JPH04281564A - Document retrieving device - Google Patents

Abstract

PURPOSE:To extract a document including a key word designated by a user at high speed. CONSTITUTION:One technical term summary 12 corresponding to respective special field Is selected from plural technical term summaries 12 in a key word dictionary 6 consisting of a common term summary 11 and the plural technical term summaries 12 by a technical term selection part 7 based on the key word and a document inputted from an input means 3, and a retrieval bit map and a document bit map in accordance with inputted key word and document are generated at a bit map generating part 9 based on selected one technical term summary 12 and common term summary 11. The matching of the retrieval bit map and the document bit map is performed at a bit map matching part 19, and the document Including an Input key word can be extracted.

Description

[Detailed description of the invention]

[Purpose of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document retrieval device that performs a full text search to extract documents containing input keywords.

0002

2. Description of the Related Art In recent years, documents have been digitized and a large amount of document data has been distributed. When extracting what the user needs from this large amount of document data, the mainstream search method is to input keywords consisting of character strings. Searches using this keyword are mainly conducted in the following two ways.
Broadly divided into methods.

(1) A method in which keywords are assigned to all documents in advance, and documents to which keywords input by the user are assigned are extracted.

(2) A full-text search method to determine whether the document contains the keyword input by the user.

[0005] In the above method (1), by limiting the number of keywords assigned to a document, character string matching processing with keywords input by the user is minimized, thereby realizing high-speed search. There is. however,
A keyword must be assigned to a document in advance, and if the user specifies a keyword that has not been assigned, it is impossible to extract the document. Moreover, the work of assigning keywords to a document is a burden on the document creator, and the method of determining keywords is sometimes left up to the document creator, making it difficult to ensure uniformity of keywords.

Furthermore, since the method (2) above extracts documents that include the character string input by the user, there are few omissions in extraction. However, since all character strings in the search document are matched, this results in a slow search device, resulting in an inefficient search for the user.

[0007] By the way, in the methods (1) and (2) above, when a document that satisfies the conditions is found, the documents containing the character string input by the user are listed and displayed, and the search results are displayed to the user. give. Furthermore, the user must scroll the display screen to determine whether each document is suitable for the purpose and select the desired document from among the large amount of output documents. At this time, the user is not shown how the keyword appears in the document.

[0008]

[Problems to be Solved by the Invention] As mentioned above, in the conventional full-text search search device, there is a burden on the document creator to assign keywords to the document. If an attempt is made to reduce the burden of assignment, the speed of the search device becomes slow, which poses a problem in that it places a burden on the user.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a document retrieval device that can quickly extract documents containing keywords specified by a user without the need to assign keywords to documents in advance. shall be.

[Configuration of the invention]

[0011]

[Means for Solving the Problems] In order to achieve the above object, the present invention provides a document retrieval device for extracting documents containing input keywords, which includes input means for inputting keywords, a list of common terms, and a plurality of input means for inputting keywords. a keyword dictionary consisting of a list of technical terms; a technical term selection means for selecting one list of technical terms corresponding to each specialized field from the plurality of technical term lists based on the keywords and documents; a bitmap creation means for creating a search bitmap corresponding to the keyword and a document bitmap corresponding to the document based on the list of technical terms and the list of common terms; and matching of the search bitmap and the document bitmap. The present invention is characterized by comprising a bitmap matching means for extracting documents containing the above-mentioned keywords, and an output means for outputting the documents extracted by the bitmap matching means.

0012

[Operation] Since the present invention is configured as described above, each specialized term is selected based on the keywords and documents inputted from the input means from a plurality of technical term lists of a keyword dictionary consisting of a common term list and a plurality of technical term lists. A list of technical terms corresponding to the field is selected, and a search bitmap and a document bitmap corresponding to the input keyword and document are created based on the selected list of technical terms and the list of common terms. By matching this search bitmap with the document bitmap and extracting documents that include the input keyword, the target document can be extracted at high speed.

[0013]

Embodiments Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of a document search device according to an embodiment of the present invention.

In the figure, reference numeral 1 denotes an external storage device such as a floppy disk device or a hard disk device, which stores document data that has already been created.The document data read from this external storage device 1 is For example, the document data memory 2 is stored in a dynamic RAM. One piece of document data consists of a text data section that includes only text information in the document, and a non-text data section that includes image data, format information, and the like.

Reference numeral 3 denotes an input unit for inputting search keywords, commands, etc., which is composed of, for example, a keyboard, a mouse, and a device for controlling these. Commands and the like are controlled by the control unit 4, such as VRA.
M and the bit information stored in this VRAM are displayed on a display section 5 consisting of a CRT display for displaying the bit information as a dot string. The display unit 5 further displays search results, document data stored in the document data memory 2, and the like.

The control unit 4 is composed of a RAM that stores system programs and is used as a buffer memory, a CPU that executes control operations, etc., and is connected to each of the above-mentioned devices or devices that will be described later by a bus, and controls each device. , and performs control and processing such as data transfer between devices. Note that this control unit 4 includes buffers and counters necessary for control and processing, such as
The number of document data stored in the external storage device 1 is stored in a document number storage buffer (not shown).

The document retrieval device of the present invention creates in advance a document bitmap indicating whether or not common keywords and specialized keywords are included for each document, and also uses the input unit 3
A search bitmap in the same format as the document bitmap is created for the search keyword input from , and documents containing the input search keyword are extracted by matching both bitmaps. Creation of these bitmaps is performed by a keyword dictionary 6 , a technical term selection section 7 , a dictionary buffer 8 , a bitmap creation section 9 and a bitmap buffer 10 .

[0019] Keyword dictionary 6 is a list of common terms 11
The common term list 11 is a collection of terms included in a document regardless of the technical field, and an example of the contents of the common term list 11 is shown in FIG. Further, the technical term list 12 is a collection of terms used in documents according to technical fields, and an example of the contents of the technical term list 12 is shown in FIG. As shown in FIG. 3, the technical term list 12 has a technical field number 13 assigned to each technical field, and includes technical term lists 12a, 12b,
It consists of 12c,..., 12n.

The technical term selection unit 7 compares the terms included in the document read into the document data memory 2 with the technical term list 12 of the keyword dictionary 6, and selects one field-specific specialized term from the technical term list 12. Select the term list 12n. Similarly, one field-specific technical term list 12n is selected from the search keywords input from the input section 3. The selected field-specific technical term list 12n is temporarily stored in the dictionary buffer 8.

The bitmap creation section 9 includes a common term list 11 that constitutes the keyword dictionary 6 and a technical term selection section 7.
The document data and the search keyword input from the input unit 3 are converted into bitmaps based on the field-specific technical term list 12n selected by . As shown in FIG. 4, the document data bitmap 14 is composed of a document name 15, a specialized field number 13, a common keyword area 16, and a specialized keyword area 17, a specific example of which is shown in FIG. The number of flags in the common keyword area 16 and the specialized keyword area 17 match the number of common terms in the common term list 11 and the number of technical terms in the field-specific technical term list 12n selected from the technical term list 12, respectively.
, one term corresponds to one flag in the order of each term list 12n. The document bitmap 14 is created using the common keyword area 16 or specialized keywords corresponding to the words and phrases included in the document that match common terms in the common term list 11 or technical terms in the field-specific technical term list 12n. This is done by setting the keyword flag in area 17 to "1". The document bitmap 14 created in this way is stored in the bitmap buffer 10. on the other hand,
As shown in FIG. 6, a bitmap 18 of search keywords
is composed of a specialized field number 13, a common keyword area 16, and a specialized keyword area 17, a specific example of which is shown in FIG. 7(A). Common keyword area 16 or specialized keyword area 17 of search bitmap 18
The keyword flag is set in the same way as the keyword flag of the document bitmap 14,
One term corresponds to one flag in the following order: common term list 11 and field-specific technical term list 12n. The search bitmap 18 created in this way is stored in the bitmap buffer 10 in the same way as the document bitmap 14.

The bitmap matching section 19 extracts documents containing the search keyword input by the user through the input section 3 . The bitmap matching unit 19
Documents containing the search keyword are extracted by comparing each document bitmap 14 created by the bitmap creation section 9 and stored in the bitmap buffer 10 with the search bitmap 18. For example, when comparing the search bitmap 18 shown in FIG. 7(A) and the document bitmap 14 shown in FIG. 7(B), first, the specialty field number 13 is compared. If the specialty field numbers 13 match, then common keyword area 16 and specialty keyword area 1
Compare the keyword flags in 7. This comparison is
This is done by checking whether a flag in the document bitmap 14 that corresponds to a term whose flag is set to "1" in the search bitmap 18 is set to "1". Therefore, if the flags in the document bitmap 14 corresponding to all the keyword flags set to "1" in the search bitmap 18 are also set to "1", all the search keywords input by the user are included. Indicates that it is a document. In the example document shown in Figure 7(B),
This indicates that the document contains two search keywords input by the user.

Next, specific processing operations of the document retrieval apparatus having the above configuration will be explained with reference to flowcharts showing the processing flow of FIGS. 8 to 10. FIG. 8 is a flowchart showing the creation process of the document bitmap 14, FIG. 9 is a flowchart showing the creation process of the search bitmap 18, and FIG. 10 is a flowchart showing the matching process of both bitmaps 14 and 18.

In FIG. 8 showing the process of creating the document bitmap 14, first, the buffers and counters in the control unit 4, the dictionary buffer 8 and the bitmap buffer 10 are
is initialized, and then one document is transferred from the document data stored in the external storage device 1 to the document data memory 2.
is loaded into. (Step S1).

Proceeding to step S2, the technical term selection section 7
is started and the terminology selection unit 7 reads the terms and keywords dictionary 6 used in one document.
Then, one of the field-specific technical term lists 12n constituting the technical term list 12 is selected. The selection of this field-specific technical term list 12n is determined by the frequency with which the terms for each field in the field-specific technical term list 12n appear in terms used in one document read. For example, a list of specialized terms 12a, 12b, 12 by field in terms used in one read document.
Assuming that the frequencies of occurrence of c are "2", "5", and "1", respectively, the field-specific technical term list 12b is selected.

Subsequently, the common term list 11 and the selected field-specific technical term list 12n are read from the keyword dictionary 6 and stored in the dictionary buffer 8. A document keyword dictionary for one read document is created using the common term list 11 and the field-specific technical term list 12n stored in the dictionary buffer 8. Along with the creation of this document keyword dictionary, the bitmap creation section 9 is activated, and the common term list 11 and field-specific technical term list 12n in the document keyword dictionary are stored in the bitmap buffer 10.
A bit area is created consisting of a common keyword area 16 and a specialized keyword area 17, each having the same number of bits as the number of terms. (Step S3, Step S4
, step S5).

When the creation of the document keyword dictionary and bit area is completed, the bitmap creation section 9 selects one term (keyword) from the document keyword dictionary and transfers it to the technical term selection section 8.
It is determined whether the read term is included in one document stored in the document data memory 2. If it is included, the bit corresponding to the term in the bit area is set to "1", that is, a flag is set. For example, if the imported term is one of the document keyword dictionary
If the term is located at the 0th position, "1" is set in the 10th bit of the bit area. Furthermore, if it is not included, this flag is not set. To read terms from the document keyword dictionary above, use the common term list 11.
The processing is performed sequentially in the order from field-specific technical term list 12n and in the order in which they are stored. The above reading, discrimination,
When the flag setting process is completed, the process moves to step S9. (Step S6, Step S7, Step S8).

In step S9, it is checked whether the processes of steps S6, S7, and S8 have been completed for all terms in the document keyword dictionary, and if there are terms that have not yet been completed and should be read. for,
Returning to step S6, the above-described reading, determination, and flag setting processes are repeated. Further, if all the terms have been completed and there is no term to be read, the process is performed in step S10.
Proceed to.

In step S10, the document name 15 and the specialized field number 13 corresponding to the selected field-specific technical term list 12n are added to the bit area, and the document name 15 and specialized field number corresponding to the one read document are added. 13, common keyword area 16, and specialized keyword area 17
A document bitmap 14 consisting of is formed on the bit area. In this way, the document bitmap 14 corresponding to one read document is created, but when the creation of the document bitmap 14 is completed, the process proceeds to step S11.

In step S11, the external storage device 1
It is checked whether the formation of the document bitmap 14 has been completed for all the document data stored in the . This check is performed while referring to a document number storage buffer (not shown) in the control section 4. If the process has not yet been completed for all document data, the process returns to step S1, and the same process as above is performed, and the document bitmap 14 corresponding to each document data is stored in the bitmap buffer 10.
is stored in Furthermore, if the above processing for all document data has been completed, the creation processing of the document bitmap 14 is ended.

With the above steps, the process of creating the document bitmap 14 is completed.

Next, referring to FIG. 9, search bitmap 1
The creation process of step 8 will be explained.

First, the user inputs a plurality of keywords consisting of character strings for searching through the input unit 3. This input keyword is stored in a buffer (not shown) in the control unit 4. (Step S21, Step S22).

[0034] When the keyword input is completed, the technical term selection unit 7 is activated, the keyword input by the technical term selection unit 7 is compared with the terms used in the keyword dictionary 6, and the technical term list 12 is compiled. One of the field-specific technical term lists 12n is selected in the same manner as in step S2 above. At this time, the plural input keywords are treated as one document. (Step S23).

[0035] Next, the above steps S3 and S4
And similar to the process in step S5, the keyword dictionary 6
The common term list 11 and the selected field-specific technical term list 12n are read, and a search keyword dictionary consisting of the common term list 11 and the selected field-specific technical term list 12n is created in the dictionary buffer 8. In the buffer 10, there is a common keyword area 1 having the same number of bits as the number of terms in the common term list 11 and field-specific technical term list 12n in the search keyword dictionary.
6 and the specialized keyword area 17 are created in an area different from the area where the document bitmap 14 is stored by the process of step S10. (Step S24, Step S25, Step S26).

[0036] When the creation of the search keyword dictionary and bit area is completed, the bitmap creation section 9 controls the control section 4.
One input keyword is read from a plurality of input keywords stored in a buffer within the computer. (Step S27)
. Next, one term (search keyword) is read from the search keyword dictionary via the technical term selection unit 8, and it is determined whether the input keyword and the character string of the term match. If they match, "1" is set in the bit in the bit area corresponding to the term, as in step S8 above. For example, if the loaded term is one of the search keyword dictionaries,
If the term is located at the 0th position, "1" is set in the 10th bit of the bit area. When this flag setting process is completed, the process moves to step S32. Also,
If they do not match, the process proceeds to step S31,
It is checked whether reading of all terms in the search keyword dictionary has been completed. If reading of all terms has not been completed, the process returns to step S88 to read new terms and repeat the same process as above. Furthermore, if reading of all terms has been completed, the process moves to step S32. Note that the terms in the search keyword dictionary are sequentially read from the common term list 11 to the field-specific technical term list 12n, and in the order in which they are stored. (Step S28, Step S29, Step S30, Step S31).

[0037] In step S32, it is checked whether or not reading has been completed for all input keywords. If reading has not been completed for all input keywords, the process returns to step S27 and a new input keyword is read. The above-mentioned terms can be read, identified,
Repeat the flag setting process. Furthermore, if reading has been completed for all input keywords, the process moves to step S33.

In step S33, the specialized field number 13 corresponding to the selected field-specific technical term list 12n is added to the bit area, and a search bitmap consisting of the specialized field number 13, the common keyword area 16, and the specialized keyword area 17 is created. 18 is formed and stored in the bitmap buffer 10.

With the above steps, the process of creating the search bitmap 18 is completed.

Next, the matching process between the document bit match 14 and the search bit map 18 will be explained with reference to FIG.

First, the bitmap matching section 19 is activated and all document bitmaps 14 stored in the bitmap buffer 10 are read. Subsequently, the search bitmap 18 is also read, and the process proceeds to step S43.
to move to. (Step S41, Step S42
).

In step S43, one document bitmap 14 among the plurality of document bitmaps 14 is compared with the specialty field number 13, common keyword area 16, and specialty keyword area 17 of the search bitmap 18. As a result of the comparison, if a flag is also set in the bit position in the 1-document bitmap 14 that corresponds to the flagged bit in the search bitmap 18, that is, 1.
If a flag is set at a position in the document bitmap 14 that matches the bitmap in the search bitmap 18,
Since all input keywords are included in one document, it is determined that it is the target document, and one document bitmap 1
4 is stored in a buffer (not shown) in the control unit 4. (Step S45). If the bitmaps do not match, the process moves to step S46.

In step S46, it is checked whether or not the matching process has been performed on all document bitmaps 14. If the matching process has not been completed on all document bitmaps 14, the process returns to step S43. The process returns to , and the same processing as above is performed. Furthermore, if the processing has been completed for all document bitmaps 14, the process moves to step S47.

In step S47, the document with document name 15 stored in the buffer in the control section 4 is read out from the external storage device 1 and output and displayed on the display section 5 as an extracted document.

With the above steps, the matching process between the document bit match 14 and the search bit map 18 is completed.

Note that in the above embodiment, documents containing all input keywords are extracted as target documents, but the extraction of target documents is not limited to this, and documents containing even one input keyword are extracted as target documents. In this case, the document display priority may be determined according to the number of input keywords.

Further, in the above embodiment, the document bitmap 14 and the search bitmap 18 are created in the bitmap creation section 9, but the invention is not limited to this, and the functions of the bitmap creation section 9 can be distributed, The document bitmap 14 and the search bitmap 18 may be created by a document bitmap creation section and a search bitmap creation section, respectively.
By performing distributed creation in this manner, parallel processing of bitmap creation becomes possible and processing speed can be improved.

Further, in the above embodiment, the dictionary buffer 8 and the bitmap buffer 10 are configured independently from the control section 4, but the invention is not limited to this, and they may be configured as buffers within the control section 4.

Further, the present invention is not limited to the above-mentioned embodiments, and it goes without saying that various modifications can be made without departing from the gist of the present invention.

[0050]

As described in detail above, according to the document retrieval device of the present invention, by selecting specialized fields for documents and keywords, it is possible to suppress the bitmap capacity per document. Since the target document is extracted by comparing and determining, it has great practical effects such as high-speed document retrieval.

[0051] Furthermore, the document creator does not need to add search keywords, and the labor of the document creator can be reduced.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of a document search device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a specific example of the format of a common term list.

FIG. 3 is a diagram showing a specific example of the format of a list of technical terms.

FIG. 4 is a diagram showing an example of document bitmap storage.

FIG. 5 is a diagram showing a specific example of a document bitmap.

FIG. 6 is a diagram showing an example of storing a search bitmap.

FIG. 7 is a diagram showing a specific example of a search bitmap.

FIG. 8 is a flowchart showing the flow of processing for creating a document bitmap.

FIG. 9 is a flowchart showing the flow of processing for creating a document bitmap.

FIG. 10 is a flowchart showing the flow of bitmap matching processing.

[Explanation of symbols]

3...Input section (input means) 5...Display section (output means) 6...Keyword dictionary 7...Technical term selection section (technical term selection means) 9...Bitmap creation section (bitmap creation means) 11...Common term list 12 …List of technical terms 14…Document bitmap 18…Document bitmap 19…Bitmap matching unit (bitmap matching means)

Claims (1)

[Claims] Claim 1: A document retrieval device for extracting documents containing input keywords, comprising: an input means for inputting keywords; a keyword dictionary consisting of a list of common terms and a plurality of lists of technical terms; a technical term selection means for selecting one technical term list corresponding to each specialized field from the plurality of technical term lists;
a bitmap creation means for creating a search bitmap corresponding to the keyword and a document bitmap corresponding to the document based on the one technical term list selected by the technical term selection means and the common term list; A document retrieval device characterized by comprising: a bitmap matching means for matching the above document bitmap and extracting a document containing the above keyword; and an output means for outputting the document extracted by the bitmap matching means. .