JPH10254876A - Document summarizing method and its device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To summarize a document by a system according to the needs of a user by selecting a summarizing system to execute according to the style of a summarizing objective document and the collecting purpose of the document. SOLUTION: A corresponding relation analyzing part 7 starts the document style of the summarizing objective document inputted from a document input part 2 to retrieve a summarizing system corresponding to the sorted document style at a corresponding relation storing part 6, which stores a corresponding relation an operator previously set-inputs at a corresponding relation input part 5. A document summarizing part 11 reads the summarized content of the retrieved summarizing system from a summarizing system storing part 9 and summarizes the summarizing objective document by the read summarized contents. A summary generating part 14 generates a summary based on an output condition stored in an output condition storing part 13 and the generated summary is outputted from a summary output part 15. The part 13 stores an output condition the operator previously set-inputs at an output condition input part 12.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document summarizing apparatus for summarizing a document to be summarized according to a preset summarization method.

[0002]

2. Description of the Related Art With the spread of personal computers and word processors that handle document data, a large number of document summarizing apparatuses for summarizing and outputting input documents have been proposed. In particular, the development of information networks such as the Internet has made it possible to easily collect a large amount of document data. There is an increasing demand for a document summarization device that can be easily handled.

For this reason, the conventional document summarization apparatus extracts a significant paragraph from a summarization target document based on a method of extracting the first few lines of the summarization target document and an evaluation of a preset frequency of words and the like. Any one of a summary method and a method of creating a table of contents of the document to be summarized based on the analysis result of the document structure is provided.

[0004] For example, in a document summarizing apparatus of a method of summarizing based on the evaluation of the appearance frequency of a word, a word set by a user is searched in each paragraph of the document to be summarized, and the word is set by the user in advance. Paragraphs that exist more than
Alternatively, a paragraph in which the word is present most often is extracted as a summary document, displayed on a display in a predetermined format, or printed by a printer.

[0005]

However, many of the conventional document summarizing apparatuses are designed to execute a single summarizing method. Even if a plurality of summarizing methods can be executed, a detailed summarizing method for each summarizing method is used. Conditions are fixed, and in such a fixed summarization method, it is often impossible to output a summary document according to a user's request. That is, the style of the document to be summarized and the purpose of collecting the document by the user are diversified, and an appropriate summarization method is different for each style and purpose. Therefore, depending on the fixed summarization method, all of the user's requirements can be satisfied. I can't let it.

For example, when listing all items of a large number of collected document data, a summarization method for extracting uppercase portions or emphasized portions in the document is suitable, and when collecting information on specific items, Is suitable for summarization, which extracts paragraphs containing the keyword specified in advance, and when reading only the conclusion, the summarization method for determining the conjunction at the beginning of each paragraph is suitable. This depends on the style of the document to be summarized and the purpose of collecting the document.

As described above, in the conventional document summarizing apparatus which can execute only a single summarization method when the processing content of the user changes according to the style of the document to be summarized and the purpose of collecting the document, the optimum summarization is performed. There is a problem that a document cannot be summarized by a method and a summary document required by a user cannot be output.

An object of the present invention is to select a summarization method to be executed according to the format of a document to be summarized and the purpose of collecting the document, and to summarize the document in a method according to the needs of the user. It is to provide a device.

[0009]

According to a first aspect of the present invention, an input of a document to be summarized is received, and a summarizing method is selected. A single or plural summarizing methods are selected from a plurality of summarizing methods stored in advance. The method is characterized in that it receives an input, summarizes a document to be summarized according to a single or a plurality of selected summarization methods, and outputs a summary document.

[0010] The invention according to claim 8 is a document input unit for receiving an input of a document to be summarized, a summary method storage unit for storing a plurality of summary methods, and a summary document output unit for outputting a summary document after the summary processing. And input of a selection of one or more summarization methods from among the plurality of summarization methods stored in the summarization method storage means, and input of a document according to the single or plural summarization methods selected and input by the summarization method selection means. And a control unit for summarizing the summarization target document input from the means and outputting the summarization document via the summarization document output unit.

According to the first and eighth aspects of the present invention, a document to be summarized is summarized by one or a plurality of summarization methods selected and input from a plurality of summarization methods stored in advance. Therefore, the document to be summarized is summarized by the summarization method selected by the user.

According to a second aspect of the present invention, a correspondence relationship between each of a plurality of document formats to be classified and a single or a plurality of summarization methods is stored, and an input summary target document is converted into any of a plurality of formats. Classification is performed, and a summarization process is performed by selecting one or a plurality of summarization methods based on the classification result and the correspondence stored in advance.

According to a ninth aspect of the present invention, each of a plurality of document formats classified by the document classifying means and a single or a plurality of summarization methods of the plurality of summarization methods stored in the summarization method storage means are used. A correspondence storage unit for storing the correspondence, wherein the control unit classifies the summary target document input from the document input unit into one of a plurality of formats, and the classification result and the correspondence storage unit The present invention is characterized in that a single or a plurality of summarization methods are selected from a summarization method storage unit based on the stored correspondence and a summarization process is executed.

According to the second and ninth aspects of the invention, the input summary target document is classified into one of a plurality of document formats, and based on the classified document format and the correspondence stored in advance. Single or multiple summarization schemes are selected. Therefore, the document to be summarized is summarized by the summarization method according to the document format.

According to a third aspect of the present invention, an input for setting the correspondence between the style of the document to be summarized and the summary method is received,
It is characterized by storing a correspondence set and input in advance. The invention according to claim 10 is characterized in that the correspondence storage unit receives an input for setting the correspondence between the style of the document to be summarized and the summary method.

According to the third and tenth aspects of the present invention, the document to be summarized is summarized by a classified document format and a summarization method selected based on the correspondence set by the user in advance.

According to a fourth aspect of the present invention, the correspondence between the format of the document to be summarized and the summary method is set based on the history of the correspondence between the document format of the document to be summarized and the summary method used for the summary processing. And storing it.

According to an eleventh aspect of the present invention, the control unit determines the format of the document to be summarized and the summarization method based on a history of the correspondence between the document format of the document to be summarized and the summarization method used for the summarization process. Is set and stored in the correspondence storage unit.

According to the fourth and eleventh aspects of the present invention, the document to be summarized is summarized by the summarization method selected based on the classified document format and the history of the correspondence between the document format and the summarization method. You.

The invention described in claim 5 is characterized in that a detailed summary condition setting input is received for each of a single or a plurality of summarization methods, and the set and input summary condition is stored.

According to a twelfth aspect of the present invention, a summary condition input unit for receiving a detailed summary condition setting input for each of a single or a plurality of summary methods, and a summary condition set and input in the summary condition input unit are stored. And a summary condition storage unit.

According to the fifth and twelfth aspects of the present invention, a single or a plurality of summarization methods are stored, and detailed summarization condition setting input is received for each of the summarization methods, and the set summarization conditions are stored. I do. Therefore, the document to be summarized is summarized according to the preset summary condition.

The invention described in claim 6 is characterized in that a plurality of input documents are summarized by the same summarization method.

According to a thirteenth aspect of the present invention, the document input unit accepts input of a plurality of documents to be summarized, and the control unit summarizes the plurality of documents input from the document input unit using the same summarization method. It is characterized by.

According to the sixth and thirteenth aspects of the invention, a plurality of documents to be summarized are summarized at once by the same summarization method.

[0027] The invention described in claim 7 is characterized in that setting input of output conditions for determining an output state of a summary document is received, and a summary document is created based on the set input condition.

According to a fourteenth aspect of the present invention, the control unit receives an input of setting of an output condition for determining an output state of the summary document, and creates a summary document based on the set and input output condition. And

According to the seventh and fourteenth aspects of the present invention, a summary document created based on preset output conditions is output.

[0029]

FIG. 1 is a block diagram showing the configuration of a document summarizing apparatus according to an embodiment of the present invention.
The document summarizing apparatus 1 is actually a processing program for realizing the functions shown in FIG. 2 mounted on hardware such as a personal computer having the configuration shown in FIG. 1 via a storage medium such as a CD-ROM. is there. Therefore, the actual document summarizing apparatus 1 is a C
A keyboard 24, a mouse 25, a display 26, a printer 27, a data line terminating device (hereinafter, referred to as DCE) 28 such as a modem and a terminal adapter, an FDD and an HDD are connected to the PU 21 via interfaces 23a to 23g.
Is connected to the external storage device 29.

FIG. 2 is a functional block diagram of the document summarizing apparatus. The document summarizing apparatus 1 includes a document input unit 2, a summary method input unit 3, a summary condition input unit 4, a correspondence input unit 5, a correspondence storage unit 6, a correspondence analysis unit 7, a summary method selection unit 8,
Summarization method storage unit 9, summarization condition storage unit 10, document summarization unit 1
1, an output condition input unit 12, an output condition storage unit 13, a summary document creation unit 14, and a summary document output unit 15.

The document input unit 2 receives the input of the text of the document to be summarized and the attached information. The additional information input to the document input unit 2 includes property data such as a storage location of the document, a title, a date and time, a creator, a sender, an addressee, and a document length. The document input unit 2 may receive input of a plurality of documents to be summarized in one input process.

The summarization method input unit 3 receives an input for setting a summarization method to be executed on the document to be summarized. The summarization condition setting unit 4 accepts setting input of summarization conditions for determining the content of the summarization process in each summarization method. Here, the summarization conditions are, for example, "short sentences and greeting sentences are not subject to summarization processing", "short sentences are sentences with 5 words or less", and "greeting sentences are" take care ". Or "Sentences containing the character string" Thank you "", "Extract 5 sentences containing frequently used words", "Word" PQ
R "means a more detailed summarization process, such as" extract a paragraph including "", "" extract an emphasized sentence ", or" the emphasized sentence is a sentence that is distinguished from other sentences by decorative characters. "

The summarization method includes “heading list”,
Summarization processing such as "Creating a table of contents", "Graph overview", or "Gathering information about XYZ Company". Each summarization method is a set of one or more of the above-mentioned summarization conditions. It is set to. In this summarization method, a specific name may be given by combining one or more arbitrary summarization conditions among the summarization conditions described above.

The summarizing method input unit 3 displays a plurality of summarizing methods 31 on the display 26 shown in FIG. 1, for example, on the initial screen D1 or D1 'shown in FIG. 3 (A) or (B).
Alternatively, 31 'is displayed in a list, and a summary mode setting input to be executed by an operation of the keyboard 24 or the mouse 25 by the operator is received. Initial screen D in FIG. 3 (A)
Reference numeral 1 denotes a display example in a case where the summarization process is included in a part of another application software as one of the menus. The initial screen D1 'in FIG. It is an example of a display when it is stored. In addition,
When there are many executable summarization methods, these are displayed, for example, in the display area 41 of the list screen D2 shown in FIG.
A predetermined number of summarization methods displayed on the initial screen D1 or D1 'of the summarization process shown in (A) or (B) may be made selectable.

The summary condition input unit 4 includes a display 26
Then, for example, condition setting screens D3 to D7 shown in FIGS. The condition setting screens D3 to D7 are shown in FIG.
Is displayed by selecting the setup menu 32 on the initial screen D1 or by operating the setup key 32 'on the initial screen D1' in FIG. 3B. On the condition setting screen D3, regarding the summarization method described in the display area 51, a sentence having a number of words of 6 or less in the first setting unit 52, a sentence composed of many characters other than the alphabet,
It is set to exclude the first five lines from the summary target,
It is set that the highlighted part is extracted together with the next line in the second setting unit 53.

A condition setting screen D4 shown in FIG.
Is displayed by operating the key 54 in the setting screen D3, and accepts the setting of the definition of the greeting excluding from the summarization target. The condition setting screen D5 shown in FIG.
It is displayed by operating the key 55 in the setting screen D3, and accepts the setting of the definition of the important part to be extracted. The condition setting screen D6 shown in FIG. 7A is displayed by operating the key 56 in the setting screen D3, and accepts the setting of the definition of the emphasized portion. The condition setting screen D7 shown in FIG. 7B is displayed by selecting the display content 63 in the setting screen D5, and accepts setting of an important word.

The correspondence input unit 5 receives an input for setting the correspondence between the document format of the document to be summarized and the summary method.
Here, the document format refers to a feature for classifying a document to be summarized in relation to each summary method, and is classified by, for example, property data such as a title, a creator, a sender, or an address. Therefore, in general, the document format of the document to be summarized can be generally specified by whether or not the specific property data includes a predetermined character string. The document format of the document to be summarized can be specified not only by property data but also by word distribution information, character decoration information, character structure information, keyword distribution information, or the like.

The correspondence input unit 5 is connected to the display 2
6 displays, for example, a relation setting screen D8 shown in FIG. This relation setting screen D8 is the initial screen D1 or D
It is displayed by selecting the menu 33 in 1 'or operating the button 33'. On the relationship setting screen D8, the setting unit 82 receives the additional setting of the correspondence between the document format and the summary method, and the setting unit 83 accepts the deletion of the already set correspondence. The display unit 8 displays the correspondence between the currently set document format and the summary format.
Displayed at 1.

The correspondence storage unit 6 stores the correspondence between the document format set and input in the correspondence input unit 5 and the summary method. This correspondence is determined by, for example, property data, character decoration information, or character structure information that can be used to specify the document format. For example, “the summarization method M is used for the summarization target document including the character string ABC in the sender data. , "Use summary method O for documents to be summarized that include a character string PQR in a font different from the others in underlined parts." Or "Summary for documents to be summarized that have no more than five paragraphs within 10 lines." Use method N ".

The correspondence analysis unit 7 performs the document format and the summarization method based on the history of the document format of the document to be summarized input from the document input unit 2 and the summarization method used for the summarization processing of the document to be summarized. Analyze the correspondence with. The analysis result of the correspondence analysis unit 7 is, for example, a correspondence storage unit in a format such as “A summary method Ia is frequently used for a summary target document including a character string XYZ in the title, and the summary method Ib is rarely used”. 6 and present it to the operator.

The summarization method selection unit 8 classifies the document format of the document to be summarized input by the document input unit 2 and reads out the summarization method corresponding to the classified summarization method from the correspondence storage unit 6. As described above, the document format of the document to be summarized can be specified by whether or not a specific character string is included in specific property data, word distribution information, character decoration information, character structure information, keyword distribution information, and the like. The summarization method selection unit 8 classifies the document format of the summarization target document by searching for a predetermined character string in specific property data of the summarization target document.

The summarization method storage unit 9 stores a plurality of executable summarization methods. The summarization scheme storage unit 9 may store some of the summarization schemes that are frequently used among the plurality of summarization schemes in a hierarchical manner at a higher level than the other summarization schemes.

The summary condition storage section 10 stores the summary conditions set and input in the above-described summary condition input section 4. The summarization condition storage unit 10 may be configured integrally with the summarization method storage unit 9.

The document summarizing section 11 reads out the summarization scheme set and input by the summarization scheme input section 3 or the summarization scheme determined by the summarization scheme determination section 8 from the summarization scheme storage section 9 and the summarization condition storage section. The summary condition storage unit 10 reads out the summary condition, summarizes the summary target document input from the document input unit 2 in accordance with the summary method and the summary condition, and outputs the summary result to the summary document creation unit 14.

The output condition input unit 12 receives an input for setting the output format of the summary document. Here, the output format refers to one or a plurality of parameters for adjusting the appearance of the output state of the summary document. This output format includes the document summarization section 1
1 can be included.

The output condition input section 12 displays, on the display 26, condition setting screens D9 and D10 shown in FIGS. 9 and 10, for example. On the setting screen D9, 1
When sequentially displaying two summary documents, an input for setting items to be displayed together with the summary documents is received. The setting screen D10 accepts setting of a condition for determining a display order when a plurality of summary documents are displayed at the same time.

The output condition storage unit 13 stores the output condition input unit 1
The output conditions set and input in 2 are stored. The output condition storage unit 13 may store output conditions corresponding to each document format of the document to be summarized.

The summary document creating unit 14 is configured to output the output condition
3 according to the output conditions read from
A summary document is created from the summary result in 1 and supplied to the summary document output unit 15. The summary document output unit 15 displays or prints out the summary document supplied from the summary document creation unit 14.

The summary document output unit 15 has a display 26
The output screen D11 shown in FIG. 11 or FIG.
Alternatively, D12 is displayed. On the output screen D11, the items set on the setting screen D9 described above are displayed on the display unit 111, and the summary document is displayed on the display unit 112. On the output screen D12, a plurality of summary documents are displayed on the display unit 121 in an order based on the conditions set on the above-described setting screen D10.

Of the units shown in FIG. 2, the correspondence storage unit 6, the summary method storage unit 9, the summary condition storage unit 10, and the output condition storage unit 13 are configured in the external storage device 29 shown in FIG. . Further, the summary document output unit 15 includes the display 26 and the printer 27 shown in FIG. 1, and displays the summary document on the display 26 and prints out the summary document.

The correspondence analysis unit 7, the summary method selection unit 8, the document summary unit 11, and the summary document creation unit 14 are configured in the CPU 21 shown in FIG. The CPU 21 executes processing for realizing the functions of the correspondence analysis section 7, the summary method selection section 8, the document summary section 11, and the summary document creation section 14 according to a processing program installed in advance, and the keyboard 24 To control the I / O devices. Further, the CPU 21 transmits, via the DCE 28 and the public telephone network 30, other data communication terminals, databases,
It is connected to a server of a data communication network such as the Internet.

The CPU 21 accepts an input of a document to be summarized via the DCE 28 and the public telephone network 30. The document is input from a storage medium such as a floppy disk or a document read by a reading device such as a scanner. Can be used as the document to be summarized. Also, by including the DCE 28 in the summary document output unit 15 and including the transmission destination of the summary document in the output condition set and input in the output condition input unit 12, the summary document is transmitted to another data communication terminal or the like. It may be.

FIG. 13 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the first and eighth embodiments of the present invention. The CPU 21 first reads the summary target document input to the document input unit 2 (s1), and initializes the initial screen D shown in FIG. 3A or 3B.
1 or D1 ', and reads the digest condition set in advance for the digest method from the digest condition storage unit 10 (s
2). Next, the CPU 21 reads out the summarization method set and input in the summarization method input unit 3 from the summarization method storage unit 9, and summarizes the document to be summarized by this summarization method (s
3). Thereafter, the CPU 21 outputs the summary result to the display 26 or the printer 27 (s4). further,
The CPU 21 confirms whether or not to change the summarization method and the summarization target document (s5, s6). If the summarization method is changed, the process returns to s2. If the summarization target document is changed, the process returns to s1. Return.

By executing the processing shown in FIG.
The document to be summarized can be summarized by a single or a plurality of summarization methods selected from a plurality of summarization methods executable in the document summarization apparatus 1. Can be summarized and output. Note that the reading of the summarization target document in s1 and the reading of the summarization method in s2 include not only reading of the summarization target document and the summarization method one by one, but also a plurality of readable summarization target documents and
A list of executable summarization methods can be displayed on the display 26, and one summarization target document and summarization method can be read based on the operator's selection. In addition, regardless of whether the processing of s1 or s2 is performed before or after, the processing of s5 and s6 is not essential. Further, prior to the processing of s5 and s6, it may be confirmed whether or not to repeat the summarization processing.

FIG. 14 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the second and ninth embodiments of the present invention. After reading the document to be summarized (s21), the CPU 21 classifies the document format of the document to be summarized (s22), and searches the correspondence storage unit 6 to determine whether or not there is a correspondence between the classified document formats. (S23). If there is a correspondence between the classified document formats in the correspondence storage unit 6, the CPU 21 reads out the summarization method from the summarization method storage unit 9 based on the correspondence and summarizes the summarization target document (s24). If there is no correspondence between the classified document formats in the correspondence storage unit 6, the CPU 21 executes a preset summarization method (s26). Thereafter, the CPU 21 ends the process after judging whether or not the operator has selected to change the document to be summarized (s25). Note that the processing of s25 is not essential.

By executing the processing shown in FIG.
A summary method suitable for the document format of the document to be summarized input from the document input unit 2 is automatically selected, and a summary document suitable for the document format of the document to be summarized can be created. In addition,
The case where the correspondence of the classified document format does not exist in the correspondence storage unit 6 in s23 includes a case where the document to be summarized cannot be classified into any document format. Also, s26
In place of the above processing, a display for prompting the operator to input the setting of the summary method may be performed.

FIG. 15 is a flowchart showing a processing procedure of a document summarizing method and a document summarizing apparatus according to another embodiment of the present invention. CPU
21. First, after accepting the setting of a plurality of summarization methods on the initial screen D1 shown in FIG. 3A or 3B (s31), the document to be summarized is read (s32), and then the contents of the method counter are read. Is set to the set number of summarization schemes (s33). Thereafter, the CPU 21 executes the summarization process by sequentially changing the summarization mode until the content of the mode counter becomes "1" (s34 to s37). When the contents of the system counter become "1" in s36, the CPU 21 checks whether or not to change the document to be summarized (s38). When the operator desires to change the document to be summarized, the process returns to s32.

By executing the processing shown in FIG. 15, it is possible to sequentially execute the summarization processing using a plurality of summarization methods on a single summarization target document. Note that, in the output of the summary document in s35, all the summary documents may be output collectively after all of the plurality of summary processes for the document to be summarized are completed. Also, the processing of s31 and s
Regardless of whether it is before or after the process of 32, the process of s38 is not essential.

FIG. 16 is a flowchart showing a processing procedure at the time of inputting correspondence in the document summarizing method and the document summarizing apparatus according to the third and tenth embodiments of the present invention. When the selection of the menu 33 or the operation of the button 33 ′ is performed on the displayed initial screen D 1 or D 1 ′, the CPU 21 firstly stores the correspondence stored in the correspondence storage unit 6 into, for example, the setting shown in FIG. The display 26 is displayed on the display 26 in the state shown in the screen D8 (s41), and a new correspondence is added by the operator on the setting screen D8, or a part or all of the correspondence stored in the correspondence storage 6 is displayed. Accept deletion (s42, s
43). When the operator additionally sets a new correspondence, the CPU 21 determines whether or not the newly set correspondence has already been stored in the correspondence storage 6, and then stores it in the correspondence storage 6 (s44). , S45), when the operator selects the deletion of the correspondence, the correspondence stored in the correspondence storage 6 is deleted (s46).
The CPU 21 continuously executes the processing of s41 to s46 until the operator inputs the end of the correspondence setting processing (s47).

By executing the processing shown in FIG.
For example, "The title includes" daily report "and the length is 1
For documents with zero or more lines, the correspondence between the document format and the summarization method such as "Use the abstraction method"tomorrow's task extraction "" can be set in advance, and the summarization method can be summarized according to the actual situation of the operator. Can be. In the confirmation process of s44, it is determined whether the additionally set correspondence does not conflict with the already stored correspondence, or whether the additionally set correspondence is not included in the already stored correspondence. You can also check. Alternatively, the processing shown in FIG. 16 may be performed only when the correspondence relationship of the input summary target document with respect to the document format is not stored in the correspondence storage unit 6.

FIG. 17 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the embodiments of the present invention. First, the CPU 21 reads the document to be summarized (s51), and classifies the document format of the document to be summarized (s52). Then, CPU2
1 analyzes the correspondence history of the classified document formats (s53). This history is, for example, “A summary method“ headline breaking news ”was used ten times in a document containing“ XYZ breaking news ”in the title”. This is realized by counting the number of times cumulatively. Therefore, in the process of s53, the CPU 21 searches the correspondence storage unit 6 for the correspondence between the classified document formats.

If there is one or more correspondences for the classified document formats in the correspondence storage unit 6, the CPU 21
Selects, for example, the summarization method that is used most frequently for the document format (s54, s55). If there is no correspondence in the correspondence storage unit 6 for the classified document format, the CPU 21 displays a list of executable summarization methods on the display 26 (s56), and accepts selection of the summarization method by the operator (S56). s57) The correspondence between the selected summarization method and the classified document format is newly stored in the correspondence storage unit 6 (s58). After this, CPU2
1 reads the summarization method selected in s55 or s57 from the summarization method storage unit 9 and executes the summarization process (s5).
9) Output a summary document (s60). Furthermore, CPU
21 updates the history of the correspondence between the document format of the document to be summarized and the summary method used this time (s61), and ends the process after confirming whether to change the document to be summarized (s6).
2). Note that the process of s62 is not essential.

By executing the processing shown in FIG.
According to the history of the correspondence between the document format of the summarization target document and the summarization method, the summarization target document can be summarized in the frequently used summarization method for each document format. Note that s
In the summarization method selection process in 55, if there is a correspondence relationship between the classified document format and a plurality of summarization methods, conditions other than the most frequently used condition, for example,
One summarization method may be selected on condition that the point of use is the latest. In s55, C
The summarization method selected by the PU 21 may be displayed on the display 26 to prompt the operator to confirm whether or not to use the summarization method.

FIG. 18 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the fifth and twelfth embodiments of the present invention. Initial screen D in FIG.
When the selection of the menu 32 or the operation of the button 32 'in 1 or D1' is performed, the CPU 21 displays one or a plurality of available summarization methods in, for example, the display area 41 of the list screen D2 shown in FIG. List display (s71),
The selection of a single summarization method by the operator (s72) and the selection of setting the summarization condition of the summarization method or deleting the summarization method are accepted (s73).
When the operator operates the button 42 or 43 on the list screen D2 and selects the setting of the summary condition, the CPU 21
Sequentially displays the setting screens D3 to D7 shown in FIGS. 5 to 7 on the display 26, continuously accepts the setting input of the summary condition by the operator (s74), and uses the set summary condition as the content of the summarization method. Summary condition storage unit 1
0 is stored (s75). When the operator operates the button 44 on the list screen D2 and selects the deletion of the digest scheme, the CPU 21 deletes the digest scheme from the digest scheme storage unit 9 (s76). The CPU 21 continuously executes the processing of s71 to s76 until the operator inputs the end of the processing (s77). Note that the process of s77 is not essential.

By executing the processing shown in FIG.
Summary information for determining the processing content in each summary method in more detail, for example, data on search words such as appearance frequency and appearance position, data on extracted words such as underlining and enlarged characters, and property data such as title and creator By setting, the document to be summarized can be summarized by a summarization method optimal for the needs of the operator. In addition, when the summary condition is stored in s75, it is inconsistent with the summary condition having the same content as the summary condition according to the current setting, the summary condition including the summary condition according to the current setting, or the summary condition according to the current setting. It may be checked whether or not the summary condition to be executed is already stored. Further, the set summarization condition may be stored in the summarization method storage unit 9 as a part of each summarization method. FIG. 19 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the embodiment of the present invention. First, the CPU 21
After reading a plurality of documents to be summarized (s81),
The summarization method is determined (s82), and the content of the document counter is set to the number of documents to be summarized (s83). After this,
The CPU 21 executes the summarization process by sequentially changing the summarization target document until the content of the document counter becomes "1" (s84).
To s87). When the content of the document counter becomes "1" in s86, the CPU 21 confirms whether or not to change the summarization method (s88). When the operator desires to change the summarization method, the process returns to s82. Note that the process of s88 is not essential.

By executing the processing shown in FIG. 19, it is possible to collectively execute the summarization processing using the same summarization method for a plurality of summarization target documents. Note that s8
The determination of the summarization method in 2 can be made by the setting input by the operator or based on the classification result of the document format of the document to be summarized. In addition, in the output of the summary document in s85, all the summary documents may be output collectively after all the summarization processes for the plurality of summary target documents are completed.

FIG. 20 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the embodiment of the present invention. When the CPU 21 completes the summarization process, it reads out the output conditions stored in the output condition storage unit 13 (s91), and creates a summary document according to the read out output conditions (s92). CPU
The display 21 displays the created summary document on the display 26 in a state shown in, for example, the output screen D11 or D12 in FIG. 11 or FIG. 12 (s93), and receives a selection input by the operator as to whether or not to change the output condition. (S9
4). For example, when the operator selects the change of the output condition by operating the button 113 or 123 on the screen D11 or D12, the setting screen D9 or D10 shown in FIG. indicate.

By executing the processing shown in FIG. 20,
The summary document created in a format desired by the operator can be output, and the utility value of the summary document can be increased. Note that when there are a plurality of summary results by the processing shown in FIG. 15 or FIG. 19, the processing of s91 to s94 can be continuously performed. In this case, the processing of s93 is performed for a plurality of summary documents at once. You can do it.
Further, the summary document displayed on the display 26 in s93 may be printed out by the printer 27 in accordance with an instruction of the operator. Further, as described above, a condition for transmitting the summary result as communication data such as an e-mail to another communication terminal device can be set as the output condition.

[0069]

According to the first and eighth aspects of the present invention, a document to be summarized is summarized by a single or a plurality of summarization methods selected and input from a plurality of summarization methods stored in advance. It is possible to select a summarization method to be executed according to the format of the target document and the purpose of collecting the documents, and there is an advantage that the document can be summarized by a method according to the needs of the user.

According to the second and ninth aspects of the present invention,
The input summarization target document is classified into one of a plurality of document formats, and the document format is selected by selecting one or a plurality of summarization formats based on the classified document format and the correspondence stored in advance. The document to be summarized can be summarized by the summarization method according to the above.

According to the third and tenth aspects of the present invention, the document to be summarized can be summarized by the classified document format and the summarization method selected based on the correspondence set by the user in advance.

According to the fourth and eleventh aspects of the present invention, a document to be summarized is summarized by a summarization method selected based on the classified document format and the history of the correspondence between the document format and the summarization method. Can be.

According to the fifth and twelfth aspects of the present invention, a single or a plurality of summarization methods are stored, and the summarization conditions set for each of the summarization methods are stored. Thus, the document to be summarized can be summarized.

According to the sixth and thirteenth aspects of the present invention, a plurality of documents to be summarized can be summarized at once by the same summarization method.

According to the seventh and fourteenth aspects of the present invention, it is possible to output a summary document created based on preset output conditions.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a document summarizing apparatus as an example of an embodiment of the present invention.

FIG. 2 is a functional block diagram of the document summarizing apparatus.

FIG. 3 is a diagram showing an example of an initial screen displayed on a display of the document summarizing apparatus.

FIG. 4 is a diagram showing an example of a setting screen of a summarization method displayed on a display of the document summarization apparatus.

FIG. 5 is a view showing an example of a summary condition setting screen displayed on the display of the document summarization apparatus.

FIG. 6 is a diagram showing an example of a setting screen of the summary condition.

FIG. 7 is a diagram showing an example of a setting screen of the summary condition.

FIG. 8 is a diagram showing an example of a setting screen related to a correspondence between a document format and a digest format displayed on a display of the document digest device.

FIG. 9 is a view showing an example of an output condition setting screen displayed on a display of the document summarizing apparatus.

FIG. 10 is a diagram showing an example of a setting screen of the output condition.

FIG. 11 is a diagram showing an example of a summary document output screen displayed on a display of the document summarization apparatus.

FIG. 12 is a diagram showing an example of an output screen of the summary document.

FIG. 13 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the first and eighth embodiments of the present invention.

FIG. 14 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the embodiment of the present invention.

FIG. 15 is a flowchart showing a processing procedure of a document summarizing method and a document summarizing apparatus according to another embodiment of the invention described in claims 1, 2, 8 and 9;

FIG. 16 is a flowchart showing a processing procedure at the time of a correspondence input process in the document summarizing method and the document summarizing apparatus according to the embodiments of the present invention.

FIG. 17 is a flowchart showing a processing procedure of a document summarizing method and a document summarizing apparatus according to an embodiment of the present invention.

FIG. 18 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the embodiments of the present invention.

FIG. 19 is a flowchart showing a processing procedure of a document summarizing method and a document summarizing apparatus according to the embodiment of the invention described in claims 6 and 13;

FIG. 20 is a flowchart showing a processing procedure of the document summarizing method and the document summarizing apparatus according to the embodiment of the present invention.

[Explanation of symbols]

 1-document summarization device 2-document input unit 3-summarization method input unit 4-summary condition input unit 5-correspondence input unit 6-correspondence storage unit 7-correspondence analysis unit 8-summarization method selection unit 9-summarization method Storage unit 10-Summary condition storage unit 11-Document summarization unit 12-Output condition input unit 13-Output condition storage unit 14-Summary document creation unit 15-Summary document output unit

Claims (14)

[Claims] 1. A method for receiving a selection input of a single or a plurality of summarization methods among a plurality of presumed summarization methods, summarizing a document to be summarized according to the single or a plurality of summarization methods selected and input, and summarizing. A document summarizing method characterized by outputting a document. 2. A method according to claim 1, wherein a correspondence relationship between each of a plurality of document formats to be classified and a single or a plurality of summary formats is stored, and a document to be summarized is classified into one of a plurality of formats.
2. The document summarization method according to claim 1, wherein a summarization process is executed by selecting one or a plurality of summarization methods based on the correspondence stored in advance. 3. The document summarizing method according to claim 2, wherein a setting input of a correspondence between the style of the document to be summarized and the summarization method is received, and the correspondence set and input in advance is stored. 4. The system according to claim 2, wherein the correspondence between the format of the document to be summarized and the summarization method is set and stored based on the history of the correspondence between the document format of the document to be summarized and the summarization method used for the summarization process. Document summarization method described in. 5. The document summarizing method according to claim 1, wherein a setting input of detailed summarizing conditions is received for each of one or a plurality of summarizing methods, and the set summarizing conditions are stored. 6. The document summarizing method according to claim 1, wherein the plurality of input documents are summarized by the same summarizing method. 7. The document summarizing method according to claim 1, wherein the setting input of an output condition for determining the output state of the summary document is received, and the summary document is created based on the output condition set and input. 8. A document input unit for receiving an input of a document to be summarized, a summary method storage unit for storing a plurality of summarization methods, a summary document output unit for outputting a summary document after summarization processing, and a summary method storage unit. A selection input of a single or plural summarization methods among a plurality of stored summarization methods is accepted, and the summarization object input from the document input means by the single or plural summarization methods selected and input by the summarization method selection means. A controller for summarizing a document and outputting a summary document via a summary document output unit. 9. A correspondence relation storing a correspondence relation between each of a plurality of document formats classified by the document classification means and one or a plurality of summarization methods among the plurality of summarization methods stored in the summarization method storage means. A storage unit, wherein the control unit classifies the summary target document input from the document input unit into one of a plurality of formats, based on the classification result and the correspondence stored in the correspondence storage unit. 9. The document summarizing apparatus according to claim 8, wherein one or more summarizing methods are selected from the summarizing method storage unit to execute a summarizing process. 10. The document summarizing apparatus according to claim 9, wherein the correspondence storage unit receives an input for setting a correspondence between the style of the document to be summarized and the summary method. 11. The control unit sets the correspondence between the format of the document to be summarized and the summarization method based on the history of the correspondence between the document format of the document to be summarized and the summarization method used for the summarization processing. The document summarizing apparatus according to claim 9, wherein the document summarizing apparatus stores the document in the correspondence storage unit. 12. A summary condition input unit for receiving a detailed summary condition setting input for each of a single or a plurality of summary methods, and a summary condition storage unit for storing the summary conditions set and input in the summary condition input unit. Claims 8 to 1 including
2. The document summarizing apparatus according to claim 1. 13. The document input unit accepts input of a plurality of documents to be summarized, and the control unit summarizes the plurality of documents input from the document input unit using the same summarization method.
13. The document summarizing apparatus according to any one of items 1 to 12. 14. The control unit according to claim 8, wherein the control unit receives an input of setting of an output condition for determining an output state of the summary document, and creates a summary document based on the set and input output condition.
3. The document summarizing apparatus according to any one of 3.