatus and document classification method according to the present invention will be described with reference to the drawings.

  1 is an overall block diagram showing a first embodiment of a document classification apparatus according to the present invention, FIG. 2 is a functional block diagram of the document classification apparatus of FIG. 1, and FIG. 3 is implemented by the document classification apparatus of FIG. FIG. 4 is a diagram illustrating an example of categories set by the document classification apparatus in FIG. 1, and FIG. 5 is a flowchart of a document acquired from a document database in the document classification apparatus in FIG. FIG. 6 is a diagram showing examples of documents and classification categories in the classification result of the document classification apparatus in FIG. 1. FIG. 7 is a diagram showing categories of digitized documents classified by the classification list acquisition unit in FIG. FIG. 8 is a table showing an example of a classification list composed of a list. FIG. 8 is a table showing an example of a classification list composed of a list of categories of digitized documents not classified by the classification list acquisition unit of FIG. For computerized document classification devices, FIG. 10 is a diagram illustrating a state in which a change category is selected from the candidate categories in the document classification apparatus in FIG. 1, and FIG. 11 is a diagram from the candidate categories in the document classification apparatus in FIG. It is a figure which shows the display state of the selected change category.

  In FIG. 1, the document classification apparatus has a CPU (microprocessor) 100 that performs operations for document classification processing, logical determination, and the like, and operates as document classification display means. The RAM 120, the input device 130, the display device 140, the scanner 150, the HD (hard disk) 160, the external storage device 170, and the communication device 180 are connected.

  The bus B includes an address signal line and a control signal line, and performs address designation and control content designation when the CPU 100 controls each component, and performs data transfer between the components.

  The RAM 120 is a writable random access memory, and is used for primary storage of various data of each component.

  The ROM 110 is a read-only fixed memory, and stores a boot program by the CPU 100 and the like. The boot program loads the control program 162 stored in the HD 160 into the RAM 120 when the document classification apparatus is started up and is executed by the CPU 100. A control program for executing the document classification process will be described in detail with reference to the flowchart of FIG.

  The input device 130 is a keyboard, a mouse, and the like, the display device 140 is a CRT, a liquid crystal display, and the like. The scanner 150 reads a document recorded on a sheet such as paper, performs character recognition, and extracts text data.

  In addition to the control program 162, the HD 160 stores a document database 164 that stores electronically created electronic documents, and a classification dictionary 166 that stores classification information for classifying the documents.

  The external storage device 170 is a drive for accessing a removable external storage medium such as a floppy disk, CD, DVD, etc., and can be used in the same manner as the HD 160. Data exchange with other document processing apparatuses can be performed through these recording media. Is possible. If necessary, the control program 162 to be stored in the HD 160 may be copied from the external storage device.

  The communication device 180 is a network controller and exchanges data with the outside via a communication line.

  The document classification device operates in response to various inputs from the input device 130. When input from the input device 130 is started, an interrupt signal is first sent to the microprocessor CPU100. The CPU 100 reads various commands stored in the ROM 120 or the RAM 110 in response to the interrupt signal, and performs various control of the components by executing the commands.

  In FIG. 2, the document classification apparatus has a configuration indicated by the following functional blocks.

  The document classification device includes a document storage unit 202, a category setting unit 203, a document classification unit 204, a classification result storage unit 205, a classification list acquisition unit 206, a candidate list acquisition unit 207, a list display unit 208, a change category selection unit 209, a change The function block of the result storage unit 210 is included.

  The document storage unit 202 corresponds to the document database 164 of FIG. 1 and stores the digitized document DD.

  The category setting unit 203 and the document classification unit 204 correspond to the CPU 100 and the classification dictionary 166, the classification result storage unit 205 corresponds to the RAM 120, the classification list acquisition unit 206, the candidate list acquisition unit 207, the change category selection unit 209, and the change result. The storage unit 210 corresponds to the CPU 100 and the RAM 120, and the list display unit 208 corresponds to the display device 140.

  The category setting unit 203 sets a category for classifying the digitized document DD, and the document classification unit 204 classifies the digitized document DD into the category set by the category setting unit 203.

  The classification result storage unit 205 stores the category of the electronic document DD classified by the document classification unit 204, and the classification list acquisition unit 206 stores a list of categories of the digitized document DD classified by the document classification unit 204. get.

  The candidate list acquisition unit 207 acquires a list of categories of the digitized documents DD that are not classified by the document classification unit 204, and the list display unit 208 displays the contents of the classification list CLL and the candidate list CDL.

  The change category selection unit 209 selects a change category in the category candidate list CDL acquired by the candidate list acquisition unit 207 and displayed on the list display unit 208 for the digitized document DD.

  The change result storage unit 210 stores the category of the digitized document DD changed by the change category selection unit 209.

  The classification list CLL is the category of the digitized document DD classified by the document classification unit 4 as a list list by the classification list acquisition unit 206, and the candidate list CDL is the category acquired by the candidate list acquisition unit 207. A list of candidates.

  In FIG. 3, the document classification apparatus executes document classification processing by a document classification method including the following steps.

  Step S301: The category setting unit 203 sets a category for classifying the digitized document DD in the document database 164 of the HD 160. At this time, the screen of FIG. 4 is displayed on the display device 140, and for example, categories such as “politics”, “economy”, “society”, “sports”, etc. are set. The screen includes a display box 400 for displaying these categories, a “create” button 401 for creating a category, an “execute” button 402 for confirming the created category, and a “cancel” button for canceling category creation. 403 is provided.

  Step S302: Subsequent to step S301, from the document storage unit 2 (document database 164), the stored electronic document DD, for example, the electronic document “circle market” DD501, the electronic document “corrected budget” DD502, electronic The document “soccer” DD503 is acquired.

  Step S303: The digitized documents DD501 to DD503 acquired in step S302 are classified by the document classification unit 204 according to the category set in step S301, and the classification result is displayed on the screen of the display device 140. For example, as shown in the screen image of FIG. 6, the digitized document “corrected budget” is classified into the category “politics”, the digitized document “yen quote” is classified into the category “economy”, and the digitized document “soccer”. Is classified into the category “sports”. On the screen, an “OK” button 601 is provided on which the user confirms the classification result.

  Step S304: The computerized documents DD501 to DD503 and each classification category in the classification result of step S303 are stored in the classification result storage unit 205.

  Step S305: The classification list acquisition unit 6 acquires a list of categories of the digitized documents DD501 to DD503 classified in step S303. For example, as shown in the table of FIG. 7, the classification category “economy” corresponding to the document name “yen quote” of the digitized document DD501, and the classification category “politics” corresponding to the document name “corrected budget” of the digitized document DD502. The classification category name “sports” corresponding to the document name “soccer” of the digitized document DD503 is acquired.

  Step S306: The category that has not been adopted as the classification category in Step S303 is obtained as a candidate category that has not been adopted by the candidate list obtaining unit 207, and a list list is generated. For example, when classified by the document classification unit 204, “economy” and “politics” are candidate categories for the document “yen exchange rate”, and “politics” is a candidate category for the document “corrected budget”. It is assumed that “sports”, “society”, and “politics” are candidate categories for “soccer”, and that the respective first candidates are selected in step S303. In this case, as shown in FIG. 8, the candidate category “politics” that has not been adopted is acquired for the document name “yen quote” of the digitized document DD501 and corresponds to the document name “soccer” of the digitized document DD502. Then, candidate categories “society” and “politics” that have not been adopted are acquired. On the other hand, since the only classification category “politics” is extracted from the electronic document DD503 “corrected budget”, there is no candidate category.

  Step S307: The contents of the category classification list (FIG. 7) acquired in step S305 and the category candidate list (FIG. 8) acquired in step S306 are displayed on the display device 140 by the list display unit 208. For example, as shown in FIG. 9, regarding the digitized document “soccer” DD503, a document name display 902 is displayed as belonging to the classification category “sports” 901 of the classification list, and the document “soccer” 902 is displayed by the user. When designated, the candidate categories “society” and “politics” 903 that are not employed when the document “soccer” is classified are displayed on the screen based on the candidate list.

  The screen of FIG. 9 is provided with an “OK” button 904, and the user presses the “OK” button 904 when adopting the category at that time.

  Step S308: If any radio button of the candidate category 903 is pressed, it is determined that the category is to be changed, and the process proceeds to step S309. If the category is not changed, the process is terminated as it is.

  Step S309: When the “OK” button is pressed after the candidate category radio button is pressed, the candidate category corresponding to the pressed radio button is determined and the change process is executed. For example, as shown in FIG. 9, when the document “soccer” is designated by a cursor or the like and the candidate category is displayed, as shown in FIG. 10, any candidate category (for example, “society”) is displayed. When the radio button is selected and designated by the user and the “OK” button 1003 is pressed, the selected candidate category “society” is identified and the change process is executed.

  In step S310, the category of the digitized document 1 changed in step S309 is stored in the change result storage unit 12 (the classification list and the candidate list are updated). Thereafter, the process returns to steps S305 to 307, and the display according to the changed category is performed. For example, as shown in FIG. 11, the category of the electronic document “soccer” DD503 is changed from “sports” to “society”. Will be displayed.

  Next, a second embodiment of the document classification method performed by the document classification apparatus of FIG. 1 will be described based on the drawings.

  FIG. 16 is a frequency list showing the frequency with which each category is used.

  In the second embodiment, an electronic document is classified and displayed as a list based on the usage frequency of each category.

  In the second embodiment, the processes in steps S303 to S307 and S310 in FIG.

  Hereinafter, processing different from that in the first embodiment will be described in the second embodiment.

  In the second embodiment, when performing document classification in step S303 in FIG. 3, candidate categories are obtained for the input electronic document, and the obtained candidate categories are determined according to the frequency of use of each category described in the frequency list. Arrange in order of frequency of use. Then, the first category is stored as a classification category of the electronic document in step S304, and candidate categories that have not been adopted are acquired as a candidate list in step S306.

  If the display of candidate categories is instructed in step S307, the candidate categories are displayed in order of use frequency according to the candidate list acquired in step S306.

  If the category is changed, in step S310, the category is updated, and the frequency list is updated assuming that the selected category is used.

  Next, a third embodiment of the document classification method performed by the document classification apparatus in FIG. 1 will be described with reference to the drawings.

  In the third embodiment, a Japanese syllabary list in which categories used for classification are arranged in the order of 50 tones is used.

  Then, when the classification list is acquired in step S305 and when the candidate list is acquired in step S306, the categories are arranged in the order of 50 tones based on the Japanese syllabary list.

  Then, when displaying the classification list as shown in FIGS. 9 to 11 and when displaying the candidate list, the categories are displayed in alphabetical order.

  Next, a fourth embodiment of the document classification method performed by the document classification apparatus in FIG. 1 will be described with reference to the drawings.

  In the fourth embodiment, when displaying the contents of the classification list and the contents of the candidate list, in order to clearly distinguish the classification category as the classification list and the candidate category in the candidate list, the classification category and the candidate Change the display color for each category. It should be noted that instead of the display color, the font or font size may be changed for display.

  Next, a fifth embodiment of the document classification method performed by the document classification apparatus in FIG. 1 will be described with reference to the drawings.

  12 is a functional block diagram showing a fifth embodiment of the document classification apparatus according to the present invention, FIG. 13 is a flowchart showing a fifth embodiment of the document classification method implemented by the document classification apparatus of FIG. 12, and FIG. FIG. 15 is a diagram showing a maximum category number setting screen in the processing of FIG. 13, and FIG. 15 is a diagram showing the relationship between the maximum number of categories and the classification list and candidate list in the processing of FIG. In the figure, the same or corresponding parts as those in the first embodiment are denoted by the same reference numerals, and description thereof is omitted.

  In the fifth embodiment, one electronic document can be classified into a plurality of categories, and the maximum number of categories into which one document can be classified is set in advance.

  In FIG. 12, the fifth embodiment includes a category number setting unit 1200 in addition to the configuration of the first embodiment.

  The digitized document DD is classified by the document classification unit 205 based on the maximum number of categories set by the maximum category number setting unit 1200.

  The processing after the document classification unit 205 is the same as in the first embodiment.

  In FIG. 13, steps S1301, S1305, and S1308 to S13011 correspond to steps S301, S304, and S1307 to S310 of the first embodiment. In the fifth embodiment, steps S1302, S1306, and S1307 are added. The contents of are significantly different from those of the first embodiment.

  Hereinafter, only steps S1302, S1304, S1306, and S1307 in FIG. 13 will be described, and description of other steps will be omitted.

  Step S1302: The category number setting unit 1200 sets the maximum number of categories for classifying the digitized document DD.

  When setting the maximum number of categories, for example, the screen of FIG.

  On the same screen, a maximum category number input box 1400, an “execute” button 1401, and a “stop” button 1402 are provided, and the maximum category number (“10” in the figure) is displayed in the maximum category number input box 1400 by an input device 130 such as a keyboard. ”Is entered.)

  When the inputted maximum category number is adopted as it is, the “execution” button 1401 is pressed to determine the maximum category number. When the maximum number of categories once input is changed, a “stop” button 1402 is pressed.

  Step S1304: The computerized document acquired in step S1303 is classified by the document classification unit 204 according to the category set in step S1301. At this time, the classification category is selected within the range of the maximum number of categories for classifying the digitized document set in step S1302.

  Step S1305: The results classified in step 1304 are stored in a list.

  Step S1306: When any of the documents is designated by the user, the classification list acquisition unit 107 acquires a classification list of the category of the specified digitized document.

  Step S1307: The category not adopted as the classification category in step S1304 is obtained as a candidate category not adopted by the candidate list acquisition unit 207, and a list list including the classification list and the candidate list is generated. This candidate category that has not been adopted is a candidate category having a lower rank than the maximum number of categories set in step S1302.

  In FIG. 15, for example, when the maximum number of categories is set to “3”, and classification is performed on the electronic document “recycle”, “administrative general”, “environment”, “politics”, “sports”, Suppose five of “Society” are extracted. At this time, since the maximum number of categories is 3, the top three categories of “administrative general”, “environment”, and “politics” are extracted as classification categories, and “sports” and “society” are not adopted as candidate categories. Categories will be extracted.

  For example, FIG. 17 shows that when the maximum classification number is set to 2, the input document “soccer” is classified into the categories “sports” and “politics”, and the document “soccer” belonging to the classification category “sports”. "Is a display screen when a list is displayed in step S1308. A category whose check box is inverted in black is a classification category, and a category whose check box is blank is a candidate category. At this time, as shown in FIG. 18, when the category “society” is designated, the maximum number of categories is exceeded, so the category “sports” of the designated document is turned off. In this way, control is performed so as not to exceed the maximum number of categories. When the OK button is pressed on the screen of FIG. 18, the category to which the document belongs is changed and displayed as belonging to the categories of “society” and “politics” as shown in FIG.

  With this configuration, it is possible to classify one document into a plurality of categories, and the user can adjust the maximum number of classification categories to be classified.

1 is an overall block diagram illustrating Embodiment 1 of a document classification device according to the present invention. (Example 1) It is a functional block diagram of the document classification device of FIG. (Example 1) It is a flowchart which shows Example 1 of the document classification method implemented with the document classification device of FIG. (Example 1) It is a figure which shows the screen image at the time of setting a category with the document classification device of FIG. (Example 1) It is a figure which shows the example of the digitized document acquired from the document database in the document classification device of FIG. (Example 1) It is a figure which shows the classification result by the document classification device of FIG. (Example 1) It is a table | surface which shows the example of the classification list which consists of a list of the categories of the digitized document classified by the classification list acquisition part of FIG. (Example 1) It is a table | surface which shows the example of the classification list which consists of a list | wrist of the category of the digitized document which is not classified by the classification list acquisition part of FIG. (Example 1) It is a figure which shows the classification category and candidate category with respect to an electronic document in the document classification device of FIG. (Example 1) It is a figure which shows the state which selects the change category from the candidate category in the document classification device of FIG. (Example 1) It is a figure which shows the display state of the change category selected from the candidate category in the document classification device of FIG. (Example 1) It is a functional block diagram which shows Example 5 of the document classification device based on this invention. (Example 5) FIG. 13 is a flowchart illustrating a fifth embodiment of the document classification method performed by the document classification apparatus in FIG. 12. FIG. (Example 5) It is a figure which shows the setting condition of the maximum category number in the process of FIG. (Example 5) FIG. 14 is a diagram illustrating the relationship between the maximum number of categories and the classification list and candidate list in the process of FIG. 13. (Example 5) It is a figure which shows the example of the frequency list of a category. It is a display screen which shows the document classified into the some category. It is a display screen when an instruction to change the classification category of a document classified into a plurality of categories is given. It is a display screen which shows the result as which the classification category of the document classified into the some category was performed.

Explanation of symbols

202 Document storage unit 203 Category setting unit 204 Document classification unit 205 Classification result storage unit 206 Classification list acquisition unit 207 Candidate list acquisition unit 208 List display unit 209 Change category selection unit 210 Change result storage unit CLL classification list CDL candidate list DD Electronic documents

Claims (14)

A classification means for classifying the input electronic document into categories,
Candidate category list acquisition means for acquiring, as a candidate list, categories that are candidates when the electronic document is classified by the classification means;
Display means for displaying a list of the plurality of electronic documents classified by the classification means so that the category into which each electronic document is classified can be determined;
When a user's desired electronic document is instructed to be corrected from among the plurality of displayed electronic documents, the list of the plurality of electronic documents displayed on the display means is displayed and Display control means for controlling to display a candidate category list;
Correcting means for correcting the classified category of the instructed electronic document based on the category selected from the candidate category list;
A document classification apparatus comprising: 2. The document classification apparatus according to claim 1, wherein when the category is corrected by the correcting unit, the display unit updates and redisplays the list based on the corrected category. . The document classification apparatus according to claim 1, wherein the candidate category list is displayed in descending order of classification frequency of the categories. The document classification apparatus according to claim 1, wherein the candidate category list is displayed in alphabetical order of category names in the category. The document classification apparatus according to claim 1, wherein when the list and the candidate category list are simultaneously displayed by the display control unit, control is performed so that each list can be identified. The document classification apparatus according to claim 1, further comprising category number setting means for setting a maximum number of categories in which each electronic document can be classified by the classification means. A classification step for classifying the input electronic document into categories,
A candidate category list obtaining step for obtaining a candidate list as a candidate list when the electronic document is classified by the classification step;
A display step of displaying a list of the plurality of electronic documents classified by the classification step so that each electronic document can be classified.
When a user-desired electronic document is instructed to be corrected from among the plurality of electronic documents displayed in the list, the list of the plurality of electronic documents displayed in the display step is displayed and the specified electronic document is displayed. A display control step for controlling to display a candidate category list;
A correction step of correcting a classified category of the designated electronic document based on a category selected from the candidate category list;
A document classification method comprising: 8. The document classification method according to claim 7, wherein when the category is corrected by the correction step, the display step updates and redisplays the list based on the corrected category. . 8. The document classification method according to claim 7, wherein the candidate category list is displayed in descending order of classification frequency of the categories. The document classification method according to claim 7, wherein the candidate category list is displayed in order of the alphabetical order of the category names in the category. 8. The document classification method according to claim 7, wherein when the list and the candidate category list are displayed at the same time in the display control step, the list is controlled so as to be identifiable. The document classification method according to claim 7, further comprising a category number setting step of setting a maximum number of categories in which each electronic document can be classified by the classification step. A computer-executable program for realizing the document classification method according to any one of claims 7 to 12 by a computer. A storage medium for storing a computer-readable program for realizing the document classification method according to claim 7 by a computer.

Images