JPS60189071A - Electronic dictionary - Google Patents

Abstract

PURPOSE:To compress data in data memory by adding hyphenation information to only words which consists of a specific number of characters or more in an electronic word dictionary in which words and their hyphenation positions are registered. CONSTITUTION:An input device 1 inputs character and word information to a storage device 2 and spelling information on a word inputted to the device 2 and hyphenation information on a word consisting of >=6 characters are stored in the electronic word dictionary 3. The dictionary 3 has an index part 1, where the numbers of all registered words are stored while classified by one-character, two-character ... eight-character words and also classified by the kinds of initials, i.e. alphabetically. Further, the dictionary 3 has memory parts for spelling check data and hyphenation data and memory parts J1, J2...J8 corresponding to indexes 1 classified by the numbers of words. Thus, the dictionary 3 is constituted and hyphenation information is added only words consisting of >=6 characters, thus compressing the contents of the data memory.

Description

[Detailed description of the invention] Technical fields> The present invention relates to an electronic word dictionary.

<Prior art> Conventional electronic word dictionaries that handle Western languages have a hypl+enate function, which means that words placed in the right margin of a sentence are placed in the middle of the word without moving the entire word to the next line or below. There is no function to input a hyphen (0 sphen) and express it in multiple lines, so the operator must remember the hyphen position of the word in his head and type it, or check the hyphen position in a dictionary and enter it. Was.

Therefore, inputting hyphens is troublesome, and
In English and American, the position of the hyphen may be different for the same word tt+ (for example, in the word [5structure
In English, it is a struct-ure (a morphological break), and in American English, it is a 5-truc-lure (a syllabic break). ), which is increasingly troublesome.

Therefore, the present inventor has developed an electronic word dictionary that has a hyphenate 142a function added in addition to the word spell check function, or hyphenate information is not necessarily required for all recorded words. An example is the word “ahead J
For example, if the number is two, add a hyphen and write ra-hea.
It is possible to divide it into two parts, dJ, but this hyphenate is not appropriate.

<Purpose of the Invention> Therefore, in the present invention, in an electronic word dictionary in which the separator (hyphen) positions of words as well as single and ill are registered, the words to which separator information is attached are only words consisting of a specific number of characters -1-. The data memory was compressed as follows.

<Example> The configuration of the present invention will be described below along with examples.

The electronic single word dictionary according to the embodiment of the present invention has word information for spell checking and hyphen position information only for words consisting of six or more characters.

FIG. 2 is a block diagram of a word search system having an electronic dictionary on the Kimio side.

In the figure, 1 is an input device, 2 is a storage device, 3 is an output device, 4 is an audio guide, and 5 is a control device.

Next, each of the above components will be explained.

Input device 1 - A device for inputting character/word information into the storage device 2. Specifically, it is a keyboard device, a tufflet device, an OCR, etc.

Storage device 2 - A device for storing character information input from the input device 1. Specifically, they are core memory, IC memory, etc.

Output device 3: A device for outputting the results saved and edited by the storage device 2. Specifically, they are printers, display devices, etc.

Dictionary 3: A device that stores spelling information for words stored in the storage device 2 and hyphen position information for words with six or more characters. Specifically, it is ROM (read only memory) and the like.

Control device 5: A device for controlling the transmission of signals between each component.

The configuration of the dictionary 4 described above is as shown in FIG.

In the same figure, ■ is an index part, and the index I first classifies all the words into the number of letters, that is, 1 letter, 2 letters, 8 letters, etc., and then the initial letter 〢A ,B,C,・,X,Y.

The words are classified by Z and the number of words after the classification is registered. In the same figure, from the top, one-letter words, two-letter words, etc.
, is an 8-letter word, and the letter "A".

"BJ, rc", . . . , rXJ, rYJ, rZJ represent the initial letters of words. Each index stores the number of registered words.

Also, in the same figure, ■ is a memory for word spell check data and high digit data, and memos corresponding to index I for each number of words are stored.
Consists of 8.

In addition, information from the first letter of each word to the second letter below is 1.
It is converted and stored in units of 5 bits per character.

The human conversion is shown in Table 1 below. The character "," in the same table is an apostrophe.

Table 1 Here are some specific examples.

When storing the 8 words in Table 2 below, the words to be memorized in Table 2 are rMJ for AMJ, rNJ for word 1 ANJ, rLLJ for word ``ALL'', rRMJ for word ARM, J, and rRMJ for word ``ALL''. rPLA, NJ teha rLANj, word r P L A Y J teha [LAYJ, word rD
EL1cATEj is "ELICATE", word rDE
LIVERYJ is rELIVERYJ.

Then, a meaningless take X (bit off) is added to match the memory's 1 byte-8 hint.

That is, r M J→0IIOIXXX →681((Hexaclecima6)rNJ→0
IIIOXXX -= 70 H r L L J → 0110001+00XXXXXX-
Snoring 6300H r RM'J-+ 1001001101XXXXXx
→9340H r L A N J → 0I100OOOOIOIIIO
X→605CH 1'-LA, YJ→0110000001 +1001
X → 6072H rELIcATEj → 00'1010110001
0010001 10000 + 10100001 010 01 100XXXXXXX → 2B1280DOA600 rELIVERYJ → oo+o+o+1ooo+C
l0IIO] ], 00010 110010110010 ] 001 0XXXXXXX -+ 2B1362CB2900 Among the above words, the words in the 6th sentence and below are IDEL I
CATE j and rDELIVERYJ.

In this embodiment, the hyphenate information is put together at the end of the spelling characters with a pitch of about 14p. That is, high 1
If there is a hyphen, it is rlJ, and if there is no hyphen A, it is "0".

In addition, in this embodiment, as a general rule, there is no hyphen before the first letter (initial letter) or the last letter of one word, so the hyphen Δ −
2) I need a hint.

”-Jl'4? -4r DELICATE J 04
In this case, if the character separation is represented by "○", it becomes DOEOLOI 0COAOTOE, or the positions where the hyphen can be set are r DE +--1-CA, TE J, so the -High Two A E1- information is , change the hyphen setting position to "
○", and the non-hyphen setting position is "・", and it becomes D OE OL @ I @ COA OT E, so it is 01100 (this is 6 bins of 1) information).

In addition, in the case of ``,rx r DEL+VERy'', if the character separation is represented by ``○'', it becomes DOEOl-○l0VOEOROY, but the possible hyphen setting values are IDE-LIV-ER-YJ, so The hyphen information is 10010 because the hyphen setting position is "○" and the hyphen non-setting position is "・", and it becomes DOE@LOI OV@EORY. This is information about the previous 6 items.)

Here, the dictionary index storing the words in Table 2 is shown in FIG. 3, and its data memory is shown in FIG. 4.

In addition, the hyphenate information constitutes the single t.
It is also possible to compose by memorizing the letters 1H16 with a hyphen, but in that case the word r DEL I C
ATE J, word r DEL I V'ERY J, the detection efficiency is low when the position of the hyphen is different even though the same characters are lined up halfway.
We adopted a processing method that first grasps the entire word and then detects hyphen information.

Inserting hyphen information between the letters of a word would impede the quickness of the spelling of the word, so the hyphen information was stored all at the end of the word.

The reason why the word spelling check is performed in byte units even though one character is converted into 5 hits is that the processing speed is faster than checking in units of hits. For this purpose, an empty area was provided in the single J data.

<Effects> As described above, according to the electronic word dictionary of the present invention, the data is considerably compressed because the delimiter information (=J is given only selected words, not all recorded words. .

[Brief explanation of drawings]

Figure f31 is a block diagram of the electronic dictionary of the present invention, Figure 2 is a block diagram of a word search system having the electronic dictionary of the present invention, Figure 3 is a diagram of the dictionary configuration of a specific word, and Figure 4 is a block diagram of the word search system having the electronic dictionary of the present invention. is a memory configuration diagram of the dictionary. Agent Patent attorney Aihiko Fukushi (and 2 others) Figure 1

Claims (1)

[Scope of Claims] An electronic word dictionary characterized in that 1° word information and hyphen positions of words only for words consisting of a specific number of characters or more are registered. 2. The electronic group dictionary according to claim 1, wherein the specific number is six.