JP5288617B2 - Document management system, document management method, and computer program - Google Patents

Description

  The present invention relates to an interface technique for easily grasping an area in which a keyword group of interest is described from document data.

  There has been proposed an interface technology that supports understanding of the description contents of document data by highlighting a keyword designated by the user from the document data (for example, Patent Documents 1 and 2). Also known is a technique for extending the conventional interface technology of this type, specifying different colors for each of a plurality of keywords, and coloring and displaying a plurality of keywords existing in document data with the specified colors. Yes.

JP 2003-44517 A JP 2003-22241 A

  The conventional interface technology is useful for understanding whether or not the specified keyword exists in the document data, and if so, it is only highlighted in the document data. Therefore, it is impossible to specify where the description content to be noticed of the document data exists.

  In addition, when there are a plurality of document data, it is impossible to grasp which part of the other document data corresponds to the content described in the one document data. This is because the two document data are created by different authors in many cases. Even if the same contents are described in terms of meaning, if the two document data are described in different words, the corresponding portions are accurately identified. This is because it is difficult to grasp. In order to make it possible to comprehensively grasp different words and phrases, it is necessary to specify synonyms of keywords to be specified in as much detail as possible, and the document comparison time depends on the skill of the user.

  SUMMARY OF THE INVENTION The main object of the present invention is to provide a document management interface technology capable of improving the readability of document data and reducing the time required for comparison of a plurality of document data. .

  The present invention aims to solve the above problems by a document management system using a spectral pattern that is a combination of color information, a document management method using a spectral pattern, and a computer program.

The document management system of the present invention accepts input of a plurality of keywords arranged in a predetermined order and color information associated with each keyword, and for each accepted keyword, a keyword ID (ID is an abbreviation of Identification, for identifying the keyword). The same) and the color information are registered in the first table in association with each other, and the designated document data is divided into predetermined areas, and each divided area is registered in the first table. A search is performed for whether or not a keyword exists. If there is a keyword, color information registered in the first table for the keyword, and if not, a spectrum pattern in which unregistered color information is arranged in the predetermined order is displayed for each divided region. The generated spectrum pattern is associated with a division ID for identifying the division region. When the spectrum data is registered in the two tables and the document data is displayed on a predetermined display device, the keywords registered in the first table of the document data are highlighted with color information unique to the keywords. And a display control means for displaying the spectrum pattern registered in association with the division ID in the second table together with the relative position information of the divided area identified by the division ID.

  In an embodiment, the system further comprises synonym management means for associating a synonym of a keyword registered in the first table with a keyword ID for the keyword and registering the synonym in a third table. A spectrum pattern of a divided region where a keyword registered in one table does not exist but a synonym associated with the keyword ID of the keyword exists in the third table is created based on color information unique to the keyword.

  In one embodiment, the similarity between each color information pattern registered in the first table and all spectrum patterns registered in the second table is determined, and the spectrum pattern having the highest similarity is determined. Further, a similar area search means for specifying a divided area of the divided ID associated with is provided.

The document management method according to the present invention is a method executed by a document management system connected to a display device, the step of displaying the first document data and the second document data on the display device, and the displayed first A plurality of keywords arranged in a predetermined order for characterizing the description content of the document data and input of color information associated with each keyword are received, and for each received keyword, a keyword ID for identifying the keyword and the color information Are registered in the first table in association with each other, and the second document data is divided into predetermined areas, and a search is performed to determine whether or not the keyword registered in the first table exists in each divided area. If it exists, the color information registered in the first table for the keyword is displayed. A spectral pattern arranged in a fixed order is generated for each divided region, and the generated spectral pattern is registered in the second table in association with a divided ID for identifying the divided region, and registered in the first table. Determining the similarity between each color information pattern and all the spectral patterns registered in the second table, and identifying the division region of the division ID associated with the highest similarity spectral pattern; Spectral patterns registered in association with the division IDs in the second table, while highlighting the keywords registered in the first table in the document data of the identified divided regions with color information unique to the keywords. Is displayed on the display device together with the relative position information of the divided area identified by the division ID. And a step, a document management method according to the spectral pattern.

The computer program of the present invention receives a plurality of keywords arranged in a predetermined order and color information associated with each keyword, and identifies the keyword for each received keyword. keyword management means for registering in the first table in association keyword ID and the said color information for, with dividing the designated document data for each predetermined region, in each divided area, it is registered in the first table If there is a keyword, the color information registered in the first table for the keyword is found. If not, the spectrum pattern in which the unregistered color information is arranged in the predetermined order is divided into regions. Generate each time, and identify the divided area by the generated spectrum pattern Spectrum management means for registering in association with the divided ID of fit to the second table, and the document data when displaying on a predetermined display device, the keywords registered in the first table of the document data Display control that highlights the color information unique to the keyword and displays the spectral pattern registered in association with the division ID in the second table together with the relative position information of the division area identified by the division ID it is a computer program to function means as.

  According to the present invention, since the presence / absence of a region related to a keyword and the relative position of the region can be visually grasped, the browsing property is improved, and when a plurality of document data are compared, the presence / absence of a related region is also easy. Since it can be grasped, it is possible to shorten the time required for comparing the document data and grasping the description contents.

1 is a configuration diagram of a document management system to which the present invention is applied. The figure which shows the example of registration of a keyword table. The figure which shows the example of registration of a synonym table. Explanatory drawing of original document data. The figure which shows the example of registration of a division | segmentation document table. Explanatory drawing of the procedure of a process which produces | generates a spectrum display pattern. The figure which showed the state before extraction of a keyword (a synonym is included) and after extraction. The figure which shows the example of registration of a spectrum table. Explanatory drawing of the procedure of the document management method by this embodiment. The figure which shows the example of a screen display in a display apparatus.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a configuration diagram of a document management system according to this embodiment. The document management system 1 includes a computer main body 10, an external storage device 20 such as a hard disk, an input device 30 including a keyboard, and a document data capturing device 40 including a reading mechanism from an OCR or USB memory. And a display device 50 such as a liquid crystal display.

  The computer main body 10 executes the computer program of the present invention installed in the external storage device 20 to make the computer main body an input / output (I / O) control unit 11, a keyword management unit 12, a synonym management unit 13, a document The management unit 14, the spectrum management unit 15, the display control unit 16, and the spectrum evaluation unit 17 function.

  In the external storage device 20, in addition to the computer program and necessary data, a keyword table 21, a synonym table 22, a divided document table 23, a spectrum table 24, and a document database (hereinafter, the database is referred to as “DB”). 25 and a synonym dictionary 26 are stored.

  The keyword table 21 stores “keyword” and “display color” in association with each “keyword ID”. The keyword ID is, for example, a reception number (serial number) or a keyword name. The synonym table 22 stores a synonym acquired from the synonym dictionary in association with each “keyword ID”. The divided document table 23 stores “document ID”, “division ID”, and “divided document data” in association with each other. The spectrum table 24 stores “document ID”, “division ID”, and “spectrum display pattern” in association with each other.

The functions realized in the computer main body 10 are as follows.
[Input / output control unit]
The input / output (I / O) control unit 11 receives processing instructions and input of search terms from the input device 30, takes in document data from the document data take-in device 40, and processes to the display device 50. Alternatively, output of processing results, data registration to the external storage device 20 and reading thereof are controlled.

[Keyword Management Department]
The keyword management unit 12 registers the keyword appearing in the target document data and the color information for highlighting the keyword by color in the keyword table 21 in association with each other. This process is performed by receiving one or more keywords and color information for coloring the keywords from the user. For each accepted keyword, a keyword ID is assigned with a serial number such as “01”, “02”..., And the received color information is registered in association with the keyword ID.
A registration example of the keyword table 21 registered in this way is shown in FIG. In the example of FIG. 2, the keyword “recognition” identified by the keyword ID “01” is colored and displayed in “red”. Similarly, the keyword “image” identified by the keyword ID “02” is identified by “yellow”, the keyword “failure” identified by the keyword ID “03” is identified by “green”, and the keyword ID “04”. The keyword “sensor” indicates that it is colored with “purple”.

[Synonym Management Department]
The synonym management unit 13 obtains synonyms by searching the synonym dictionary 26 for all keywords registered in the keyword table 21, and sets each synonym with the keyword ID of the search source keyword for each synonym. Register in the table 22.
A registration example of the synonym table 22 is shown in FIG. In the example of FIG. 3, since the keyword ID “01”, that is, the search source keyword is “recognition”, and synonyms “identification” and “specific” are obtained from the synonym dictionary 26, the keyword ID “01” and the synonym “ “Identification”, the keyword ID “01”, and the synonym “specific” are registered in the synonym table 22.
Similarly, since the search source keyword is “image” and the synonyms “image”, “still image”, and “frame” are obtained from the synonym dictionary 26, the keyword ID “image” of the search source keyword “image” is obtained. 02 ”is registered in the synonym table 22. For other search source keywords, synonyms are registered in the synonym table 22 in the same procedure.

[Document Management Department]
The document management unit 14 takes in, divides, edits, stores, and reads out document data.
That is, the document data is fetched from the document data fetching device 40 and stored in the document DB 25 together with the document ID. The captured document data is referred to as “original document data” for convenience. A serial number is used as the document ID.

  The document management unit 14 also divides the original document data based on a predetermined rule. The divided document data is referred to as “divided document data” for convenience. The division rule is performed by pattern extraction, for example. For example, in the case of a patent document, the pattern extraction detects a pattern starting from a black bracket with a paragraph number and starting from a number, a black bracket with a black number, a document, and immediately before the black bracket with the next paragraph number. By doing. The divided document data obtained as a result is the document data of the area identified by the paragraph number. In addition, you may make it divide | segment into a meaningful range manually.

The document management unit 14 assigns a division ID (for example, a serial number) to each divided document data obtained as described above, and registers it in the divided document table 23 together with the document ID of the original document data and the divided document data.
For example, if the document data identified by the document ID “001” is as shown in FIG. 4, the contents of the divided document table for the divided document data divided by the document dividing unit 14 are as shown in FIG. Become. In the example of FIG. 5, division IDs “1”, “2”... Are assigned to the divided document data divided from the document data with the document ID “001”, and are identified by a set of the document ID and each division ID. The contents of the divided document data are stored in association with each other.

  The document management unit 14 also adds position information to the head of each divided document data in the original document data, and registers the position information ID for identifying the position information in the divided document table 23 in association with the division ID. . The position information is an anchor tag when an HTML document is taken as an example, and the position information ID can be a name attribute of the anchor tag, for example, a character string concatenating the document ID and the division ID.

  The document management unit 14 further creates modified document data in which character strings appearing in the original document data and the divided document data are colored with color information registered in the keyword table 21. This function will be described later.

[Spectrum Management Department]
The spectrum management unit 15 checks whether a keyword exists for each divided document data in a predetermined order (for example, keyword registration order). Specifically, a spectrum display pattern is output and registered in the spectrum table 24 in the procedure shown in FIG.
That is, it is checked whether or not there is a keyword registered in the keyword table 21 for all divided document data (step S101). If the keyword exists (step S101: Yes), or the keyword does not exist, but the synonym for the keyword ID of the keyword exists in the synonym table 22 (step S101: No, S102: Yes), the keyword table 21 The color information associated with the keyword ID is output (step S103).
On the other hand, when neither a keyword nor a synonym exists (step S101: No, S102: No), an unused color is output (step S104). The unused color may be any color as long as it is not specified in the keyword table 21, but in this example, “black” is used for convenience of explanation. If keywords remain in all the divided document data, the process returns to step S101 (step S105: No), and the process ends when the search for all keywords is completed (step S105: Yes).

  As a result, a spectrum display pattern is generated for each divided document data in a sequence of color information registered in the keyword table 21. The spectrum management unit 15 registers the spectrum display pattern in the spectrum table 24 together with the document ID and the divided document ID.

FIG. 7 is a diagram showing a state before and after extraction of keywords (including synonyms) for the divided document data shown in FIG. In the illustrated example, in the divided document data (paragraph 0015) with the division ID 1, the first keyword “recognition” 411 among the five types of keywords registered in the keyword table 21 appears, and is output. The color will be “red”. Since the “image” registered second in the keyword table 21 does not appear, the output color is “black”. Although the third keyword “failure” does not appear, but the synonym “trouble” 413 appears, the output color is “green” assigned to “failure”. Since the fourth registered keyword “sensor” and the fifth registered keyword “IC chip” do not exist in the divided document data, the output color is both “black”.
Also for the divided document data (paragraph 0015) with the same original document data and the division ID 2, the registered keyword “recognition” 411 appears first, and then the second registered keyword “image” 415 also appears. However, since there are no other keywords, the output color is “black”.

As a result, as shown in FIG. 8, a spectrum display pattern of “red, black, green, black, black” is registered in the spectrum table 24 for the divided document data with the division ID “1”, and the division ID “2” is registered. As for the divided document data “”, a spectrum display pattern “red, yellow, black, black, black” is registered.
When synonyms are used instead of keywords, the positions of the keywords and the synonyms in the spectrum display pattern may be the same as in this example, but they may be distinguished from each other.

[Display control unit]
The display control unit 16 performs control for causing the display device 50 to display document data (original document data and divided document data) and a spectrum display pattern. Specifically, the keyword registered in the keyword table 21, the spectrum display pattern registered in the spectrum table 24, and the document data stored in the document DB 25 are acquired, in addition to the highlighted display of the specified keyword, The spectral display pattern in the document data is set as one set so that one or a plurality of document data can be displayed on the display device 50 in parallel.

[Spectrum evaluation section]
The spectrum evaluation unit 17 calculates the distance between the spectrum display pattern in which synonyms of keywords registered in the keyword table 21 are arranged and the spectrum display pattern of the spectrum table 24 of each divided document data, and for each divided document data. The division ID of the divided document data giving the spectrum display pattern with the shortest distance is specified.
As an example of the distance scale, for example, for two character strings having the same number of characters, a Hamming distance that gives the number when the characters at corresponding positions are different as the distance may be used. Taking FIG. 2 as an example, the division ID of the division document that gives the spectral display pattern having the shortest distance from the red, yellow, green, violet, and blue spectral display pattern is specified. The identified division ID is transmitted to the display control unit 17 together with the corresponding spectrum display position.
In addition, you may employ | adopt the form which notifies a user the spectrum position most similar to a designated keyword.

[Operation]
Next, the operation of the document management system 1 in this embodiment will be described.
Here, two patent documents are simultaneously displayed on the display device 50, and the area of the other patent document (second document data) in which contents similar to the description contents of one patent document (first document data) are described. An example of an interface in which the degree of relevance between patent documents can be visually grasped by highlighting on the display device 50 is shown.
FIG. 9 is an explanatory diagram of the procedure executed by the document management system 1 in this case, and FIG. 10 is a screen display example on the display device 50.

  The document management system 1 causes the display device 50 to display a screen for inputting two patent document names and keywords. The left side of FIG. 10 is an entry area 510 for these pieces of information. The input area 52 for file name 1 in Patent Document 1, the input area 54 for file name 2 in other Patent Documents, and a plurality (in the illustrated example, five areas). ) Keyword input areas 511 to 515 are formed. Unique color information is assigned in advance to the input areas 511 to 515 for each keyword.

  When two patent documents are designated by the operator, the document management system 1 takes in each designated patent document, assigns each document ID, and stores it in the document DB 25 (step S1).

  Thereafter, Patent Document 1 is displayed in the first document area 520 of FIG. 10, and Patent Document 2 is displayed in the second document area 540 (step S2).

  When an operator inputs five keywords into the entry areas 511 to 515 in descending order of characterizing the description contents of the patent document 1 with reference to the description contents of the claims column 521 of the displayed patent document The document management system 1 accepts these keywords and the color information assigned to the entry areas 511 to 515, and associates the keyword ID for identifying the keyword with the color information and the color information for each keyword. (Step 3).

  The fetched Patent Documents 1 and 2 are respectively divided according to a predetermined rule, and a search is performed as to whether or not a keyword registered in the keyword table 21 exists in each divided area. A spectrum display pattern in which color information registered in the keyword table 21 or the synonym table 22 is registered for the keyword if it exists, or unregistered color information (black) in the above-described order is generated in the above order. The display pattern is associated with the division ID and registered in the spectrum table 24 (step S4).

After dividing each patent document and registering it in the spectrum table 24, the document management system 1 arranges the patent documents 1 and 2 together with the relative position information of each divided document data and displays them on the display device 50 (step S5).
“Place the patent documents 1 and 2 in parallel with the relative position information of each divided document data” means that in the example of FIG. 10, the area adjacent to the first document area 520 is one divided document data of the patent document 1. The first spectrum area 530 for displaying the spectrum display pattern in one column for the number of divided document data, the second document area 540 adjacent to the first spectrum area 530, and one adjacent divided document data for Patent Document 2 Are arranged as a second spectral region 550 for display for displaying the same number of divided document data as the number of divided document data. Elements 531 to 535 of the spectrum display pattern are formed in the first spectral region 530, and the same number of elements 551 to 555 are formed in the second spectral region 550.

  The arrangement of the spectrum display patterns in each spectrum region 530, 550 is such that the division ID and the sequence corresponding to the appearance order of the designated keyword intersect so that the keyword location in the divided document data can be known. For example, relative position information of the divided document data is arranged on the vertical axis of the display device 50, and the elements 531 to 535 and 551 to 555 of the spectrum display pattern are arranged two-dimensionally on the horizontal axis. At this time, a path to the position information of the corresponding divided document data is set for each division ID. Taking an HTML document as an example, a hyperlink (for example, an anchor tag) is set to divided document data corresponding to the division ID.

Thereafter, the spectrum evaluation unit 17 discriminates the similarity between each color information pattern registered in the keyword table 21 and all the spectrum display patterns registered in the spectrum table 24, and associates it with the spectrum display pattern having the highest similarity. The divided area of the divided ID is specified. Then, among the divided document data of the specified divided area, the keyword registered in the keyword table 21 is highlighted with color information unique to the keyword, and the spectrum registered in the spectrum table 24 in association with the division ID. The display pattern is highlighted on the display device 50 together with the relative position information of the divided area identified by the division ID (step S6).
For example, in the case of an HTML document, keyword highlighting is performed by coloring by specifying color information with a font attribute value. On the other hand, the highlighting of the relative position information of the divided areas is performed by coloring in another color so that the positions 536 and 556 separated by the rectangular lines in FIG. 10 can be identified, or by superimposing the cursor.

Thus, for example, the first spectral region 530 characterized by five keywords is considered to be highly related to the description content of the specific region 521 of Patent Document 1 which is known to exist in the third divided region 536. According to the second spectral region 550, the described part of the patent document 2 is the specific region 541 of the patent document 2 that can be located in the first half region as a whole, and the spectrum display pattern of the specific region 541 is a divided region 556. Can be confirmed.
Therefore, since the comparison of the relationship between Patent Documents 1 and 2 can be grasped visually and quickly, the time required for the document comparison can be significantly shortened.

  In the screen example of FIG. 10, when a specific spectrum display pattern or element is selected, the corresponding divided document data may be displayed. For example, when the spectrum corresponding to the document ID “001” and the division ID “1” is selected with a mouse or the like, the document display position is displayed in accordance with the division ID “1” when the document ID “001” is displayed. Anyway.

  Further, when the user gives an instruction to the divided document data of Patent Documents 1 and 2 in which the spectrum is displayed as color information corresponding to each predetermined keyword on the display device 50, the document management unit 14 performs the division. Each keyword of the document data may be output to the external storage device 20 as a file such as an HTML file colored with a corresponding color and stored in an area not shown. Thereby, by creating the modification data colored with the color information stored in the keyword table 21, the comparison result can be easily referred to later.

As described above, according to the document management system 1 of the present embodiment, not only the attention area of a plurality of document data to be compared, but also related portions between the document data can be easily grasped by spectrum display, and has excellent browsing properties. An interface can be realized.
The document management system 1 according to the present embodiment performs synonym expansion using the synonym dictionary 26 when the designated keyword does not exist in the comparison target document data. The location can be easily grasped.

[Modification]
In the above description, an example in which both the first document data and the second document data are divided and two spectrum regions are formed in accordance with the number of document data has been described. However, the division is performed only on the second document data. You may make it do.
In the description of FIG. 9, the document data is divided after the keyword group and the unique color information are registered in the keyword table 21. However, this process may be reversed.
In this embodiment, the spectrum display pattern has been described with five color combinations. However, since this is because five designated keywords are used, the number of color combinations can be arbitrarily changed according to the number of keywords. It is possible.
In the above description, the display colors for the input areas 511 to 515 of each keyword are determined in advance. However, the user may set the display colors on the screen.
Furthermore, the synonym for the designated keyword may be set on the screen by the user, and only the designated synonym may be highlighted.

DESCRIPTION OF SYMBOLS 1 ... Document management system, 10 ... Computer main body, 11 ... Input / output (I / O) control part, 12 ... Keyword management part, 13 ... Synonym management part, 14 ... Document management unit, 15 ... spectrum management unit, 16 ... display control unit, 17 ... spectrum evaluation unit, 20 ... external storage device, 21 ... keyword table, 22 ... synonym table, 23 ... Divided document table, 24 ... Spectrum table, 25 ... Document DB, 26 ... Synonym dictionary, 30 ... Input device, 40 ... Document data capture device, 50 ... Display device 52... File name 1 input area 54... File name 2 input area 510... Entry area 511 to 515. Document area, 530 First spectral region, 540... Second document region, 550... Second spectral region, 531 to 535, 551 to 555... Spectrum display pattern element, 536 and 556. Information representing the location.

Claims (5)

A keyword that receives input of a plurality of keywords arranged in a predetermined order and color information associated with each keyword, and registers a keyword ID for identifying the keyword and the color information in the first table for each received keyword Management means;
The designated document data is divided into predetermined areas, and a search is performed to determine whether or not a keyword registered in the first table exists in each divided area. If there is a keyword, the keyword is stored in the first table. A spectral pattern in which registered color information or unregistered color information is arranged in the predetermined order is generated for each divided region, and the generated spectral pattern is associated with a divided ID for identifying the divided region. Spectrum management means to be registered in the second table,
When displaying the document data on a predetermined display device, the keyword registered in the first table in the document data is highlighted with color information unique to the keyword, and is divided into the second table. Control means for displaying the spectrum pattern registered in association with the ID together with the relative position information of the divided area identified by the divided ID;
A document management system comprising: Synonym management means for associating a synonym of the keyword registered in the first table with a keyword ID for the keyword and registering the synonym in the third table;
The spectrum management means includes
Creating a spectral pattern of a divided region in which the keyword registered in the first table does not exist but the synonym associated with the keyword ID of the keyword exists in the third table based on color information unique to the keyword;
The document management system according to claim 1. The similarity between each color information pattern registered in the first table and all the spectrum patterns registered in the second table is determined, and the division ID associated with the spectrum pattern having the highest similarity is determined. It further comprises a similar area search means for specifying a divided area.
The document management system according to claim 2. A method executed by a document management system connected to a display device, comprising:
Displaying the first document data and the second document data on the display device;
A keyword for identifying a plurality of keywords arranged in a predetermined order for characterizing the description content of the displayed first document data and color information associated with each keyword and identifying the keyword for each received keyword Registering the ID and the color information in the first table in association with each other;
The second document data is divided into predetermined areas, and a search is made as to whether or not a keyword registered in the first table exists in each divided area. If there is a keyword, the keyword is stored in the first table. A spectral pattern in which registered color information or unregistered color information is arranged in the predetermined order is generated for each divided region, and the generated spectral pattern is associated with a divided ID for identifying the divided region. Registering in the second table,
The similarity between each color information pattern registered in the first table and all the spectrum patterns registered in the second table is determined, and the division ID associated with the spectrum pattern having the highest similarity is determined. Identifying a segmented area;
Spectral patterns registered in association with the division IDs in the second table, while highlighting the keywords registered in the first table in the document data of the identified divided regions with color information unique to the keywords. And displaying on the display device together with relative position information of the divided area identified by the division ID,
Document management method by spectral pattern. A computer having a first table and a second table;
A plurality of keywords arranged in a predetermined order and input of color information associated with each keyword are received, and for each received keyword, a keyword ID for identifying the keyword and the color information are associated and registered in the first table. Keyword management means,
The designated document data is divided into predetermined areas, and a search is performed to determine whether or not a keyword registered in the first table exists in each divided area. If there is a keyword, the keyword is stored in the first table. A spectral pattern in which registered color information or unregistered color information is arranged in the predetermined order is generated for each divided region, and the generated spectral pattern is associated with a divided ID for identifying the divided region. Spectrum managing means for registering in the second table, and
When displaying the document data on a predetermined display device, the keyword registered in the first table in the document data is highlighted with color information unique to the keyword, and is divided into the second table. A computer program that functions as display control means for displaying a spectrum pattern registered in association with an ID together with relative position information of a divided area identified by the divided ID.

Images