JP2865446B2 - Sentence processing equipment - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sentence processing apparatus provided with a kana-kanji conversion function, and more particularly to a sentence processing apparatus provided with an example dictionary for improving conversion efficiency.

[0002]

2. Description of the Related Art A conventional example dictionary is, as shown in FIG.
The three elements of the preceding word in the example (hereinafter referred to as a pre-phrase word), the word in the rear of the example (hereinafter referred to as a post-phrase word), and the particles located between the words in the example are stored as one set of information. I have.

[0003]

However, in the conventional example dictionary, if 16 types of particle information (eg,,,,,,, etc.) are provided, one example requires 16 bits of data. In the case of thousands of examples,
There is a problem that 640 kilobits are required as particle information, and the capacity of the example dictionary increases. The present invention has been made in view of the above circumstances, and provides a text processing apparatus capable of compressing the capacity of an example dictionary.

[0004]

FIG. 1 is a block diagram showing a basic configuration of the present invention. In the figure, the present invention provides an input means 1 for inputting a word reading, a word dictionary means 2 for storing a number of words including kanji in correspondence with the reading, and an effective combination of each word. An example dictionary means 3 for storing a phrase word, a post-phrase word and a combination of particle information for connecting the words, and a word candidate corresponding to the reading of the word input from the input means 1 are called from the word dictionary means 2 and It comprises a conversion means 4 for retrieving candidates and particle candidates between words from the example dictionary means 3 and thereby converting the reading of words into kanji-kana-kanji sentences, and an output means 5 for outputting the converted kanji-kana-kanji sentences. All the particle information stored in the example dictionary means 3 is coded in units of a pattern composed of one or more particles classified according to the frequency of appearance. One or a plurality of coded particle information is stored in association with one combination of words, and the conversion means 4 further searches all the particles represented by the one or more codes. A sentence processing apparatus characterized by being targeted. Further, the present invention provides an example dictionary of a sentence processing apparatus which is referred to in order to obtain the validity of a combination of a preceding phrase word and a subsequent phrase word when converting an input word reading into a word including a corresponding kanji. A first memory for storing a pre-phrase word and a post-phrase word
And the particle information that connects the pre-phrase word and the post-phrase word
Consists of one or more particles classified by current frequency
The second is to encode and store the pattern to be processed as a unit.
A combination of a memory and a word stored in the first memory
In the second memory of the particle pattern corresponding to each of
It is composed of an address pointer indicating a storage location.
A pre-phrase word and a post-phrase word in one memory;
One or more particles in the memory are
And stored by associating them with pointers.
An example dictionary is provided.

As the input means 1, conversion means 4, and output means 5 in the present invention, those known in the art can be used.
For example, the input means 1 only needs to be capable of inputting a kana character string, and can be constituted by a keyboard, a mouse, a tablet input device, an OCR, or the like. The conversion means 4 includes a microprocessor as a main component, a program memory storing a program for controlling the microprocessor, and the like. Output means 5 is CRT, LCD
Etc., and the output means 5 includes a printer.

[0006]

According to the present invention, when a kana character string is input from the input means 1 and a conversion key is pressed, a conversion candidate word string is read out from the word dictionary means 2 and then the example dictionary means is referred to. , A pre-phrase word candidate and a post-phrase word candidate are selected, and further, a particle candidate connecting the pre-phrase word and the post-phrase word is selected from the collation result with the particle information stored from the frequently occurring particles. Act as if

[0007]

DETAILED 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 by this. The example dictionary in the kana-kanji conversion is intended to reduce the total number of key operations by enabling the conversion result to be converted at once without selecting the next candidate.
For example, if a kana character string such as “Kissen” is input and converted, the notation corresponding to “Kichi” will be “Kiki”, “Known”, “Base”, “Kiichi”, “Dangerous land”. There are many homonyms (hereafter homonyms) such as etc., and simply considering the frequency of the number of conversions of a word increases the number of key operations required to select the homonym. ,
In this case, the correct conversion result can be obtained by one conversion operation by storing as an example such as "No, No" or "Construction".

The present invention relates to an improvement in the above-mentioned example dictionary structure, and greatly improves the memory efficiency as compared with the example dictionary having the conventional configuration shown in FIG. FIG. 2 shows a configuration of a word processor to which one embodiment of the present invention is applied. Reference numeral 10 denotes an input device including a keyboard. Reference numeral 11 denotes a character string control device mainly composed of a CPU chip. Reference numeral 12 denotes a storage device that includes a ROM, a RAM, a floppy disk, or the like, and stores processing data and control information. Reference numeral 13 denotes a display device that includes a CRT, an LCD, and the like, and that displays a processing result. Reference numeral 14 denotes a kana-kanji conversion processor for converting a kana character string input from the input device 10 into a kanji-kana mixed sentence. Reference numeral 15 denotes a dictionary search device that searches a conversion dictionary 16 used for kana-kanji conversion. The conversion dictionary 16 includes a word dictionary 161 that stores word information, and an example dictionary 162 that stores a combination of words, particles, and words.

FIG. 3 shows the structure of the example dictionary 162 described above. What differs from the configuration of the conventional example is the portions 34 and 35. That is, 35 is a particle data memory storing particle data of the example, and 34 is an address pointer indicating the storage location of the particle data of the corresponding example word.
Taking the frequency of the appearance frequency of the particle data of each example with respect to all the examples, it is found that there is a large deviation in the frequency distribution as shown in FIG. 4, and this is used to construct the example dictionary 162. ing. That is, the particle data memory 35 is arranged in the form of arranging particle data in descending order of frequency, for example, arranging up to the 56th frequency, followed by particle patterns necessary for synthesizing low frequency data. I have.

FIG. 5 shows an example of a particle pattern that can be synthesized in a pattern up to the 56th degree. In the case of the example “store” “sales”, the particle patterns are [de ”, nothing, reverse]. These are particle patterns [none] (frequency 2nd) and [de,
Inverse] (frequency 19th).

FIG. 6 shows an unsynthesizable particle pattern [Ni,
To]. In this case, the pattern of [,, to] is stored as it is at the 57th place or higher. Here, the particle pattern [none] indicates that the words are directly connected to each other without intervening particles between the example words, and [reverse] means that the words of the preceding phrase and the following phrase of the example are not connected. Indicates that connection is possible even if is reversed.

FIG. 7 shows an example of the frequency of a particle pattern. In the drawing, [+] in the particle pattern indicates that the connected particle is not particularly limited (particle free). For example, when a kana character string “Kichi is sent” is input, the character string control means 11 stores the character “ki” and “chi” in the storage means 12 as a character string to be converted for each input. The input character string is displayed on the display means 13. When the conversion key is pressed in this state, the character string control means 11 activates the kana-kanji conversion processing means 14, and some conversion candidates, for example, "build a known", "build a wit", and "build a base". "Construct" is generated. Thereafter, the kana-kanji conversion processing means 14 searches the example dictionary 162 shown in FIG. A specific example will be described below.

Pre-phrase word 3 using pre-phrase word “known” as an example
1 and temporarily stores the address indicated by the address pointer 32 to the succeeding phrase word at the corresponding position and the value of the address pointer immediately below, and stores the subsequent phrase word 33 existing between the two address pointers. Search for. However, since the pattern of “known” and “construction” is not registered in the example dictionary in this search, the search failed, and only the combination of “base” and “construction” succeeded among the candidates, which is It remains as a conversion candidate.

After "base" matches "construction", the address pointer 3 of the particle information corresponding to the later phrase word "construction"
4 and the particle information indicated by this pointer is
5 and temporarily store the particle data. Since there are cases where two particle data are combined, the data corresponding to “construction” of the later phrase word 33 is compared with the word after “construction”. In this case, since the particle is a combination of [,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, / B / F / F-A-B-C The retrieval of the particle data is completed by taking two ORs.

The particle data is compared with the particle of the conversion candidate, and "" is matched, and as a result, "construct a base".
Remain as the first conversion candidates and are displayed on the screen by the display unit 13. The above is a specific example, and the processing flow is shown in FIG. In FIG. 3 showing the structure of the example, the pre-phrase word 31 of the example and the address pointer 32 to the post-phrase word of the example are connected after the pre-phrase word 31 of the example if the head position is known. No need. The same applies to the relationship between the post-phrase word 33 of the example and the address pointer 34 to the particle information of the example. By applying the present invention, in the case of 67,000 actual measurements, a dictionary capacity of about 800 kilobits can be compressed as compared with the conventional example dictionary.

[0016]

According to the present invention, a table in which particle patterns that can be connected between example words are arranged in the order of frequency of occurrence of particles, instead of simply having 16 types of bits, is prepared. For those having a low frequency, by synthesizing a pattern having a high frequency so that particle information can be searched, the memory capacity of the example dictionary can be significantly reduced.

[Brief description of the drawings]

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

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

FIG. 3 is an explanatory diagram showing a configuration of an example dictionary according to the embodiment.

FIG. 4 is a table showing a frequency distribution of particle patterns in an example dictionary.

FIG. 5 is an explanatory diagram showing an example of dividing a particle pattern.

FIG. 6 is an explanatory diagram showing an example in which a particle pattern cannot be combined.

FIG. 7 is an explanatory diagram showing the frequencies of all examples of particle data.

FIG. 8 is a flowchart illustrating the operation of the embodiment.

FIG. 9 is an explanatory diagram showing a configuration of a conventional example dictionary.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input means 2 Dictionary means 3 Example dictionary means 4 Conversion means 5 Output means

Claims (2)

(57) [Claims] 1. An input means for inputting a reading of a word, a word dictionary means for storing a number of words including kanji corresponding to the reading, and an effective combination of each word, An example dictionary means for storing a phrase word and a combination of particle information for connecting words to each other; a word candidate corresponding to the reading of a word input from the input means being called from the word dictionary means; From the example dictionary means, thereby converting the reading of the word into a kanji kana mixed sentence, and an output means for outputting the converted kanji kana mixed sentence, and stored in the example dictionary means. All of the particle information is coded in units of a pattern composed of one or more particles classified by the frequency of appearance, and is combined into a single word. And one or a plurality of coded particle information are stored in association with each other, and the conversion means further searches all particles represented by the one or more codes. A sentence processing device characterized by the following. 2. A sentence processing apparatus referred to in order to obtain the validity of a combination of a preceding phrase word and a subsequent phrase word when converting the reading of an input word into a word including a corresponding kanji.
An example dictionary of the first embodiment, wherein a first memory storing a pre-phrase word and a post-phrase word, and particle information connecting the pre-phrase word and the post-phrase word,
Consists of one or more particles classified by degree
Second memory for coding and storing patterns in units
And each of the combinations of words stored in the first memory
The storage location of the corresponding particle pattern in the second memory is
A first address word , a second address word, and a second address word in the first memory.
The one or more particles in the second memory
The fact that the information is stored in association with the address pointer
Example dictionary to be used as a symbol .