JPH1145289A - Document processor, storage medium storing document processing program and document processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a document processor capable of preparing a summary based on the preference of a user such as a utilization purpose or the like, a storage medium storing a document processing program and a document processing method. SOLUTION: A key word and the importance are obtained from the contents of a processing document in the past and a GP(group personalizing) matrix for which one of the plural users and the key word is turned to a row, the other is turned to a column and the importance of the respective key words to the respective users is turned to an element value is obtained. Important words (a), (b),... from a summary preparation object document and the importance from the appearing frequency or the like are obtained, a term vector V for which the importance is an element is shifted by the GP matrix and a preference term vector V' is obtained. Preference important sentences F(Z) are extracted from the summary preparation object document based on the element (=preference importance) of the preference term vector V', arranged in an appearing order in the summary preparation object document and turned to a preference summary.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document processing apparatus, a storage medium storing a document processing program, and a document processing method, and more particularly, to creation of an abstract based on user's preference such as purpose of use. .

[0002]

2. Description of the Related Art Conventionally, computers have been used to automatically create abstracts (including abstracts) for various documents such as books, papers, and reports. For the automatic summarization of documents, for example, “Extraction and processing of semantic information from full-text information” (Information Processing Society of Japan 38th Annual Conference Proceedings,
P. 222; 1989). In this method, an important word in a document is first extracted from information such as a character type and a verb, and the most important word is obtained from the appearance frequency of the important word. Next, by obtaining an important sentence based on whether or not the important word and the most important word appear, it is possible to automatically create a summary. Also, by reflecting the nature of paragraphs in sentences,
A method described in Japanese Patent Application Laid-Open No. 3-191475 for more accurately preparing an abstract has also been proposed.

[0003]

However, even in the same document, if the purpose of use, such as for business use or technical materials, or the user's preference is different, differences occur in important parts in the document. Then, even if an abstract is created by the conventional document processing as described above, there is a problem that an abstract based on the user's preference cannot be obtained.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and has been made in consideration of the above-described circumstances. Storage medium storing
And a document processing method.

[0005]

According to a first aspect of the present invention, there is provided a document acquisition unit for acquiring a document including a plurality of sentences;
An important word and phrase extracting means for acquiring an important word and its importance from the document acquired by the document acquiring means; and a preference important part for selecting a preference important part reflecting user's preference from the document based on the important word. The above object is achieved by providing a document processing apparatus comprising: a selection unit; and a preference summary creating unit that creates a summary of the document based on the preference important part selected by the preference important part selection unit. . The invention described in claim 2 is the first invention.
In the document processing device described in the above, the important word extraction means, from the document acquired by the document acquisition means, the candidate word of the important word and a candidate word acquisition means for acquiring the importance thereof, and represents the user's preference A preference vector having the importance values of a plurality of keywords as element values, or one of a plurality of users and a plurality of keywords representing the preferences of each user in a row, and the other as a column, the importance of each of the keywords with respect to each of the users. A preference matrix for acquiring a GP matrix as an element value, and using the preference vector or the GP matrix acquired by the preference acquisition module, the candidate phrase acquired by the candidate phrase acquisition module. The important words are extracted from the importance shifted in importance, and the preference important part selecting means, based on the importance words and their importances, By providing the document processing device to select the main components, to achieve the above object. According to a third aspect of the present invention, in the first aspect of the present invention, the important word extracting unit obtains the candidate word of the important word and its importance from the document acquired by the document acquiring unit. Extracting the important phrase according to the importance of the candidate phrase, and the preference important part selecting means includes a preference vector having an importance value of a plurality of keywords representing the user's preference as an element value, or a plurality of users and a plurality of users. A preference matrix for acquiring one of a plurality of keywords representing preferences of each user in a row and the other as a column, and a GP matrix in which the importance of each keyword for each user is an element value; Using the preference vector or the GP matrix obtained by the means, the importance obtained by shifting the importance of the important words obtained by the important word extraction means is changed to Ri to achieve the object by providing a document processing apparatus for selecting the critical parts. The invention according to claim 4 is a document acquisition function for acquiring a document including a plurality of sentences, an important word extraction function for acquiring an important word and its importance from the document acquired by the document acquisition function, A preference important part selecting function for selecting a preference important part reflecting user's preference from the document based on the important word; and a summary of the document based on the preference important part selected by the preference important part selecting function. The above object is achieved by providing a storage medium storing a computer-readable document processing program for causing a computer to implement a preference summary creation function. According to a fifth aspect of the present invention, in the storage medium according to the fourth aspect, the important word / phrase extraction function acquires the important word candidate words and their importance from the document acquired by the document acquisition function. A candidate word acquisition function and a preference vector with the importance values of multiple keywords representing user preferences as element values, or one row of multiple users and multiple keywords representing each user's preference as rows and the other as columns A preference acquisition function for acquiring a GP matrix with the importance of each of the keywords for each user as an element value,
Using the preference vector or the GP matrix acquired by the preference acquisition function to extract the important phrase from the importance shifted from the importance of the candidate phrase acquired by the candidate phrase acquisition function. The preference important portion selection function achieves the above object by providing a storage medium storing a document processing program for selecting the preference important portion based on the important word and its importance. According to a sixth aspect of the present invention, in the storage medium according to the fourth aspect, the important word extraction function acquires the candidate word of the important word and its importance from the document acquired by the document acquisition function. The important phrase is extracted based on the importance of the candidate phrase, and the preference important part selecting function is a preference vector having the importance of a plurality of keywords representing the user's preference as an element value, or a plurality of users and a plurality of users. A plurality of keywords representing preferences of each user in a row, and the other as a column, a GP matrix having element values of the importance of each keyword for each user as element values, Using the preference vector or the GP matrix acquired by the acquisition function to shift the importance of the important words acquired by the important word extraction function Document processing program for selecting the critical parts to achieve the above object by providing the stored storage medium by. The invention according to claim 7 acquires a document composed of a plurality of sentences, acquires an important word and its importance from the acquired document, and reflects a user's preference from the document based on the important word. The object is achieved by providing a document processing method for selecting a preference important part and creating an abstract of the document based on the selected preference important part.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of a document processing apparatus, a storage medium storing a document processing program, and a document processing method according to the present invention will be described below in detail with reference to FIGS. I do. (1) Overview of Embodiment In this embodiment, a keyword and its importance are acquired from the contents of a past processed document, and one of a plurality of users and the keyword is set as a row, and the other is set as a column. A GP matrix with the importance as an element value is obtained. .., And the importance g (p), g (q),... From the appearance frequency and the like are obtained from the summary creation target document, and the term vector V = (g
(P), g (q),...) Are shifted by a GP matrix to obtain a preference term vector V ′. Elements of preference term vector V '(= preference importance) g' (p), g '
Based on (q),..., The preference important sentences are extracted from the summary creation target document, and are arranged in the order of appearance in the summary creation target document to obtain a preference summary.

(2) Details of the Embodiment FIG. 1 shows an embodiment of a document processing apparatus according to the present invention, and a computer from which the document processing program according to an embodiment of the present invention is read. FIG. 2 is a block diagram showing the configuration of FIG. As shown in FIG. 1, the document processing apparatus (computer) includes a control unit 11 for controlling the entire apparatus. The control unit 11 includes a keyboard 12 and a mouse 13 as input devices, a display device 14, a printing device 15, a storage device 16, a storage medium driving device 17, a communication control device 18 via a bus line 21 such as a data bus. , And an input / output I / F 19 and a character recognition device 20 are connected. The control unit 11 includes a CPU
111, a ROM 112, and a RAM 113. R
The OM 112 is a read-only memory in which various programs and data for the CPU 111 to perform various controls and calculations are stored in advance.

The RAM 113 is a random access memory used as a working memory by the CPU 111. In the RAM 113, the target document storage area 1 is used as an area for performing the preference summarizing process according to the present embodiment.
131, summary parameter storage area 1132, important word storage area 1133, term vector storage area 113
4. A matrix storage area 1135, a preference term vector storage area 1136, a summary storage area 1137, and other various areas are secured.

[0009] The target document storage area 1131 stores a document for which a preference summary is to be created. The summary parameter storage area 1132 stores the value of the summary parameter obtained by input from the operator or the like or the default value of the summary parameter read from the data storage unit 163 described later. As the summary parameter input by the operator, for example, the ratio of the summary to all documents (1 to 99%)
, Date / time, price information, quantity priority such as physical quantity (size, weight, temperature, etc.) / None, URL (Uniform Resou
rce Locator) weighted long sentence with / without priority /
Values such as whether to select or not to be used are stored. The term vector storage area 1134 stores the term vector of the document for which the preference summary is created, which is acquired in the present embodiment. In the summary storage area 1135, the important sentences acquired according to the present embodiment are stored in the order of the preference summary creation target document.

The keyboard 12 has various keys such as a kana key and a numeric keypad for inputting kana characters, a function key for executing various functions, a cursor key, and the like. The mouse 13 is a pointing device, and is an input device for designating a corresponding function by left-clicking a key, an icon, or the like displayed on the display device 14.
As the display device 14, for example, a CRT or a liquid crystal display is used. The display device 14 displays the contents of a document for which a preference summary is to be created, the preference summary created by the present embodiment, and the like. Printing device 15
Is for printing a sentence displayed on the display device 14, a document stored in the document database 165 of the storage device 16, and the like. Various printing apparatuses such as a laser printer, a dot printer, an ink jet printer, a page printer, a thermal printer, and a thermal transfer printer are used as the printing apparatus.

The storage device 16 comprises a readable and writable storage medium and a drive device for reading and writing various information such as programs and data on the storage medium.
As a storage medium used for the storage device 16, a hard disk is mainly used, but a readable and writable storage medium among various storage media used in a storage medium driving device 17 described later may be used. Good. The storage device 16 stores the kana-kanji conversion dictionary 161 and the program storage unit 1
62, data storage unit 163, important word database 16
4. Document database 165, matrix database 16
8, other storage units not shown (for example, storage units for backing up programs, data, and the like stored in the storage device 16) and the like. In the program storage unit 162, in addition to various programs such as a preference summary creation processing program in the present embodiment, a kana-kanji conversion program for converting a kana character string input using the kana-kanji conversion dictionary 161 into a kanji mixed sentence. Etc. are stored. The data storage 163 stores various data such as default values of summary parameters. As the default value of the summary parameter, for example,
Summary ratio for all documents = "25%", date and time,
Importance of quantity such as price information, physical quantity (size, weight, temperature, etc.) = "No" or URL (Uniform Resource Locato)
r) emphasis = "no", long single sentence emphasis = "no",
A value such as "/" / "/""=" No "is stored.

In this embodiment, the keyword database 164 stores, in the present embodiment, a keyword (processing keyword) obtained based on a document processed during a predetermined period in the past, and the importance of the keyword (processing keyword). (Processing importance) are stored in correspondence with each other. The document database 165 includes a document created by the kana-kanji conversion program and a storage medium drive 17 and a communication controller 1 created by another device.
8 is stored. The format of each document stored in the document database 165 is not particularly limited, and a text format document, HTML (Hype
Documents in various formats, such as rTextMarkup Language) format documents and JIS format documents, can be stored. Further, in the document database 165, the user who has processed the document and the number of times of processing have been stored in association with each document.
The value of the processing count is reset to 0 every predetermined period.

The matrix database 168 stores matrices Ga, Gb, and Gc obtained based on the contents of the document processing performed in a predetermined period in the past. These matrices G
A GP (Group Personalize) matrix is acquired from a, Gb, and Gc, and the importance of important words (including phrases) of the document to be summarized is shifted (importance is converted) by the GP matrix. 2A to 2C are explanatory diagrams illustrating an example of the matrices Ga, Gb, and Gc.

As shown in FIG. 2A, the matrix Ga is a matrix in which processing important words extracted from processed documents processed within a predetermined period in the past are arranged in rows and the processed documents are arranged in columns. The element represents the processing importance f (x) of the processing important word. As shown in FIG. 2B, the matrix Gb is a matrix in which the processed document is taken in rows and the user is taken in columns, and each element is
This is the number of times that the user has processed each document within the predetermined period. In the matrix Gc, as shown in FIG. 2C, both the row and the column indicate the importance coefficient of each user. The matrix Ga and the matrix Gb are rewritten every predetermined period, and the matrix Gc is appropriately rewritten by an input from the operator.

The storage medium drive 17 is a drive for the CPU 111 to read a computer program or data including a document from an external storage medium. The computer programs stored in the storage medium include programs for various processes executed by the document processing apparatus of the present embodiment, and dictionaries and data used therein. Here, the storage medium refers to a storage medium in which a computer program, data, and the like are stored, and specifically, a magnetic storage medium such as a floppy disk, a hard disk, and a magnetic tape, and a semiconductor storage medium such as a memory chip and an IC card. A storage medium such as a CD-ROM, an MO, a PD (phase change rewritable optical disk) or the like, from which information can be read optically, and a paper such as a paper card or a paper tape (and a medium having a function equivalent to the paper) are used. It includes a storage medium and a storage medium in which a computer program or the like is stored by various methods. As a storage medium used in the document processing apparatus of the present embodiment, a CD-ROM or a floppy disk is mainly used. Storage medium drive 17
In addition to reading computer programs from these various storage media, it is possible to write data and the like stored in the RAM 113 and the storage device 16 to a writable storage medium such as a floppy disk.

In the document processing apparatus of this embodiment, the control unit 1
One CPU 111 reads a computer program from an external storage medium set in the storage medium drive 17 and stores (installs) it in each unit of the storage 16. When executing various processes such as similarity calculation according to the present embodiment, the corresponding program is
The data is read into the AM 113 and executed. However, not from the storage device 16 but the storage medium drive device 17
It is also possible to read the program directly from the external storage medium into the RAM 113 and execute it. Also, depending on the document processing apparatus, the preference summary creation processing program of the present embodiment or the like is stored in the ROM 112 in advance, and this is stored in the CPU 111.
May be executed.

The communication control device 18 is capable of transmitting and receiving various types of documents such as text format and HTML format and various data such as bitmap data to and from other personal computers and word processors. I have. The input / output I / F 19 is an interface for connecting various devices such as a speaker that outputs audio, music, and the like. The character recognition device 20 is a device for recognizing characters written on paper or the like in various formats such as a text format or HTML, and is configured by an image scanner, a character recognition program, and the like.

In this embodiment, in addition to a document created by an input operation on the keyboard 12 (stored in a predetermined storage area of the RAM 113), a document created externally and stored in a predetermined storage medium is read from the storage medium driving device 17. Documents, documents stored in advance in a document database, documents downloaded from the communication control device 18, and the character recognition device 2.
It is possible to acquire various documents such as a document whose characters have been recognized as 0 as a target document (document acquisition means).

Next, one embodiment of the document processing method according to the present invention, which is a preference summarizing process performed by the document processing apparatus having the above-described configuration, will be described with reference to FIGS.

In this embodiment, a new processing important word and a new processing importance are acquired for each predetermined period based on the contents of the document processing performed during the predetermined period, and the matrices Ga and Ga in the matrix database 168 are obtained. The matrix Gb is rewritten.

FIG. 3 is a flowchart showing the operation of the matrix Ga, Gb rewriting process. The CPU 111 sequentially obtains documents (processed documents) processed within a predetermined period from the document database 165, stores them in a predetermined work area of the RAM 113 (step 11), and sets keywords (including processing key words (including phrases)) for each document. )) And its importance (processing importance) (step 12).

FIG. 4 is a flowchart showing the operation of the processing important word / processing importance acquisition processing for each document. As shown in FIG. 4, the CPU 111 extracts a self-contained word from the processed document by performing morphological analysis on the document acquired from the document database 165 (step 12).
Along with 1), candidate words (phrases) including noun phrases, compound noun phrases, etc. are extracted from the processed document (step 122). Next, the processing importance f (x) of each candidate word (phrase) is obtained from the appearance frequency of the extracted candidate word (phrase) in the processing document and the evaluation function (step 123). Here, as the evaluation function,
For example, when a predetermined important word is specified in advance, weighting for the important word, weighting based on the type of candidate word (phrase) such as a word, a noun phrase, a compound noun phrase, and the like are used.

Further, the CPU 111 obtains processing important words a, b, c,... From the candidate words (phrases) based on the obtained value of the processing importance f (x) (step 124). a, b, c,... and their processing importances f (a), f
(B), f (c)... Are stored in the important word database 164. When the processing important words and the processing importance thereof are obtained for all the processing documents, the process returns to the matrix Ga, Gb rewriting processing routine shown in FIG.

Next, the CPU 111 operates the matrix database 1
68, the processing important words a, b, c,... In rows, the processing documents in the predetermined period in columns, and the processing importance f
(X) is rewritten to take each element (Step 1)
3). Then, the CPU 111 sets the document database 16
5, the number of times of processing of each document is obtained (step 14).
The matrix Gb is rewritten with the number of processes obtained from the document database 165 as elements, with the processed documents within a predetermined period as rows, and the matrix Ga, Gb rewriting process ends.

FIG. 5 is a flowchart showing the main operation of the preference summary creation processing. In the summary creation process,
The CPU 111 acquires a document for which an abstract is to be created (summary target document) and stores it in the target document storage area 1131 of the RAM 113 (step 21). The document to be summarized is stored in the RAM 113, the document database 165 of the storage device 16, the storage medium drive 1
7 or from the communication control device 18. Then CP
U111 obtains an input value when the user inputs a summary parameter from the keyboard 12 or the like, and obtains a default value of the summary parameter stored in the data storage unit 163 when there is no input by the user. It is stored in the parameter storage area 1132 (step 2
2).

Next, the CPU 111 obtains a term vector V for the digest target document stored in the target document storage area 1131 (step 23). FIG. 6 is a flowchart showing the operation of the term vector acquisition processing. CP
The U111 first performs a morphological analysis to extract independent words included in the document to be summarized (step 231), and also extracts candidate words (phrases) including noun phrases, compound noun phrases, and the like from the document to be summarized and RAM 113. (Step 232). Next, the objective importance g (y) is determined from the summary parameters stored in the summary parameter storage area 1132 of the RAM 16, the frequency of appearance of the extracted candidate words (phrases) in the document to be summarized, the evaluation function, and the like. (Step 233). Here, as the evaluation function, for example, when a predetermined important word is specified in advance, weighting for the important word, weighting according to the type of a candidate word (phrase) such as a word, a noun phrase, a compound noun phrase, and the like are used. used.

Then, important words p, q, r,... Are obtained based on the objective importance g (y) (step 234).
Objective importance g (p), g of important words p, q, r, ...
A term vector V having (q), g (r),... As elements is obtained (step 235), and the process returns to the preference summary creating routine shown in FIG.

Subsequently, the CPU 111 acquires the matrix Ga from the matrix database 168, and performs dimension matching between the term vector V and the matrix Ga (step 24). That is,
The number of dimensions of the term vector V and the number of rows of the matrix Ga are defined as the number of union of the important words of the document to be summarized and the rows of the processing important words represented by the rows of the matrix Ga. The element value of Ga and the element value of the term vector V for the key word included only in the matrix row are defined as “0”.

For example, if the important word of the document to be summarized is “important,
Important word, importance, ... ", the processing important word represented by the row of the matrix Ga is" important, ..., politics, ... ", and the term vector V = (1, 18, 19, ...) of the document to be summarized, matrix G
If one column of a is (18,..., 21,...)
When the dimensions are matched, the term vector V =
(1, 18, 19, ..., 0, ...), and one column of the matrix Ga is (18, 0, 0, ..., 21, ...). The matrix Ga and the term vector V after the dimension matching are
The data is stored in the matrix storage area 1135 and the term vector storage area 1134 of the RAM 113.

Subsequently, the CPU 111 calculates the matrices Gb, Gc
Is obtained from the matrix database 168, and a GP matrix is obtained from the matrices Ga and Gb and Gc whose dimensions have been adjusted (step 25). The GP matrix is obtained according to the following equation. GP = Ga.Gb.Gc Therefore, the GP matrix in the present embodiment is a row where the dimension matching of the Ga matrix is taken as it is, and each member of the user is a column, and each element of the GP matrix is a member. The numerical value is obtained by adding the importance of each member to the processing importance f (x) of the processing important word in each past document processing.

When the GP matrix is obtained, the CPU 1
11 calculates a GP vector based on the GP matrix (step 26).

FIG. 7 is an explanatory diagram conceptually explaining a process of calculating a GP vector from a GP matrix. CPU11
First, the average value of the elements of each element of each element gij of the GP matrix (i = 1 to the number of members m, j = 1 to the number k of the union of the key word and the processing key word of the document to be summarized) is calculated. Calculation to obtain a column vector (total GP vector) (FIG. 7 (1) →
(2)). In this total GP vector, the frequency of appearance of each element gi in the past document processing in the entire user group for each important word (however, the weight of a predetermined important word of each important word and the importance of members are taken into account) ) Is reflected. The CPU 111 further calculates the total GP
Each element gi of the vector is divided by the total number of document processing times to obtain a GP vector in one column (FIG. 7 (2) →
(3)). As described above, dividing the total GP vector by the total number of document processing times prevents the matrix Gb from including the document processing number as an element, and prevents the GP vector from increasing as the processing number increases. That's why.

Then, the CPU 111 obtains a preference term vector V 'by multiplying each element of the GP vector by the element of the term vector V corresponding to each element (step 27). Each element of the preference term vector V ′ is a preference importance g ′ (y) obtained by weighting the user's preference for the term to the objective importance g (y).

Subsequently, the CPU 111 acquires the preference part importance (favorite sentence importance F (Z)) included in the document to be summarized based on the preference importance g ′ (y) of the important word (step 28). Then, the summarization ratio of the summarization parameter (for example, the total number of sentences in the target summarization document) is ranked from the top of the portion (sentence) with the highest preference part importance (favorite sentence importance F (Z)) of each part (each sentence) (Sentence) within the top 25% of
Is listed as a preference important part (taste important sentence) (step 29), and the listed sentences are arranged in the order of appearance in the summary target document to be a preference summary of the summary target document, and the summary is stored in the RAM 113. Area 113
7 (step 30), and terminates the preference summary creation processing according to the present embodiment.

As described above, in this embodiment, the user's preference for the important word is grasped based on the frequency of appearance in the past document processing and the like, and the objective importance g ( y) is converted into a preference importance g ′ (y) weighted by reflecting the user preference, and an important sentence is acquired based on the preference importance g ′ (y) to create a summary. Therefore, according to the present embodiment, a summary reflecting the user's preference is created. According to this embodiment,
Term vector V with objective importance of key words as elements
Is obtained and converted using a GP matrix that reflects the user's preferences, thereby obtaining a preference term vector V ′ having the preference importance as an element. The present invention can be easily applied to a general document processing apparatus having a core engine adopting the standard.

According to the present embodiment, the GP matrix for shifting the term vector V to the preference term vector V ′ is a matrix Ga, Gb,
Since it is obtained by multiplying Gc, it is possible to easily construct a GP matrix taking into account various features and shift the term vector V. According to the present embodiment, the GP matrix for shifting the term vector V to the preference term vector V ′ is obtained by dividing a group including a plurality of users into several groups because each column reflects the interest of the user. A GP matrix of a group or a GP matrix (vector) of individual users can be easily obtained. According to the present embodiment, the matrix Ga, the GP matrix of which is rewritten every predetermined period based on a document processed in the past by the user,
Since the term vector V is obtained based on Gb and Gc, the term vector V is shifted to the preference term vector V ′ corresponding to the change of the user's preference over time, and the preference summary following the change of the user's preference is created. .

It should be noted that the present invention is not limited to the above-described embodiment, and can be appropriately changed without departing from the gist of the present invention. In the above embodiment, a computer is used as the document processing apparatus. However, the present invention is not limited to the computer, and may be a word processor or the like.

It is also possible to acquire the preference importance for all the important word candidates acquired from the document to be summarized, and acquire the important words from the important word candidates based on the preference importance. If you want to get preference importance based on objective importance,
Preference importance can also be obtained by performing appropriate processing for each objective importance without vectorizing the objective importance. Further, even when the objective importance is vectorized, the method of converting the term vector into the preference term vector does not need to use the GP matrix. As the candidate word extracting means and the important word acquiring means, a processing important word extracting means for extracting a processing important word from the document for which the summary is to be created can be used.

In the above-described embodiment, the GP matrix includes the number of past document processings (matrix Ga) for each user, the frequency of appearance of key words in each document (matrix Gb), and the importance of each user (matrix Gc). However, it may be obtained based only on the number of past document processings (matrix Ga) for each user and the frequency of appearance of the key words in each document (matrix Gb). In addition, for example, it may be obtained in consideration of the processing time of each document, the number of cases cited in creating another document, and the like. Further, when the GP matrix is obtained by multiplication of matrices such as the matrices Ga to Gc in the same manner as in the above-described embodiment, the elements of the matrices such as the matrices Ga to Gc indicate the frequency of occurrence of the important word in the document. Or a number that reflects the number of times a user has processed each document,
It does not need to directly represent the frequency of appearance or the number of processing itself.

In the above embodiment, the matrices Ga to Gc
Is acquired from the contents of past document processing, but may be acquired from the user and stored in the matrix database 168. In the above-described embodiment, the matrices Ga to Gc are rewritten every predetermined period. However, the matrices Ga to Gc may be appropriately rewritten every document processing or by a judgment of an operator or the like. A GP vector display means for displaying GP vectors on a display device may be provided so that the entire group of users and the user's preferences can be visually grasped. in this case,
GP vectors may be stored in a matrix database or a dedicated GP vector database in chronological order so that changes over time can be grasped. In the above-described embodiment, the importance is compared in sentence units according to the preference importance of important words, and the preference important sentence is selected as the preference important part. A preference important paragraph or a preference preference title may be selected as a part.

[0041]

As described above, according to the present invention,
For the important words in the document to be summarized, the preference importance based on the user's preference is obtained, the important part is selected based on this preference importance, and the summary is created from this important part. Preference such as user's interest, degree of attention, and purpose is reflected.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a computer which is an embodiment of a document processing apparatus of the present invention and which reads a document processing program of the present invention from an embodiment of a storage medium storing the program.

FIG. 2 shows matrices Ga, Gb, Gc in the embodiment of FIG.
It is explanatory drawing which shows an example of.

FIG. 3 is a flowchart showing an operation of matrix Ga, Gb rewriting processing in the embodiment of FIG. 1;

FIG. 4 is a flowchart showing an operation of processing important word / processing importance acquisition processing for each document in the embodiment of FIG. 1;

FIG. 5 is a flowchart showing a main operation of a preference summary creation process in the embodiment of FIG. 1;

FIG. 6 is a flowchart illustrating an operation of a term vector acquisition process in the embodiment of FIG. 1;

FIG. 7 is an explanatory diagram conceptually illustrating a process of acquiring a GP vector from a GP matrix in the embodiment of FIG.

[Explanation of symbols]

 11 control unit 112 ROM 113 RAM 1131 target document storage area 1132 summary parameter storage area 1133 important word storage area 1134 term vector storage area 1135 matrix storage area 1136 preference term vector storage area 1137 summary storage area 12 keyboard 13 mouse 14 display device 15 print Device 16 Storage device 161 Kana-Kanji conversion dictionary 162 Program storage 163 Data storage 164 Key word database 165 Document database 168 Matrix database

Claims (7)

[Claims] 1. A document acquisition unit for acquiring a document composed of a plurality of sentences; an important word extraction unit for acquiring an important word and its importance from the document acquired by the document acquisition unit; Preference important part selecting means for selecting a preference important part reflecting user's preference from the document, and preference summary creating means for creating a summary of the document based on the preference important part selected by the preference important part selecting means A document processing apparatus comprising: 2. An important word / phrase extracting unit, comprising: candidate word / phrase obtaining unit for obtaining a candidate word of the important word and its importance from the document obtained by the document obtaining unit; and a plurality of keywords representing user preference. The preference vector with the importance of the element value as an element value, or one of a plurality of users and a plurality of keywords representing the preference of each user in a row, and the other as a column, the importance of each of the keywords with respect to each of the users as an element value. A preference matrix that acquires a candidate matrix acquired by the candidate acquisition module using the preference vector or the GP matrix acquired by the preference acquisition module. Extracting the important words from the shifted importance, the preference important part selecting means selects the preference important parts based on the important words and their importance The document processing apparatus according to claim 1, characterized in Rukoto. 3. The important phrase extracting unit acquires candidate words of the important word and its importance from the document acquired by the document acquiring unit, and extracts the important word based on the importance of the candidate word. The preference important portion selecting means may include a preference vector in which importance values of a plurality of keywords representing user preferences are used as element values,
Or, preference acquisition for acquiring a GP matrix in which one of a plurality of users and a plurality of keywords representing the preferences of the plurality of users is row and the other is a column, and an importance value of each keyword for each user is an element value. Means, using the preference vector or the GP matrix acquired by the preference acquisition means, to select the important part according to the importance shifted from the importance of the important words acquired by the important word extraction means. The document processing apparatus according to claim 1, wherein: 4. A document acquisition function for acquiring a document including a plurality of sentences; an important word extraction function for acquiring an important word and its importance from the document acquired by the document acquisition function; A preference important part selection function of selecting a preference important part reflecting user's preference from the document; and a preference summary creation function of creating a summary of the document based on the preference important part selected by the preference important part selection function. And a computer-readable document processing program for causing a computer to realize the above. 5. The important word extracting function includes: a candidate word acquiring function for acquiring candidate words of the important word and its importance from the document acquired by the document acquiring function; and a plurality of keywords representing user preferences. The preference vector with the importance of the element value as an element value, or one of a plurality of users and a plurality of keywords representing the preference of each user in a row, and the other as a column, the importance of each of the keywords with respect to each of the users as an element value. And a preference acquisition function of acquiring the GP matrix. The preference vector or the GP matrix acquired by the preference acquisition function is used to determine the importance of the candidate phrase acquired by the candidate phrase acquisition function. Extracting the important words from the shifted importance, the preference important part selecting function selects the preference important parts based on the important words and their importance Storage medium document processing program according to claim 4, characterized in Rukoto is stored. 6. The important word extracting function acquires candidate words and important values of the important words from the document acquired by the document acquiring function, and extracts the important words based on the importance of the candidate words. The preference important part selection function includes a preference vector having the importance values of a plurality of keywords representing the user's preference as element values,
Or, preference acquisition for acquiring a GP matrix in which one of a plurality of users and a plurality of keywords representing the preferences of the plurality of users is row and the other is a column, and an importance value of each keyword for each user is an element value. Having a function, using the preference vector or the GP matrix acquired by the preference acquisition function, selecting the important part according to the importance shifted from the importance of the important word acquired by the important word extraction function A storage medium storing the document processing program according to claim 4. 7. A document including a plurality of sentences is acquired, an important word and its importance are acquired from the acquired document, and a preference important part reflecting user's preference is reflected from the document based on the important word. A document processing method comprising: selecting and creating a summary of the document based on the selected preference important part.