JPH02292662A - Document processor with proper noun converting function - Google Patents

Abstract

PURPOSE:To improve operability in proper noun conversion by reading out two proper nouns, and batch-converting inputted read information as one proper noun only in the case where these two proper nouns are of combination capable of being coupled. CONSTITUTION:In the case where two corresponding proper nouns are retrieved from a proper noun dictionary 10, pattern check to check whether an attribute stored in each information area of two proper noun is included in patterns capable of being coupled stored in a coupled pattern storage part 11 or not is executed. Then, in the case of an effective pattern, on the assumption that a conversion candidate exists, a character string from the start position to the clause pointer of an input KANA (Japanese syllabary) character string and the character string from the clause pointer to an end position are converted respectively into the proper nouns, and batch-conversion is executed as considering them one proper noun by connecting those. Thus, only when the input KANA character string constituted of two proper nouns stored in an input buffer 5 is in the effective combination capable of being coupled, it is batch- converted as one proper noun.

Description

DETAILED DESCRIPTION OF THE INVENTION (A) Field of Industrial Application The present invention relates to a document processing device equipped with a proper noun dictionary for converting input reading information into proper nouns.

(CJ) Conventional technology Conventionally, in this type of document processing device, in addition to a general dictionary that registers all so-called single edicts, such as independent words and adjunct words, including common name-1, name It is equipped with a dictionary for registering special proper nouns such as , place names, etc., and when using this proper noun dictionary, conversion is performed in units of phrases.

For example, the reading information “Yamada゜” is “Yamada”.
When you want to convert it into a proper noun, use “yamada”
By inputting the character string ``Yamada'' and pressing the conversion key, the proper noun ``Yamada'' is read out from the proper noun dictionary.

(c) Problems to be solved by the invention However, in such conversion in units of phrases, 2
Reading information that connects two proper nouns, for example, reading information such as "Yamada Taro", cannot be converted all at once to "Yamada Taro", so enter "Yamada" one phrase at a time. There was a problem in that the user had to enter "Taro" again to convert it to "゜Yamada" and then convert it to "Taro", making the operation cumbersome.

In addition, when converting proper nouns consecutively, they are unconditionally connected and converted, so for example, "famous person's surname" or "town name of 10 prefectures" may be converted in a situation that is impossible in reality. There have been cases where the words have been converted into meaningless words.

This invention was made in consideration of the above circumstances, and it connects two words in a proper noun dictionary and converts them into one proper noun. During the conversion, the connection between the two words is checked. The present invention provides a document processing device with a proper noun conversion function that prevents words from being converted into meaningless words.

(2) Means for Solving the Problems FIG. 1 is a block diagram showing the construction of this invention. As shown in the figure, this invention has a dictionary section 101 that stores a large number of proper nouns together with their pronunciation information. and a combination U/storage means 102 that stores combinations of proper nouns that can be combined with each other.
An input information storage means 104 that stores the reading information input from the input means 103, a setting means lO5 that sets a clause position in the reading information stored in the input information storage means 104, and a clause position set by the setting means +05. Dictionary part 1 stores two proper nouns corresponding to the previous and subsequent continuation information.
01, reading means 106, and reading means 106.
A search means 107 searches whether the combination of two proper nouns read out by is included in the combinations stored in the combination storage means 102, and when the combination of two proper nouns is included as a result of the search by the search means 107, This is a document processing device with a proper noun conversion function, which includes a conversion means 10g that converts the reading information stored in the input information storage means 104 into one proper noun by connecting two proper nouns.

Note that the dictionary section 101 and the storage means 1 in this invention
02 may be any device that can store a large number of proper nouns together with their pronunciation information, or that can store combinations of proper nouns that can be combined with each other. For example, a Uiro storage medium such as a floppy disk device or a magnetic disk device, ,R
An internal memory such as OM is used.

Further, the input means 103 may be any device that can input kana characters as reading information, and for example, a keyboard device, a tablet device, or the like may be used.

Further, the entering information storage means 104 may be any device as long as it can store the kana characters input from the input means 102, and usually a RAM such as a core memory or an IC memory is used.

It is generally convenient to use a microprocessor as the setting means 105, the reading means 106, the searching means 107, and the converting means 10g.

(E) Effect As shown in FIG. 1, according to the present invention, the reading information stored in the input information storage means 104 has a bunsetsu position set by the setting means 105, and is read by the reading means 106 before the bunsetsu position. 2 corresponding to each reading information of and after.
One proper noun is read from the dictionary section 101.

Then, the search means 107 searches whether the combination of the two proper nouns is included in the combination storage means 102, and when the combination of the two proper nouns is included, the conversion means 108 searches the input information storage means 102.
The reading information stored in is converted into one proper noun in which two proper nouns are connected.

Therefore, it is possible to combine two words in a proper noun dictionary and convert them into one proper noun, and what is more, the word that is converted as one proper noun is a meaningless connection of proper nouns. This will be prevented.

(f) Examples Hereinafter, the present invention will be described in detail based on examples shown in the drawings. Note that this invention is not limited by this.

Figure 2 is the fruit of this invention M! FIG. 1 is a block diagram showing an example of a block diagram of a block diagram;

In this figure, 1 is a keyboard equipped with a kana key, a conversion key for converting kana into a fixed noun, etc., and kana characters as 95-character gorge information are input into the keyboard 911-ro 2.

The control unit 2 is composed of a microprocessor, and has 110
In accordance with the system 911 program stored in the program memory 3 consisting of M, various data processing to be described later is performed.

4 is I? AM, and internally includes an input buffer 5 that temporarily stores input data such as kana character strings input from the keyboard 1, and a text buffer 6 that stores sentences converted from kana character strings into sentences with kanji and kana. is provided.

7 is a display device consisting of a CR i' display, an LC (liquid crystal) display, an EL display, or the like.

Reference numeral 8 denotes an IIT section consisting of an external memory such as a disk drive, a magnetic disk drive, r{OM, etc.

This dictionary 1!8 includes a general dictionary 9 that stores all so-called words such as independent words and adjunct words that are used in ordinary document processing devices, including old joint names, and a general dictionary 9 that stores frequently used proper nouns in their pronunciations. A fixed f1 noun dictionary IO that is stored together with information fu, and a combination pattern storage section 11 that stores combinations that can be combined between proper nouns are set.

Fixed a noun dictionary! 0, as shown in Figure 3, has areas for entry words, registered edicts, and information, with reading information in the entry word area and unique information in the registered word area. For example, the notation of the noun is displayed in the information area.
If the notation is “Taisho”, the “ward name゜ identification mark,
In addition, if the notation is "Yamada", the identification mark of "family name" is stored, and the presence of each noun is memorized.

As shown in FIG. 4, the combination pattern memory 11 stores, as a table, patterns that can be combined between a song and two subsequent solid nouns.

Based on this table, for example, [Yuko Taishi,
When a kana character string of kuJ is input and transmuted by the Koshi noun dictionary 10, it is possible to convert it to the notation "Yukou Taisho", but the headword "Yukou" shown in Figure 3 can be considered. The information for the headword ``Taisho゜'' is the ``ku name'', and the combination of ``first name'' and ゜ku name'' is not included in the table shown in Figure 4. Conversion to "Taisho" will not be carried out, and there will be no conversion pine supplement.

On the other hand, when the kana character string "Yamata Taro" is input and converted by the proper noun dictionary 10, the information for the headword "Yamada" shown in FIG. ti"Jho is a "first name", and the combination of this ゜surname" and "first name" is included in the table shown in Figure 4, so it is possible to convert ``Yamada Taro'' into the notation ゜Yamada Taro''. It is possible.

The control unit 2 stores the kana character string input from the keyboard 1 in the large buffer 5, and uses the general dictionary 9 or the proper noun dictionary 10 in the dictionary 91 section 8 to input the kana character string input into the input buffer 5. The string is converted into a character string containing kaji and kana characters, that is, a so-called normal sentence r3, and stored in the sentence buffer 6.

” In the conversion using Futami's unique master craftsmanship 4 dictionary 10,
The controller 11 unit 2 first temporarily sets a clause position for the input kana character string stored in the input buffer 5.

For example, if the character string "Yamada darou" is stored in Hitokababufa 5, this setting will search all the headings for "ya" in the proper noun dictionary 10, and search for "Yamada tarou". This is done by finding the so-called longest match position, in which the phrase is set at the end of the longest word whose pronunciation matches the character string .

As a result, if a search is made for the heading ``Yamada'', for example, ``Yamada/Taro'' will be searched.
Set the phrase pointer to the phrase position indicated by the “/゛ symbol, and then set the unique character string (“Taro” in the above example) that corresponds to the character string after the phrase pointer, that is, from the phrase pointer to the last position (end position). Nouns are searched from the proper noun dictionary IO by heading.

At this time, if the corresponding proper noun is not registered and cannot be searched, the position of the clause pointer is moved one position to the beginning of the input Kana character string, and in the above example, it is set to ゜yama/dadarou''. Next, the proper noun dictionary 1O is searched for a proper noun corresponding to the character string before the clause pointer, that is, from the first position (start position) of the input kana character string to the clause pointer (in the above example, "Yama").

In this search, if the corresponding proper noun is not registered, the position of the phrase pointer is moved one by one to the beginning of the input character string, and the character string from the start position to the phrase pointer is Search for the corresponding proper noun.

If the corresponding proper noun is retrieved, then the proper noun corresponding to the character string after the phrase pointer is retrieved from the proper noun dictionary ■0.

In this way, control 9111 part 2 receives input buffer y5
Set the clause position for the input kana character string stored in , and set the proper nouns corresponding to each kana character string before and after the clause position to the first proper noun (in the above example, "Yamada"). ”) and the second proper name (“Taro゜” in the above example) from the proper noun dictionary 6, the identification mark stored in each information area of the first proper name and second proper noun. In other words, the pattern is checked to see if it is included in the combinable patterns stored as a table in the combination pattern storage unit 1! shown in FIG. The first proper noun and the second proper noun are connected and converted into one proper noun (in the above example, "Yamada Taro").

The contents of the processing operation of the control section 2 will be explained according to the flowchart shown in FIG. 3.

When converting an input kana character string stored in the human buffer 5 into a proper noun, first search the proper noun dictionary IO and find the position of the longest match for the input kana character string (
Step 31).

Once the position of the longest match is determined, a phrase pointer is set at that position (step 32), and then a proper noun corresponding to the character string from the phrase pointer to the end position is searched in the proper noun dictionary 10 (step 33). , proper noun dictionary 1
It is determined whether it is at 0 (step 34).

In this determination, the corresponding proper noun is
If it is not 0, the position of the clause pointer is moved one position to the beginning of the input kana character string (step 35), and it is determined whether the position of the clause pointer is the start position of the input kana character string (step 35). 36).

Here, if the position of the phrase pointer is not the start position of the input kana character string, the proper noun corresponding to the character string from the start position to the phrase pointer is searched from the proper noun dictionary 10 (step 37), and 10 (step 38).

In this determination, the corresponding proper noun is
If it is not in O, step 35 until it is found.
- Step 38 is repeated, and the position of the phrase pointer is successively changed to the beginning side until the phrase pointer position reaches the start position of the input kana character string, and the search is repeated.

Then, in the judgment of step 36, if the position of the phrase pointer is at the start position of the input kana character string,
The proper noun corresponding to the input kana character string is a proper noun dictionary lO
In other words, there is no conversion candidate (step 39), and the conversion process ends.

In addition, in the determination at step 38, if the corresponding proper noun is found in the proper noun dictionary 10, the process returns to step 33 and the proper noun corresponding to the character string from the bunsetsu pointer to the end position is searched again. A search is made from the noun dictionary IO, and steps 33 to 38 are repeated until two proper nouns corresponding to each kana character string of the phrase pointer いもえ and subsequent are found in the proper noun dictionary IO.

Then, in the determination at step 34, if two corresponding proper nouns are retrieved from the proper noun dictionary IO,
A pattern check is performed to determine whether the attributes stored in each information area of the two proper nouns are included in the combinable pattern stored in the combination pattern storage unit If shown in FIG. 4 (step 40). , if the pattern is valid (step 4I), the character string from the start position to the phrase pointer and the character string from the phrase pointer to the end position of the input kana character string are uniquely identified as conversion candidates. The nouns are converted into nouns, and they are connected to form a single proper noun for batch conversion (step 42).

In this way, the input kana character strings consisting of two proper nouns stored in the input buffer 5 are collectively converted into one proper noun only when they are a valid combination that can be combined.

(g) Effect of the invention According to this invention, a bunsetsu position is set for input reading information, two proper nouns corresponding to each reading information before and after the bunsetsu position are read out, and they are combined. We verified whether the combination is possible, and only when the combination is combinable, convert all the input reading information into one proper noun, so the dictionary section can display two independent proper nouns. Even if it is remembered as
It is possible to enter two proper nouns at once and convert them all at once, and it also prevents two proper nouns from being converted as meaningless words. The operability in noun conversion can be improved.

In addition, the dictionary section includes, for example, mountains, mountain ranges (...mountains, ... mountain ranges, etc.), universities, high schools (...
・University. If the common noun ``high school, etc.'' is memorized and the combination U゛ storage means is made to memorize possible combinations of proper nouns and that common noun, the number of words registered in the dictionary can be reduced by a large number of words. For example, you can convert proper nouns and common nouns such as ``Hidaka Mountains'', ゜Ou Mountains'', or ``University of Tokyo'' and ``Osaka High School'' without increasing the number of names other than surnames and place names. It also makes it possible to convert a wide range of proper nouns.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the configuration of this invention, Fig. 2 is a block diagram showing an embodiment of this invention, Fig. 3 is a schematic diagram of the stored contents of a proper noun dictionary, and Fig. 4 is a combination pattern memory. FIG. 5 is a flowchart showing the operation of the embodiment. 1...Keyboard, 2...Control unit, 3
...Program memory, 4...RAM
, 5... Human cover, 6... Text buffer, 7... Display device, 8... Dictionary section, 9... General dictionary. , 10... Proper noun dictionary, ■! ......Connection pattern storage section. 7 1 Figure 2 Figure Figure Figure Figure

Claims (1)

[Claims] 1. A dictionary section that stores a large number of proper nouns together with their pronunciation information, a combination storage means that stores combinations of proper nouns that can be combined with each other, and an input information storage means that stores the pronunciation information input from the input means; A setting means for setting a phrase position in the reading information stored in the input information storage means, and a readout for reading two proper nouns corresponding to the reading information before and after the phrase position set by the setting means from the dictionary section. means, a search means for searching whether the combination of two proper nouns read out by the reading means is included in the combinations stored in the combination storage means, and a combination of the two proper nouns as a result of the search by the search means. 1. A document processing device with a proper noun conversion function, comprising a conversion means for converting reading information stored in an input information storage means into one proper noun by connecting two proper nouns.