JPH0540748A - Word processor - Google Patents

Abstract

(57) [Summary] [Object] The present invention relates to a document creation device for performing kana-kanji conversion, and an object thereof is to simplify learning registration. [Structure] A dictionary means for storing word candidates including word numbers, reading information, and notations, and adjunct word candidates including adjunct numbers, reading information, notations, and the like, and cancels and reconverts erroneously converted character strings. Range specifying means for specifying a range for reading, a number reading means for reading a word number and an attached word number from a dictionary means for a correct word string obtained by re-conversion within the specified range, and a read word It is characterized by comprising a long word dictionary means for storing a sequence of numbers and attached word numbers, and a reading means for preferentially reading a word corresponding to the input reading information from the long word dictionary means.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document preparation device,
More specifically, the present invention relates to a document creation device having a function of learning a conversion result.

[0002]

2. Description of the Related Art Conventionally, in a document preparation apparatus having a learning function, when erroneously converted contents are canceled and correct words are obtained by reconversion, the words are read as "headings" as reading information. It is registered in the dictionary separately from the "registered word". Specifically, for example, when the reading "do not" is erroneously converted to "city expense" and reconverted to "do not", the heading "do not" and the word "do not" correspond to each other. The word “non-person” registered in the dictionary is registered in the dictionary by a character-based code string.

[0003]

However, in the method of registering a headline and a notation composed of a word string by the above-described conventional document creation apparatus, the problem that the registration process becomes complicated, and the headline and the notation are registered. However, there is a problem that the capacity of the dictionary becomes large.

The present invention has been made in consideration of the above circumstances. Instead of registering a headline and registering a character code for each character in a word string, the word string is stored as a sequence of word numbers and attached word numbers. By doing so, it is possible to provide a document creation device that can realize learning registration with a small number of operations and a small dictionary capacity.

[0005]

FIG. 1 is a block diagram showing the basic configuration of the present invention. In the figure, the present invention is directed to a document creation device configured to perform kana-kanji conversion by reading one or a plurality of word candidates corresponding to input reading information and selecting one of the read word candidates. , The word means consisting of word numbers, reading information, notations, etc. and the adjunct word candidate consisting of adjunct word numbers, reading information, notations, etc. and the dictionary means 1 for canceling and reconverting the erroneously converted character string. A range specifying means 2 for specifying a range for reading, a number reading means 3 for reading a word number and an attached word number from the dictionary means 1 for a correct word string obtained by re-conversion within the specified range, and read out. Long word dictionary means 4 for storing a sequence of word numbers and attached word numbers, and reading means 5 for preferentially reading words corresponding to the input reading information from the long word dictionary means 4. A document creating device including a.

[0006]

According to the present invention, when a word that has been erroneously converted is converted into a correct word by re-converting, the sequence of word numbers and adjunct word numbers that form the correct word is long word dictionary means 4.
Memorized in, and then input the reading of the word to do the conversion,
First, it is checked whether or not the reading corresponds to a series of word numbers and auxiliary word numbers stored in the long word dictionary means 4, and if they match, the word corresponding to the series is read from the dictionary means 1. If they do not match, one or more word candidates are read from the dictionary means 1.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail based on the embodiments shown in the drawings. The present invention is not limited to this. FIG. 2 is a block diagram showing the configuration of a Japanese word processor to which the present invention is applied. In the figure,
10 is an input device including a keyboard,
Input kana character strings as reading information, and perform editing operations such as conversion, cancellation, cursor movement, etc.

Reference numeral 11 is a character string control device for judging information sent from the input device 10 and performing a series of control.
It is composed of a chip CPU, ROM, RAM and the like. As shown in step (1) of FIG. 3, when a kana character string is input, the character string control device 11 stores the character string in the work buffer 121 of the storage device 12, and the result is displayed on the LCD or CRT.
It is output to the display device 13 including the above. The above-mentioned processing is repeated for the kana character string "a person who doesn't even do that" in this embodiment.

When the conversion key of the input device 10 is pressed, the character string control device 11 causes the kana-kanji conversion processing device 14 to operate.
And the kana-kanji conversion processing device 14 activates the dictionary device 15
The information of the word written in is read and a conversion candidate is generated.

The dictionary device 15 is composed of a ROM chip because it needs to be searched at high speed. Therefore, the body dictionary itself cannot be changed by the operator. Therefore, a word existing in the main dictionary is used for learning registration as a new word, and when the same reading is input again, the learning registered word is output as the first candidate for conversion, and conversion efficiency is improved. Is configured to increase.

When generating a conversion candidate, the long word learning buffer 122 in the storage device 12 is referred to and compared with the conversion candidate, and if they match, the candidate is output as the first candidate.

The procedure (2) in FIG. 3 is a display example of the conversion result. At this stage, the kana character string "A person who doesn't even do that"
Has not been learned. If cancellation, retreat, or deletion is performed in this state, the procedure (3) in FIG. 3 is performed by the same control procedure.
The display contents are as shown in.

When the function + non-conversion key of the input device 10 is pressed, the character string control device 11 stores the position of the range designation in the work buffer 121 of the storage device 12, and the characters ""And" and "are displayed in reverse video.

When the function + space key is pressed in this state, the character string control device 11 causes the reading (kana character string) and the word number stored in the work buffer 121 of the storage device 12,
It is checked whether a series of adjunct word numbers are associated, and if they are associated, the word number,
The series of attached word numbers is stored in the long word learning buffer 122 of the storage device 12.

The long word registered for learning is always referred to in the subsequent kana-kanji conversion, and when it matches the conversion candidate, it is output and displayed as the first conversion candidate. In the above embodiment, the case where the word “shi”, the adjunct word “not”, and the word “person” are learned as long words has been described as an example.
It is not limited to one word, but may be three or more words. Similarly, the number of attached words is not particularly limited.

FIGS. 4 and 5 show the above-mentioned word numbers and attached word numbers. The dictionary device 15 has reading information, notation information, and part-of-speech information for each word number, and when the notation is selected from the reading of a word, the word number can be checked. The auxiliary word numbers have the same structure.

FIG. 6 is a flow chart for explaining the control operation of the character string control device 11 described above. First, when an input is made from the input device 10 (step 30), it is judged whether or not it is a conversion key (step 31), and the judgment result is N.
In the case of O, it is determined whether or not the range designating key has been pressed (step 32). In the case of YES, the range-designated character string is stored in the work buffer 121, and the first and last characters in the range-designated character string are stored. Is highlighted (step 33).

Next, when the function of the learning registration key + the space key is pressed (step 34), it is judged whether or not the range is designated (step 35). If YES, then all the characters within the range are determined. It is judged whether or not it has been converted (step 36), and when the registration display state is NO, that is, when the key such as cancel is not pressed (step 37), the continuous word number is set to the long word learning buffer 1.
The process of registering in 22 is executed (step 38), and the registered contents are displayed (step 39).

When a character is next input and the conversion key is pressed while the continuous number is registered, kana-kanji conversion is executed (step 40), and first, the connected word having the registered connection number is referred to. If there is a matching candidate (step 41), the candidate is determined as the first conversion candidate (step 42) and displayed on the display device 13 (step 39).

When a normal character string is input in step 34, the character string is stored (step 43 →
44), if the cancel key is pressed, cancel processing is performed (steps 45 → 46), and if the cursor movement key is pressed, cursor movement processing is performed (steps 47 → 48),
When another key such as an edit key or a special key is pressed, other processing is performed (step 49). The processing of these steps 43 to 49 is the same as the conventional processing.

Further, if YES in step 37, for example, if the cancel key is pressed, cancellation processing of consecutive word numbers is performed (step 50).

[0022]

According to the present invention, when a meaningless character string is generated by Kana-Kanji conversion and is correctly converted, learning registration is performed without performing a special operation for registering it, and subsequent Kana-Kanji conversion is performed. In the case where the same character string appears again in, the learning-registered word string is output as the first candidate, so that the number of operations for conversion can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing a basic 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 display example by learning registration according to the embodiment.

FIG. 4 is an explanatory diagram showing word numbers in the embodiment.

FIG. 5 is an explanatory diagram showing accessory word numbers of the embodiment.

FIG. 6 is a flowchart illustrating the operation of the embodiment.

[Explanation of symbols]

 1 dictionary means 2 range designation means 3 number reading means 4 long word dictionary means 5 reading means

Claims (1)

[Claims] 1. A document creation apparatus configured to perform kana-kanji conversion by reading out one or more word candidates corresponding to input phonetic information and selecting one of the read word candidates. , Dictionary means storing word candidates consisting of reading information, notation, etc., and adjunct word numbers consisting of adjunct numbers, reading information, notation, etc., and a range for re-conversion by canceling erroneously converted character strings. Range specifying means to specify, number reading means to read word number and adjunct word number from dictionary means for correct word string obtained by re-conversion within the specified range, read word number and adjunct word number A document creation device comprising a long word dictionary means for storing a sequence of words and a reading means for preferentially reading a word corresponding to input reading information from the long word dictionary means. .