JPH06290219A - Document processor with character retrieving function - Google Patents

Abstract

PURPOSE:To improve the retrieval efficiency by enabling partial matching retrieval in paragraph units of a KANJI(Chinese character)-KANA(Japanese syllabary) mixed sentence. CONSTITUTION:When a converting means 102 converts reading information inputted by an input means 101 into the KANJI-KANA mixed sentence, a paragraph information adding means 103 adds delimiting information on paragraphs to the KANJI-KANA mixed sentence. For the addition of the delimiter information, special codes which are independent of character codes are added, the information is embedded as the special flag in the character codes, or specific character strings are stored, paragraph by paragraph. Then, when a retrieval means 105 performs retrieval, a sentence which matches with the KANJI-KANA mixed sentence in one paragraph or more among the paragraphs of the KANJI- KANA mixed sentence is retrieved in the document stored in a document storage means 104. Therefore, since the partial matching retrieval is performed in paragraph units, meaningless retrieval conditions are not generated and the generation of fruitless retrieving operation is suppressed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document processing apparatus such as a word processor or a personal computer capable of creating a document in Japanese, and more particularly to a character searchable character string in a document. The present invention relates to a document processing device with a search function.

[0002]

2. Description of the Related Art Conventionally, in a document processing apparatus of this type, as shown in FIG. 7, when searching for a specific character string in a document, a method of making a perfect match with a specified character string the condition is used. There is a method that uses a partial match that allows only part of the specified character string to match.

As shown in FIG. 7 (a), when the specified character string is M-shaped, the method that uses perfect matching as a condition is that all positions from the 1st to the M-th character are determined as matching conditions. Only when the characters match in, it is judged that they match.
That is, the matching condition is the designated character string itself.

On the other hand, the method that uses partial matching as a condition is shown in FIG.
As shown in (b), if the designated character string is M-letter, as a condition for determining that the characters match, when the characters match at any one of the first to M-th positions, it is determined that they match. That is, it is a method of easing the condition by ignoring the mismatch of the characters at any one or more positions of the 1st to Mth characters.

[0005]

However, such a conventional method that uses a perfect match as a condition only searches for a character string that exactly matches a specified character string, and the operator's memory is ambiguous. If the wrong character string is entered, a completely matching character string may not be found, or an unexpected character string may be searched for. This is especially likely to occur in long texts or documents created over a long period of time.

On the other hand, the conventional method that uses partial matching as a condition not only uses the specified character string as a condition as it is, but also automatically generates a relaxed condition. Was a mistake in the memory of the operator,
Alternatively, input mistakes can be absorbed. However, since the designated character string is recognized as a set of one character to the last, the characters are all equivalent, that is, one character is independent and the condition is automatically generated. Therefore, a meaningless condition having a very low existence probability is generated, and meaningless search work occurs. Also, if the conditions are relaxed assuming that each character is independent,
Many conditions are determined to match, and many unexpected character strings are extracted.

Due to the above disadvantages, the conventional complete matching method requires the accuracy of the memory / input of the operator, and the partial matching method requires the time required for the work.

The present invention has been made in consideration of such circumstances, and when a character string consisting of a plurality of clauses is searched, it is possible to perform a partial match search in clause units, thereby improving the search efficiency. It is an object of the present invention to provide a document processing device with a character search function which is improved.

[0009]

FIG. 1 is a block diagram showing the configuration of the present invention. As shown in the figure, the present invention is
It has an input means 101 for inputting word reading information over a plurality of phrases and a kana-kanji conversion dictionary. By referring to the dictionary, the input reading information is converted into a kanji kana mixed sentence consisting of a plurality of phrases. The converting means 102, the clause information adding means 103 for adding bunsetsu delimiter information to the kanji kana kana mingling sentence when the input reading information is converted to the kanji kana kana merging sentence, and a large number of bunsetsu delimiting information added By referring to the document storage unit 104 that stores the document including the Kanji-Kana mixed sentence and the paragraph delimiter information added to the Kanji-Kana mixed sentence, the conversion unit 102 is referred to.
The document storage means 104 stores a kanji-kana mixed sentence that matches the kanji-kana mixed sentence of any one or more of the plurality of kanji-kana mixed sentences converted by.
Search means 105 for searching from the documents stored in
Output means 10 for outputting the search result by the search means 105
Provided is a document processing device having a character search function, which is provided with 6.

An input device such as a keyboard device or a tablet device is used as the input means 101 of the present invention.

Conversion means 102, phrase information addition means 10
3 and the search means 105, CPU, ROM, R
It is convenient to use a microcomputer system including an AM and an interface, and the document storage means 104
As for, the RAM therein is usually used.

As the output means 106, a display device such as a CRT display device or an LCD (liquid crystal display device) and various printers are used.

[0013]

According to the present invention, when the reading information input by the input means 101 is converted into the kanji kana mixed sentence by the converting means 102, the phrase information adding means 10
By 3, the phrase delimiter information is added to the kanji kana mixed sentence. Then, when a search is performed by the search means 105, the kanji-kana mixed part that matches the kanji-kana mixed statement of any one or more of the plurality of kanji-kana mixed sentences of the kanji kana mixed sentence is stored in the document storage. Means 104
Are retrieved from the documents stored in.

Therefore, since the partial match search is performed in phrase units, meaningless search conditions are not generated, and thus unnecessary search work is suppressed, and the search work can be simplified and made more efficient.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the embodiments shown in the drawings. The present invention is not limited to this.

FIG. 2 is a block diagram showing the configuration of an embodiment in which the present invention is applied to a Japanese word processor. As shown in this figure, the Japanese word processor of the present invention is
CP that executes programs and controls and controls the entire system
U (Central Processing Unit) 1 and RAM for storing document data
2, a ROM 3 in which a control program executed by the CPU 1 and dictionary data for Kana-Kanji conversion are stored, a printer 5 as a printing unit, a printer controller 4 for controlling the printer 5, and a display including a CRT or a liquid crystal display. A device 7, a display controller 6 for controlling the display device 7 to display data on the display device 7, a keyboard 9 as an input unit, a key interface 8 serving as an interface of the keyboard 9, and data, programs and the like are stored. FD (floppy disk device) 11 and IC card 13 as means, and FD1
FD controller 10 for controlling the IC card 1 and the IC card 13
And an IC card interface 12 and an EEPROM 15 for storing system data.

The RAM 2, ROM 3, printer controller 4, display controller 6, key interface 8, FD controller 10, IC card interface 12, and EEPROM 15 for storing system data.
Are connected to the CPU 1 via the address bus / data bus 14.

From the keyboard 9, word reading information is input over a plurality of phrases. CPU1 is ROM3
By referring to the kana-kanji conversion dictionary stored in, the input reading information is converted into a kanji-kana mixed sentence composed of a plurality of clauses. Further, at the time of the conversion, bunsetsu delimiter information is added to the kanji kana mingled sentence, and a document including a large number of kanji kana mingled sentences to which the bunsetsu delimiting information is added is stored in the RAM 2.

When a search character string consisting of a plurality of clauses is specified from the keyboard 9, the delimiter information of the clause added to the kanji kana kana sentence is referred to, so that a plurality of clauses of the search character string can be selected. , A character string matching the search character string of one or more clauses is RA
The document stored in M2 is searched, and the search result is displayed on the display device 7.

The phrase delimiter information is added as follows. That is, the CPU 1 refers to the kana-kanji conversion dictionary at the time of kana-kanji conversion, and automatically recognizes the delimiter of the bunsetsu of the input character string and converts it into a kanji-kana mixed sentence (so-called continuous phrase conversion). At the time of confirming the conversion, the delimiter of the phrase is added to the designated character string as delimiter information and stored in the RAM 2.

The following methods are available for adding this information. -Add a special code that is independent of the character code (Fig. 3
(A)). Embed as a special flag in the character code (see (b) of FIG. 3). -The delimiter information is not provided as a code or a flag, but a specified character string is stored for each delimiter of a phrase (see (c) of FIG. 3).

Next, how this delimiter is used to relax search conditions will be described with reference to FIG. If the initially specified M character string is divided into bunsetsu A, bunsetsu B, and bunsetsu C (one bunsetsu is a string of one or more characters), partial matching is performed for each bunsetsu. Exact match means that all clauses A, B, and C match, and partial match means that one or more clauses of clause A, clause B, and clause C match. .

As a first relaxation, any one clause is removed from the matching condition. For example, there are three conditions: the phrase A and the phrase B match, the phrase A and the phrase C match, and the phrase B and the phrase C match. Thus, the matching condition is relaxed by increasing the number of clauses to be excluded from the matching condition one by one. This relaxation is performed until the matching condition becomes one clause (in the example in the figure, matching of clause A only, clause B only, and clause C only), and this limit is set.
In the figure, "*" is an arbitrary character.

The contents of the processing operation in such a configuration will be described with reference to the flowchart shown in FIG. First, in step S1, the search process is started by the operator,
In step S2, the operator inputs a character string to be searched for in the document from the keyboard 9 (see screen 1 in FIG. 6). next,
In step S3, the input character string is converted into a kanji kana mixed sentence (see screen 2 in FIG. 6). That is, the Kana-Kanji conversion program stored in the ROM 3 causes the CPU 1
Refers to the kana-kanji conversion dictionary stored in ROM3, automatically divides the input character string into singular and plural clauses, converts each clause into kanji, and converts into kanji kana mixed sentences. . Here, the input character string including a single character and a plurality of characters is divided into N clauses 1 to N. Next, the converted character string is determined in step S4, conversion delimiter information is added in step S5, and the specified character string is stored. Then, the work is started in the search mode by exact match.

In step S6, the comparison position Q is set. As the initial value, the beginning of the document (first character: Q = 1) is set. Next, in step S7, the comparison position Q
The document data and the designated character string are compared with each other, and when all of the designated character strings match, the process proceeds from step S8 to S11, and the surrounding document data having the matching character string is displayed (see screen 3 in FIG. 6). , And proceeds to step S12.

If they do not match in step S8, the process proceeds to step S9, and if the search position is not reached to the end of the sentence, that is, if there is document data to be searched, the process proceeds to step S10 and the comparison position. Advances one (Q = Q + 1). If it is the end of the sentence, step S
Proceed to 13.

At step S12, the continuation / stop of the search work is selected (see screen 4 in FIG. 6), and when the stop of the search work is selected, the process ends. On the other hand, when continuation of the search work is selected, steps S12 to S9 are performed.
Go to and continue the search.

The search up to this point is a search mode based on perfect matching. If there is no matching character string in the exact match mode, or if the end of the sentence has been reached continuously, step S9
To step S13, the search mode for partial matching is automatically set.

In the partial match mode, first, step S13.
Assuming that the character string checked by the perfect match in K is composed of K clauses, the criterion L for judging the match is K. Then, in step S14, 1 is subtracted from the reference L, and step S
Check L at 15. Since L is the criterion of the number of matching clauses that is judged to be a match, if L <1, it is determined that relaxation is impossible, and the work ends (see screen 7 in FIG. 6). That is, when the character string input as the initial condition is composed of one clause, the search operation by partial matching is not performed.

If L is 1 or more, that is, if the condition character string is composed of 2 or more clauses, it is determined that the clause can be relaxed, and the process proceeds from step S15 to step S16. Step S16
Then, the search position value P representing the number of characters from the beginning is set to 1. Then, in step S17, the initial value of M indicating which number of clauses is being searched is 1, the number of clause matches N = 0,
The auxiliary pointer R = 0.

Then, in step S18, the position (P +
In R), it is checked whether or not the character string of the Mth clause from the beginning matches the character string in the document data at that position. That is, each of the 1st to Hth characters of the Mth clause and the beginning of the document data is (P + R + 1-1) to (P + R +).
If all the characters of (H-1) match, the process proceeds from step S19 to step 20. If even one character does not match in step S19, the process proceeds to step S21.

In step S20, 1 is added to the phrase matching number N, and the number of characters of the phrase M currently being compared is added to the auxiliary pointer R in order to continuously check the matching of the next phrase. In step S21, the number of mismatched clauses (MN) is compared with the number of mismatch judgment reference clauses (KL), and the value of (MN) is (K-
If it is less than or equal to the value of (L), the process proceeds to step S22, M is compared with the total phrase number K, and if they match, the process proceeds to step S25. In step S21, the value of (MN) is (KL)
If it is larger than the value of, it is determined that they do not match, and the process proceeds to step S23. On the other hand, M is compared with the total phrase count N, and if they do not match, the process proceeds to step S24, 1 is added to M, and step S1
Proceed to step 8 to compare the next clause.

If M and K are equal at step S22,
Since all the phrases have been compared, the process proceeds to step S25, and it is determined that the partial match is established, and the document data in the vicinity thereof is displayed (see screen 5 in FIG. 6). After confirming the result, the continuation / stop of the process is selected in step S26 (see screen 6 in FIG. 6). If the process is stopped, the process ends (see screen 7 in FIG. 6).

On the other hand, in the case of continuing, the process proceeds to step S23, and the position for searching one is advanced. Then, the process proceeds to S27 to check whether or not there is document data to be searched. If P
= In the case of the end of the sentence, the process proceeds to step S28, and the matching condition is relaxed by subtracting 1 from L, which is the criterion for determining matching. Then, the process proceeds to step S15, the conditions are relaxed, and the partial matching search is performed again from the beginning of the document. On the other hand, if the end of the sentence has not been reached, the process proceeds to step S17 to advance the comparison position and continue the work under the current conditions.

The above processing is roughly divided as follows.・ Steps S1 to S5: Search character string input (kana-kanji conversion) work ・ Steps S6 to S12: Search work by exact match ・ Steps S13 to S28: Search work by partial match

An actual input example will be described based on the display example of the screen of FIG. 6. In step S1, the process is started, and in step S2, as shown in the screen 1 of FIG. "Ichichikensaku" is input, Kana-to-Kanji conversion is performed in step S3, and as shown in screen 2 of FIG. 6,
Convert to "exact match search". With this kana-kanji conversion,
It is assumed that the sentences are automatically converted into three clauses such as "perfect", "match", and "search".

When determined in step S4, the segment delimiter information is embedded in the search character string "exact match search" and stored. Then, the search by exact match is started. First, here, a pointer for checking from the beginning (first character) of the document is set, and in step S7, the first character and "end", the second character and "all", the third character and "one", ... ,
When the fifth character and "inspection" are compared with each other and the sixth character and "search" are compared, if all of them match, that portion is displayed as in screen 3 of FIG. 6 in steps S8 to S11, and step S12
6, the screen 4 shown in FIG. 6 is displayed to inquire about continuation of the process, and when the operator selects continuation, the process proceeds from step S9 to step S10 to advance the pointer for comparison. Thus, steps S7, S8, S9, and S10 are repeated until the next matching character string is found or the end of the sentence is reached. On the other hand, if it is the end of the sentence in step S9, the process proceeds to step S13 to enter the search mode by partial matching.

In the search mode by partial matching, first,
Substitute 3 for the "exact match search" clause (step S
13), and because L-1 = 2 (step S14),
The process proceeds from step S15 to step S16. Next, the pointer is set to 1, and the position to be compared is set to the beginning.
Then, after setting the necessary numerical values in step S17, in step S18, the position of the document data indicated by the pointer is set to
For example, if the character string "partial match search" is present, first, "partial" and "complete" are compared, and since they do not match, the process proceeds to step S21 and (MN) = 1 and (K -L) = 1 are compared, and since they are equal, step S22
Go to, and since M = 1 and K = 3, steps S24 and S1
Proceed to 8. Here, as a result of comparing the phrases “match” and “part”, they are not matched, and (MN) = 2, (KL) =
Since it is less than 1 and the matching criterion, the comparison at that position is stopped, the process proceeds to step S23, and the comparison position pointer P is advanced. At this time, it is also checked in step S27 whether or not it is the end of the sentence, and this time, it is compared with the character string "by minute match search" of the document data. In this case also, there is a mismatch, and the comparison position pointer P is further advanced to compare with the character string "in match search" of the document data.

First, the phrases "complete" and "match" are compared, and there is no match. Then, "match" and "match" are compared and they match, so N = 1 is set and the phrase is stored in the auxiliary pointer R. Add the number 2 of "match" (step S20), then
The (P + 2) th character of the character string “in matching search” of the document data, that is, “search” and the phrase “search” are compared (step S18). Since they match, N = 2 and auxiliary pointer R = 4 are set (step S20). Next, in step S21, (M−N) = (3−2) = 1, (K−L) =
(3-2) = 1 and (MN) and (KL) become equal.

When all the three clauses have been compared (step S22), partial matching is established and step S
25, and that portion is displayed as on screen 5 in FIG. Then, select whether to continue (step S2
6) If the work is not continued, the work is finished. If the work is continued,
The same process is repeated until there is a matching character string or the comparison position is at the end of the sentence.

The relaxation of the actual conditions is based on the criterion L that is regarded as a match.
Decrease by 1. Further, when L becomes 0, it is determined that the relaxation is not possible, and the process ends.

In this way, when searching for a character string that partially matches a specific character string, by using the delimiter information of the bunsetsu at the time of conversion into a kanji kana mixed sentence, a partial match in bunsetsu units It becomes possible to perform a search, and the efficiency of search work can be improved.

[0043]

According to the present invention, the following effects can be obtained. That is, since the relationship between the characters in the designated search character string is clarified, it is possible to suppress the automatic generation of conditions that the operator does not expect. Also, wasteful work,
That is, the generation of conditions that the operator does not expect and the search work are reduced, and the work efficiency and the work time can be shortened. Furthermore, since the generation of meaningless conditions is suppressed, unnecessary extraction of character strings is eliminated, and the work efficiency is improved. Then, since the phrase delimiter used at the time of conversion refers to the internal dictionary data, an efficient search can be performed by an apparatus that performs conversion using the same dictionary data.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of the present invention.

FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention.

FIG. 3 is an explanatory diagram showing a method of embedding segment break information.

FIG. 4 is an explanatory diagram showing a method of using delimiter information.

FIG. 5 is a flowchart showing the operation of the embodiment.

FIG. 6 is an explanatory diagram showing a display example of a screen.

FIG. 7 is an explanatory diagram showing determination of matching according to a conventional technique.

[Explanation of symbols]

 1 CPU 2 RAM 3 ROM 4 Printer controller 5 Printer 6 Display controller 7 Display device 8 Key interface 9 Keyboard 10 FD controller 11 FD 12 IC card interface 13 IC card 14 Address bus / data bus 15 System data storage EEPROM

Claims (1)

[Claims] 1. An input means for inputting reading information of a word over a plurality of phrases, and a kana-kanji conversion dictionary. By referring to the dictionary, the input reading information is mixed with kanji kana composed of a plurality of phrases. A conversion means for converting the sentence into sentences, and when the input reading information is converted into a kanji-kana mixed sentence, a phrase information addition means for adding the phrase delimiter information to the kanji kana-mixed sentence, and the phrase delimiter information is added. By referring to the document storage means that stores a document containing a large number of kanji and kana mixed sentences and the delimiter information of the clauses added to the kanji and kana mixed sentences, the plurality of kanji and kana mixed sentences converted by the conversion means are referred to. A searcher that searches the documents stored in the document storage unit for a kanji-kana mixed sentence that matches the kanji-kana mixed sentence of any one or more of the clauses. When, character search function document processing apparatus comprising an output means for outputting the search result by the search unit.