JPH07192010A - Document processing unit - Google Patents

Abstract

PURPOSE:To reduce the time till an object document is obtained by extracting a keyword group based on the retrieval result, arranging keyword based on the evaluation and allowing the user to select a keyword to be added among them. CONSTITUTION:When plural retrieval results are stored in a retrieval result storage section 104, whether or not a keyword is to be evaluated is inquired of the user. When the user give an instruction to implement keyword evaluation, a keyword evaluation section 106 implements keyword evaluation processing and a keyword display section 110 displays the keyword obtained as a result in the order of more appeared documents together with number of documents. Then, the user selects a keyword from the keyword group. This is executed by inputting the keyword by a keyword selection section 11. Then, a document including the keyword selected by the user is extracted and it is used for the retrieval result. Since the time to obtain an object document is reduced, the document processing unit with a document retrieval function used easily is obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document processing device, and more particularly, to a document processing device applicable to a full-text search device for searching a whole text.

[0002]

2. Description of the Related Art Due to the spread of document databases and the improvement of computer processing capability, a document retrieval apparatus for retrieving a document containing a designated keyword from a large amount of document databases has come to be used.

In recent years, in particular, in order to improve flexibility in searching, it is possible to specify a free keyword by searching the entire text instead of using the keyword attached to the document in advance. The full-text search method described above has come to be used. This method has the feature that anyone can search it because there is no restriction that any keywords must be used.

In the full-text search method, the target search result is rarely obtained by specifying the keyword once, and generally, the process of gradually narrowing down the search result is performed while adding the keyword and repeating the search. Is normal.

[0005]

[Problems to be Solved by the Invention]

(1) However, conventionally, when a user adds a keyword, there is a problem that it is difficult to know what kind of keyword should be given to effectively narrow down the search results. Therefore, it is often the case that an inappropriate keyword is given to narrow down the search results too much, or the search results cannot be narrowed down at all.

Therefore, a first object of the present invention is to eliminate the above-mentioned drawbacks and to provide a document processing apparatus constructed so that a desired document search can be performed easily and quickly.

(2) Further, documents satisfying the search condition designated by the user (hereinafter referred to as conforming documents) usually vary depending on field factors or temporal factors. For example, if it is a newspaper article, it is concentrated on a certain "face" or a certain "period". Therefore, when the documents are sequentially searched, the time intervals at which the matching documents are obtained vary. That is, during the retrieval process, many matching documents can be obtained at one time at one time, but no matching documents can be obtained at another time. This is inconvenient when it is desired to see the search results at any time.

Further, even when the final number of relevant documents is predicted at an intermediate stage from the number of documents retrieved so far and the number of relevant documents among them, the time interval for obtaining the relevant documents varies. If so, there is a problem that the prediction becomes inaccurate.

Therefore, a second object of the present invention is to provide a document processing apparatus which, in view of the above-mentioned drawbacks, eliminates the variation in the time interval for obtaining a conforming document and enables an easy-to-use document search. It is in.

(3) As described above, when performing a document search using the document search device, the search result is too small because the search condition is too strict, or the search result is too loose because the search condition is too loose. It is difficult to search for a desired document with one search, because there are often too many. Therefore, the searcher usually makes the search condition strict or loosens the search condition according to the amount of the search result and performs the search again. However, in this type of conventional device, all search processing is performed once, and the search condition is changed by the searcher or the device based on the amount of the search result. There is a drawback that performing the search process under the condition becomes useless. In particular, in a document processing device such as a full-text search device that requires a long time for a search process, this has been a cause of significantly increasing the time required for the entire search.

Therefore, a third object of the present invention is to solve the above-mentioned conventional drawbacks and to prevent the useless search processing even if the search conditions are too strict or too loose, so that the entire search is performed. It is to provide a document processing device that reduces the time.

[0012]

[Means for Solving the Problems]

(1) In order to achieve the first object of the present invention, a document database having a plurality of documents, search condition input means for inputting a condition for searching a document, and a document satisfying the search condition from the document database. Search means for searching by full-text search; first extracting means for extracting a keyword group from a document obtained as a result of the search by the searching means;
Aligning means for evaluating the keyword group and arranging the keywords according to the evaluation value, selecting means for selecting a specific keyword from the keyword group, and a document including the keyword selected by the selecting means as the search result. The second extraction means for extracting from is presented, and the keywords that can narrow down the search results of the full-text search are presented,
The search results are narrowed down by the presentation.

(2) In order to achieve the second object of the present invention, document saving means for saving a document, search condition input means for inputting a condition for searching a document, and the search condition from the document holding means. A search means for searching documents satisfying the above condition and a control means for controlling the order of searching the documents by the search means are provided so that the order of searching the documents is not biased.

(3) In order to achieve the third object of the present invention, a search target document holding unit for holding a document to be searched, a search condition holding unit for holding a search condition, and the search target document holding unit. Retrieval processing means for retrieving documents satisfying the retrieval conditions retained in the retrieval condition retaining means from the retrieval target documents retained in the means, and retrieval result retaining means retaining the results retrieved by the retrieval processing means A document processing apparatus having: a configuration including search continuation possibility determination means for determining whether or not the search process can be continued by predicting that the number of retrieved results will be too small or too large in the middle of the process. And Here, in addition, the search continuation possibility determination means creates a search condition change instruction when the search continuation possibility determination means determines not to continue the search, and the search condition change instruction creation means creates the search condition change instruction. It is preferable to have a search condition change instruction holding unit that holds the instruction to change the search condition.

[0015]

[Action]

(1) According to the present invention, a keyword group is extracted from a search result, the keywords are evaluated, the keywords are arranged according to the evaluation value, and a user selects a keyword to be added from among the keywords, thereby narrowing down the search result. be able to. Thus, it is possible to shorten the time required to obtain the target document.

(2) Further, according to the present invention, means for controlling the order of searching the documents is provided so that the order of searching the documents is not biased in terms of type and timing of the documents. It can be expected that the obtained time intervals will not vary.

(3) Further, according to the present invention, the search processing is interrupted when the search conditions are too strict or too loose by predicting that the number of search results will be too small or too large during the processing. Therefore, it is possible to eliminate unnecessary search processing and reduce the time required for the entire search.

[0018]

Embodiments of the present invention will now be described in detail with reference to the drawings.

Embodiment 1 FIG. 1 is a block diagram showing the configuration of an apparatus according to an embodiment of the present invention.

In FIG. 1, 101 is a search condition input section for inputting search conditions, 102 is a search processing section for searching documents satisfying the search conditions by full-text search, 103 is a document database having documents to be searched, and 104 is a search. Search result holding unit that holds the results obtained by the processing unit 102, 1
Reference numeral 05 is a search result output unit that outputs search results, 106 is a keyword evaluation unit that extracts and evaluates keywords from the search results, 107 is a morphological analysis processing unit that performs morphological analysis when extracting keywords, and 108 is a keyword and each keyword , A keyword holding unit that holds the document number of the document in which 109 appears, 109 is a document keyword holding unit that holds the keyword for each document, 110 is a keyword display unit that displays the keywords extracted and evaluated by the keyword evaluation unit 106, and 111 is It is a keyword selection unit for selecting a specific keyword from the keywords displayed on the keyword display unit.

FIG. 2 is a system configuration diagram in this embodiment. In the figure, 201 is a control memory for storing the control procedure shown in FIGS. 3 and 4. This may be ROM or RAM. Reference numeral 202 denotes a central processing unit that performs processing in accordance with the control means stored in the control memory 201. Reference numeral 203 denotes a memory, which includes a search result holding unit that holds search results, a keyword holding unit that holds keywords, and a document keyword holding unit that holds keywords for each document. A keyboard 204 is used by the user to input search conditions and selected keywords. A disk device 205 has a document database. 206 is a display, which may be a CRT or a liquid crystal display. This is used to display search results and keywords. 20
7 is a bus.

FIG. 3 is a flow chart showing the processing procedure of the operation in the apparatus shown in FIG. The operation of the embodiment of the present invention will be described with reference to the figure.

First, in step S301, the search processing unit 102 performs a search process for searching the document database 103 for a document satisfying the search condition input by the user from the search condition input unit 101. The search processing here may be performed by a generally known full-text search method. The document number of the obtained search result document is held in the search result holding unit 104 in step S302. Then step S
At 303, it is checked whether or not there are a plurality of search results held in the search result holding unit 104. If there is no stored search result or if there is only one search result, the process ends.

If there are a plurality of search results in step S303, the process moves to step S304 to inquire of the user whether or not to perform keyword evaluation. If the user has instructed not to perform keyword evaluation, the process ends.

In step S304, when the user gives an instruction to perform keyword evaluation, step S305.
The keyword evaluation process is performed in step S3, and the keywords obtained as a result are displayed in the keyword display unit 110 together with the number of documents in the order of the number of documents that appear. Since it is difficult to display all of them, it is preferable to display only a predetermined number from the top. The procedure of the keyword evaluation process will be described later.

Next, in step S306, the user is made to select a keyword from the keyword group. This is done by inputting from the keyword selection unit 111.
In step S307, the document including the keyword selected by the user is extracted and used as the search result. Then, the process returns to step S302 and the process is repeated.

FIG. 4 shows in detail the processing procedure of the keyword evaluation processing S305 among the processing procedure of the operation in the apparatus shown in FIG. The procedure of the keyword evaluation process will be described with reference to this figure.

First, in step S401, one document number is retrieved from the search result holding unit 104, and step S40
In step 2, it is checked whether the document has already registered the keyword. If the keyword is already registered, the process proceeds to step S406.

If the keyword is not registered, the document is retrieved from the document database 103, and step S4
At 03, morphological analysis processing is performed on the document. As the morphological analysis method, any generally known method such as the longest matching method or the minimum number of clauses may be used.

In step S404, keywords are extracted from all the words obtained as a result of morphological analysis, excluding unnecessary words (particles, auxiliary verbs, etc., which are predetermined) that are not extracted as keywords, and are duplicated in the document. The keywords that are shown are grouped together. In addition, keywords that already exist in the search conditions specified by the user are also excluded. Step S405
Then, for this document, the keywords appearing in the document are registered in the document keyword holding unit 109.

Next, in step S406, the document number of this document is registered in the keywords appearing in this document among the keywords in the keyword holding unit 108. If a keyword that does not exist in the keyword holding unit 108 exists in this document, the keyword is registered first, and then the document number is registered.

In step S407, it is checked whether or not an unprocessed document remains, and if it remains, the process returns to step S401, and the processing up to this point is repeated. If all the documents have been processed, the process moves to step S408.
In step S408, for all the keywords existing in the keyword holding unit 108, the keywords are sorted according to the number of registered documents (that is, the number of documents in which the keyword appears).

Next, a more specific description of the present embodiment will be given by showing an actual example.

The contents of the document keyword holding unit 109 are as shown in FIG. 5, for example. The document keyword holding unit lists the keywords that appear in each document. In this example, "publishing, CD-ROM, personal computer, ..." exists as a keyword in the document of document number 01011, and "personal computer, workstation, OS, ..." as a keyword in the document of document number 000168. Exists.

The contents of the keyword holding unit 108 are as follows.
For example, as shown in FIG. In the keyword holder,
For each keyword, list the document number of the document in which it appears. In this example, the keyword “personal computer” has document numbers 000011, 000168, 000487,
Exists in the document of “...,” and the keyword “word processor” has the document numbers 00254, 000614, 000713 ,.
Present in the document.

The keyword evaluation section 106 sorts the keywords existing in the keyword holding section 108 in descending order of the number of documents and displays them together with the number of documents in the keyword display section 110. This example is shown in FIG. In this example, the most common keyword is "PC"
(The number of documents is 243), and the second most abundant keyword is “publish” (the number of documents is 118). If the user selects the keyword “publish”, the number of documents will be narrowed down to 118.
By repeating the same process, the target document can be efficiently obtained.

Next, modified examples (different embodiments) of the above embodiment will be listed.

1. In the above embodiment, the words obtained by the morphological analysis were used as the keywords when the keywords were extracted, but the keywords may vary. To deal with this, a synonym dictionary / different notation word dictionary may be prepared, and the synonym / different notation word may be mapped to one type of keyword for keyword extraction. For example, when the keywords “computer”, “computer”, “computer”, and “computer” are obtained as a result of morphological analysis, these are referred to as “computer”.
Map to. Even in this case, the present invention can be implemented in exactly the same manner.

2. In the above-described embodiment, the keyword is evaluated using the number of documents that appear, but this may be another evaluation value. For example, a numerical value weighted by the number of times it appears in the document may be used instead of simply the number of documents that appear.

3. In the above embodiment, the keywords are displayed together with the number of documents when displayed, but the ratio of narrowing down the number of documents may be indicated instead of the number of documents. Further, the keywords and the number of documents may not be shown in the form of a table, but may be displayed by a graphical means such as a graph.

4. In the above-mentioned embodiment, a certain number of keywords are displayed from the top, but it is also possible to set an appropriate range of the narrowing down ratio and show the keywords within that range.

5. In the above embodiment, the keyword was extracted by performing morpheme analysis, but in the case of a document to which a word index is added, even if the word index is used as a keyword, except for the keyword extraction process by morpheme analysis. Good. Further, keyword extraction based on character types may be performed instead of morphological analysis.

6. Although the input means is the keyboard in the above embodiment, it may be other means such as a mouse.

Embodiment 2 FIG. 8 is a block diagram showing the arrangement of an apparatus according to the second embodiment of the present invention.

In FIG. 8, reference numeral 801 is a search condition input section for inputting search conditions, 802 is a search condition analysis section for analyzing the search conditions input from the search condition input section 801, and 803 is a search for searching documents satisfying the search conditions. A processing unit, 804 is a search result output unit that outputs the results obtained by the search processing unit 803, 805 is a constraint condition holding unit that holds constraint conditions for the time when the document was written, the document type, and the like, and 806 is the document. A search order control unit that controls the search order, 807 is a random number generation unit that generates a random number used to determine the document search order, and 808 is document management information of the document to be searched (when the document was written, A selection information holding unit that holds information such as type), a document database that holds documents to be searched 809, and document management information 810 for each document. It is a document management information file to be lifting.

The contents of the document management information file 810 are as shown in FIG. 9, for example. This is about newspaper articles, which are categorized by surface and time of article. Each article is given an ID and an address on the document file is given.

FIG. 10 is a flow chart showing the processing procedure of the operation in the apparatus shown in FIG. The operation of this embodiment will be described with reference to this figure.

First, in step S501, the search condition input unit 801 inputs search conditions. Next, step S5
02, the input search condition is searched for by the search condition analysis unit 80.
Analyze in 2.

In step S503, among the search conditions,
Check whether there is a constraint condition that represents constraints such as when the document was written or the type of document. If there is a constraint condition, the constraint condition is stored in the constraint condition holding unit 80 in step S504.
Hold at 5. If there is no constraint condition, all documents are targeted, so in step S505, the constraint condition holding unit 805 holds the constraint condition "all documents are targeted".

Next, in step S 506, the management information of the document satisfying the constraint condition held in the constraint condition holding unit 805 is held in the selection information holding unit 808. Then, in step S507, one document is selected from the documents whose document management information is held in the selection information holding unit 808 by the random number generated by the random number generation unit 807. The document management information of the selected document is deleted from the selection information holding unit 808 in step S508. Subsequently, in step S509, the selected document is read from the document database 809, and a search process for checking whether the search condition is satisfied is performed.

In step S510, the selection information holding unit 8
Check whether the document management information still remains in 08,
If any remain, the process returns to step S507 and repeats. If not remaining, the process ends.

Next, a more specific description of the present embodiment will be given by taking the case of searching newspaper articles as an example. It is assumed that the search condition is “What is the article about 1990 selling computers?”. By the search condition analysis, the constraint condition of “Article in 1990” is obtained. 1 referring to the document management information file 810
The document management information of the document corresponding to the article of 990 is held in the selection information holding unit 808.

Under the search conditions of this example, it is considered that the matching documents are economically concentrated. With traditional methods,
Because the economics were searched collectively, there was a possibility that the time interval for obtaining the matching documents could vary.

In this embodiment, the search order controller 806
The random number generated by the random number generator 807 controls the document search order. This order is, for example, "Political January No. 0
It becomes a random one such as “00623” → “Social September No.000713” → “Economic February No.000223”. As a result, it can be expected that the time intervals for obtaining the matching documents are even.

Next, modified examples (different embodiments) of the above embodiment will be listed.

1. In the above embodiment, each document is used as a unit for determining the search order, but a unit of several documents may be used as a unit.

2. Although a random number is used as a method for determining the search order in the above embodiment, a method other than a random number may be used as long as an order with no bias can be obtained.

3. In the above embodiment, as the document management information,
Although the time when the document was written and the type of document were used, other factors such as the author who wrote the document may also be used together. Moreover, you may use some of these.

4. In the above embodiment, the selection and the search process are repeated for each document, but the search process may be performed after first determining all the search orders.

Embodiment 3 FIG. 11 is a block diagram showing the arrangement of an apparatus according to the third embodiment of the present invention.

In FIG. 11, 1 is a search target document holding unit that holds a document to be searched, 2 is a search condition holding unit that holds search conditions, and 3 is a search target that is held in the search target document holding unit 1. A search processing unit that searches documents that satisfy the search conditions held in the search condition holding unit 2,
Reference numeral 4 indicates a search result holding unit that holds the results searched by the search processing unit 3, and 5 indicates that the search processing is performed by predicting that the number of results searched by the search processing unit 3 is too small or too large during processing. A search continuation possibility determination unit that determines whether continuation is possible, a search condition change instruction generation unit 6 that generates a search condition change instruction when the search continuation possibility determination unit 5 determines not to continue the search, and a search condition 7 It is a search condition change instruction holding unit that holds the search condition change instruction created by the change instruction creating unit 6.

FIG. 12 shows the system configuration of this embodiment. Here, 21 is a control memory, and FIG. 13 stores a control program according to the control procedure as shown in the flowchart. Reference numeral 22 is a central processing unit for making judgments / calculations and the like according to the control procedure stored in the control memory 21. Reference numeral 23 denotes a memory, which has a search condition holding unit CB, a search result holding unit RB, and a search condition change instruction holding unit IB. Reference numeral 24 denotes a disk device, which is a search target document holding unit T
Have B. 25 is a bus.

Next, the operation of this apparatus will be described with reference to the flow chart shown in FIG.

First, step S1 is repeated until the search condition holding unit 2 holds the search condition. When the search condition is held in the search condition holding unit 2, the process proceeds to step S2.

In step S2, the retrieval target document holding unit 1
It is checked whether or not the unprocessed search target document remains, and if the unprocessed search target document remains, the process proceeds to step S3. If no unprocessed documents to be searched remain, all the processes are terminated.

In step S3, the retrieval target document holding unit 1
One document is retrieved from the unprocessed documents to be retrieved and stored in the retrieval processing unit 3, and the process proceeds to step S4.

In step S4, the document retrieved in step S3 is searched by checking whether or not the search condition held in the search condition holding unit 2 is satisfied. Information that can identify a document is held in the search result holding unit 4 as a search result. Then, the process proceeds to step S5.

In step S5, it is checked whether or not the number of retrieval processings, which is the number of documents processed by the retrieval processing unit 3, is equal to a prescribed value for determining whether or not to continue the retrieval. If they are equal, the process proceeds to step S6. If it is not equal to the specified value, the process returns to the beginning of step S2.

In step S6, it is checked whether or not the number of search results, which is the number of documents searched by the search processing unit 3, is within a specified range of whether or not to continue the search. Returns to the beginning of step S2. If it is not within the specified range, the process proceeds to step S7.

For the processing in steps S5 and S6, for example, a counter for the number of search processings and a counter for the number of search results as shown in FIG. 14 and a prescribed value for the number of search processings and the number of search results as shown in FIG. A table that holds the upper and lower limits of the specified range is combined is used.

In step S5, the value of the counter of the number of search processing cases is compared with the specified value of the number of search processing cases in the table. For example, when the table shown in FIG. 15 is used and the counter value is as shown in FIG. 14, the process returns to the beginning of step S2.

When the counter value is as shown in FIG. 16 or 17, the process proceeds to step S6.

In step S6, the value of the counter of the number of search results is compared with the upper and lower limits of the specified range of the number of search results corresponding to the specified value of the number of search processing that matches the counter of the number of search processing. In the above example, when the counter value is as shown in FIG. 16, the process returns to the beginning of step S2. When the counter value is as shown in FIG. 17, the process proceeds to step S7.

In step S7, an instruction to change the search condition, which means that the search condition is made strict when the number of search results exceeds the specified range, is made loose when the number of search results is less than the specified range. An instruction to change the meaning search condition is created and held in the search condition change instruction holding unit 7. Then, all the processes are finished.

Next, modifications (different embodiments) of the above embodiment will be listed.

1. In the above-described embodiment, the search condition change instruction creating unit 6 and the search condition change instruction holding unit 7 are provided as an embodiment of the document processing apparatus (corresponding to claim 4), but the search condition change instruction creating unit 6 and the search are performed. It is also possible to omit the condition change instruction holding unit 7, omit the search condition change instruction holding unit IB from the memory 23 of FIG. 12, and omit step S7 from the flowchart of FIG. 13 (becomes the document processing device of claim 3). .

2. In the above embodiment, when it is determined whether or not the search can be continued by predicting whether the search results will be excessive or insufficient by comparing the specified range of the number of search results in the specified number of search processing with the actual number of search results. However, the present invention is not limited to this, and always compares the specified range of the search result count with the actual search result count, and compares the ratio of the actual search result count and the search processing count with the specified range. Whether or not the search can be continued may be determined by using another method of predicting whether or not the search results are excessive or insufficient.

3. In the above embodiment, the case in which the search continuation possibility determination unit 5 determines whether or not the search can be continued is described only when the number of search processing cases is equal to the specified value. However, the present invention is not limited to this, and determination is made at regular time intervals. The determination may be made at any time during the processing, such as making a determination each time one block of the search target document is processed or making a determination each time the search processing for one document is completed.

4. In the above-described embodiment, a case has been described in which the search continuity determination unit 5 uses both the decrease and the excessive number of search results as the conditions for determining the search continuity. However, the present invention is not limited to this. Alternatively, the determination may be performed on the condition that either one of the two is too small or too large.

5. In the above-described embodiment, the case where the search processing unit 3 performs processing in units of documents has been described, but the present invention is not limited to this, and other units such as blocks in the disk device may be used.

6. In the above embodiment, the case where the search target document holding unit 1 is realized by the disk device and the search condition holding unit 2, the search result holding unit 4, and the search condition change instruction holding unit 7 are realized by the memory has been described. However, the present invention is not limited to this, and may be another storage device such as a magneto-optical disk device.

[0082]

【The invention's effect】

(1) As described above, according to the present invention, the time required to obtain a target document can be shortened, so that there is an effect that a document processing apparatus with a document search function that is easy to use can be obtained.

(2) Further, according to the present invention, it can be expected that there will be no variation in the time intervals at which the matching documents are obtained, so that there is an effect that a document processing apparatus with a document search function that is easy to use can be obtained.

(3) Further, according to the present invention, the search processing is interrupted when the search condition is too strict or too loose by predicting that the number of search results will be too small or too large during the processing. Therefore, it is possible to obtain an effect that unnecessary search processing is not performed and the time required for the entire search can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an apparatus according to a first exemplary embodiment of the present invention.

FIG. 2 is a block diagram showing a system configuration according to the first embodiment.

FIG. 3 is a flowchart showing a processing procedure according to the first embodiment.

FIG. 4 is a flowchart showing in more detail a processing procedure of a part of the processing of the flowchart shown in FIG.

FIG. 5 is a diagram showing an example of contents of a document keyword holding unit according to the first embodiment.

FIG. 6 is a diagram showing an example of contents of a keyword holding unit according to the first embodiment.

FIG. 7 is a diagram showing an example of a display of a keyword display unit according to the first embodiment.

FIG. 8 is a block diagram showing a configuration of an apparatus according to a second exemplary embodiment of the present invention.

FIG. 9 is a diagram showing an example of document management information according to the second embodiment.

FIG. 10 is a flowchart showing a processing procedure according to the second embodiment.

FIG. 11 is a basic configuration diagram showing a document processing apparatus according to a third embodiment of the present invention.

FIG. 12 is a block diagram showing a system configuration of a third embodiment.

FIG. 13 is a flowchart showing a processing procedure of a third embodiment.

FIG. 14 is a diagram illustrating a counter for the number of search processing cases and the number of search result cases in the third embodiment.

FIG. 15 is a diagram illustrating a table holding a specified value of the number of search processing cases and a specified range of the number of search result cases in the third embodiment.

FIG. 16 is a diagram illustrating a state different from that of FIG. 14 of the counter of the number of search processing cases and the number of search result cases in the third embodiment.

FIG. 17 is a diagram illustrating a state different from that of FIG. 16 of the counter of the number of search processing cases and the number of search result cases in the third embodiment.

[Explanation of symbols]

 1 search target document holding unit 2 search condition holding unit 3 search processing unit 4 search result holding unit 5 search continuation possibility determination unit 6 search condition change instruction creation unit 7 search condition change instruction holding unit 101 search condition input unit 102 search processing unit 103 Document database 104 Search result storage unit 105 Search result output unit 106 Keyword evaluation unit 107 Morphological analysis processing unit 108 Keyword storage unit 109 Document keyword storage unit 110 Keyword display unit 111 Keyword selection unit 201 Control memory 202 Central processing unit 203 Memory 204 Keyboard 205 Disk device 206 Display 207 Bus 801 Search condition input unit 802 Search condition analysis unit 803 Search processing unit 804 Search result output unit 805 Constraint condition holding unit 806 Search order control unit 807 Random number generation unit 808 Selection information storage Part 809 document database 810 document management information file

Claims (4)

[Claims] 1. A document database having a plurality of documents, a search condition input means for inputting a condition for searching a document, a search means for searching a document satisfying the search condition from the document database by a full text search, and the search. First extracting means for extracting a keyword group from a document obtained as a result of searching by the means; aligning means for evaluating the keyword group and aligning the keywords according to the evaluation value; and a specific keyword from the keyword group. And a second extracting unit for extracting a document including the keyword selected by the selecting unit from the search result, and presenting a keyword capable of narrowing the search result of the full-text search, A document processing apparatus, wherein search results are narrowed down by the presentation. 2. A document storage unit for storing a document, a search condition input unit for inputting a condition for searching a document, a search unit for searching a document satisfying the search condition from the document holding unit, and the search unit. A document processing apparatus, comprising: a control unit that controls the order in which documents are searched, so that there is no bias in the order in which documents are searched. 3. A search target document holding unit for holding a document to be searched, a search condition holding unit for holding a search condition, and a search condition holding from the search target document held in the search target document holding unit. What is claimed is: 1. A document processing device comprising: a search processing unit that searches for a document that satisfies the search condition held in the unit; and a search result holding unit that holds the result searched by the search processing unit. The document processing apparatus is provided with a search continuation possibility determination unit that determines whether or not the search processing can be continued by predicting that the number of cases is too small or excessive during processing. 4. The document processing apparatus according to claim 3, further comprising: a search condition change instruction creating unit that creates a search condition change instruction when the search continuability determination unit determines that the search is not continued. A document processing apparatus, comprising: a search condition change instruction holding unit that holds the search condition change instruction created by the search condition change instruction creating unit.