JPS6274153A - Retrieving method for electronic dictionary - Google Patents

Abstract

PURPOSE:To shorten the time required for checking a spelling, by adjusting the number of words per one group by increasing or decreasing the number of digits of the index of a hash value, when grouping a word group and registering it to a dictionary. CONSTITUTION:Character and word information inputted and edited by an input device 1 is stored in a memory device 2. An electronic dictionary 4 for a spelling check which is connected to the memory device 2 is provided with an operation processing part for a spelling check processing, and offers information related to whether the spelling of an inputted word is correct or not, in accordance with an inquiry from the memory device 2. This operation processing part which is used exclusively for the spelling check processing is constituted so that the number of words per one group can be adjusted by increasing or decreasing the number of digits of the index of the hash value, when grouping the words and registering them in the dictionary 4.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application! J!P) The present invention relates to a search method for electronic dictionaries attached to various language processing devices including word processors, typewriters, and the like.

(Technical background) In electronic word dictionaries attached to Roman word processors, Roman typewriters, etc., a spell check function is important, and there is a demand for spell check processing such as parentheses to be performed as quickly as possible. Therefore, when performing a spell check, for example, as shown in Table 1, the group of eulogy words to be registered in the dictionary has an initial letter and a number of characters of 7, - therefore 2.
Mars-like 1. It is conceivable that the corresponding ororj8 words are searched only within the group 4, and the number of words in each group iJ shown in Table 1. 3. (Problem to be solved by the invention) However, in the search method described in 1.1L, although the Qj word group registered in electronic meter -) is used as a group number, 2) , because the number of 91 words that belong to 7 individual groups is large, and there are many words (for example, a word with the initial letter "C" and 8 letters (7th letter 154.17i), it is necessary to search for the corresponding word. -4゛rol!j There is a problem that the interval becomes long 9,
Therefore, 1. L real quino,
Processing (, 2 is difficult to provide.

Father In the above search power method, each word has 1 out of 1 letters! It has been converted into a code 1 and the electronic meter 11 has been registered in Japan. =, the storage capacity required to store one word varies depending on the number of characters, and especially in the case of 111 words, which have a large number of characters, there is a problem that the storage capacity per word becomes large. Moreover, for 11 words with a large number of characters, the search time per word would be quite long3.In addition, according to the above search method, in the case of a wrongly spelled word, After matching all words in the group,
jtl, it's a misspelled word and the judgment is 4 or so, so
The purpose of the present invention is to solve the various inconveniences mentioned in 1. Therefore, the electronic dictionary search method according to the present invention stores a plurality of words in the electronic dictionary I) consisting of a memory device and performs a search. , r
For each alphabetic character, write the letters U-r, ,,1'
When l- is defined, each position r) in the Qi word is set to t), ,
(Determine +- and add 1 to the first letter of each Qi word.) Then, multiply the character (,') j1' by the position weight, divide the value by a predetermined prime number, find the remainder, and continue to Add the remainder of the curvature of the character wa, it and the position ugo, the power of it'c>value 11, and (divide it by the predetermined prime number 7:') to find the remainder. The process is repeated for each character fii up to the letter ``hi'', and the final remainder for the final character is divided into the lower constant number of inte tx and the lower constant number of digits to calculate the value of each word. Register the data in a group hj with a common index in the electronic dictionary, obtain the above-mentioned value for the word to be searched, and combine the data of the search word and f=
The present invention is characterized in that the index of the search word in the electronic dictionary is checked for match or mismatch with the data group of the group corresponding to the index.

In this case, it is preferable to register the data groups belonging to each groove number in ascending order of numerical value.

(Example) Please refer to the attached drawings and charts below.1. The present invention will be explained based on examples.

The drawings show the general structure of a processing device and a word δP to which the method of the present invention can be applied. In the same figure, ■ is an input device for inputting characters and word information to the language processing device, and specifically, for example, a keyboard device, a tablet device, an OCI device, etc.
z (optical character reader), magnetic tape device, etc. are used.

Reference numeral 2 denotes a storage device which is connected to the human-powered device 1 and stores the character and word information edited by the human-powered device 1, such as a core memory, an IC memory, a magnetic disk device, etc. 9 , 3 are connected to the storage device 2,
It is an output device for outputting information stored in the storage device 2.
For example, various printers, display devices, magnetic tape devices, magnetic disk devices, etc. are used. 4.4 is an electronic dictionary for spelling tongue ivy connected to the storage device 2.
), for example, it is composed of: l memory, IC memory, ROM (rung 1, access memory), magnetic disk device, etc. As will be described later, the dictionary 4 is equipped with an arithmetic processing section dedicated to spell check processing,
1, it is possible to provide information on whether the spelling of the word that has been deduced is correct or not.

Reference numeral 5 denotes a control device connected to each device 1 to 4, which is configured by a combi coater, for example, and is connected to each device 1 to 4. Controls the transmission and reception of signals at this station.

Next, encoding of European words using the hashco method in the present invention will be described.

When encoding this, first, as shown in Table 3, which shows part of the human alphabet, each character has two characters.
A unique character way consisting of a base number) (conveniently expressed in decimal notation) is determined. Although omitted in Table 3,
Character weights are similarly determined for lowercase alphabets, numbers, and the like.

At the same time, as shown in Table 4, for each character position within a word, a unique position weight (
(expressed in hexadecimal).

Table 4 shows position weights for the first to seventh characters, but similar position weights are determined for the eighth and subsequent characters.

Next, based on the above character weight and position weight,
Calculate the hash value of each eulogy using the following procedure.

(i) That is, first, the character weight and position weight of the first character (initial character) of each word are multiplied.

For example, in the case of the word "AIR", the character weight of "A" is "60" and the positional weight of the first character is "ooogoo".
oo". In that case, it is preferable to divide the character weight into 3-bit units and multiply by the position weight. Then, divide the multiplied value by the prime number closest to 227 to obtain the remainder. Remember the remainder.

(11) Next, add the remainder obtained for the first character to the value obtained by multiplying the character weight and position weight of the second character, and divide that value by a prime number to calculate a new remainder. Thereafter, the same operation is repeated for each character up to the last character (the end of the word), and the finally obtained remainder is used as the hash value of the word. Here, since the prime number closest to 227 is selected as the divisor for each division, all the values are expressed as binary numbers within 27 digits.

Note that by changing the divisor, the number of digits of the hash value can be changed arbitrarily.

Table 5 shows the results of calculating the hash value using the above procedure for the beginning of the alphabet.

Table 6 shows the first part of the hash values obtained in this way, rearranged in ascending order (in ascending order). In Table 5.6, hash values are expressed in octal for convenience.

Once the encoding using the hash method is completed as described above, the hash value of each word is divided into the upper 11 digits (in the case of binary representation) of the index part (hereinafter simply referred to as index) and the lower 16 digits of the data part ( (hereinafter simply referred to as data), and grouped by words with common indexes. For example, among the word groups shown in Table 6, the 43 word groups with hash values less than 2+6 (13 in octal notation, 200,000)
r is classified into the first group with index "0". In addition, the Haranoko value is 21'1 or more and 21
Words with an index of less than 7 are classified into the second group with the index "bi". In this way, all words are classified into 2048 (-2') groups corresponding to the number of digits in the index. , the number of groups can be increased or decreased arbitrarily by changing the number of digits of the index.

Table 2 shows the number of words belonging to each group (expressed in hexadecimal for convenience) when the number of groups is 2048. The vertical axis outside the margin of Table 2 shows the first three digits of the group number (each digit is expressed in hexadecimal), and the horizontal axis outside the margin shows the lowest digit of the group number (in octal notation). ) is shown. As is clear from the table, by increasing the number of groups, the number of words belonging to each group decreases.1. ing.

By the way, as shown underlined in Table 2, the maximum number of words in the I group in nature is 88 (+ 6
58) in the display, therefore, all spell checks can be performed with a maximum of 88 searches.

The word data grouped as described above is stored in the dictionary 4 in ascending order for each group. Furthermore, the start address of each group is determined and stored based on the number of words in each group shown in Table 2 (see Table 7). These addresses are used to select the relevant group during spell checking.

The spell check process using the dictionary 4 will be described below.

When a word to be spell-checked (hereinafter referred to as a search word) is sent from the storage device 2 to the dictionary 4, an arithmetic processing unit (not shown) in the dictionary 4 performs a search 11" for the i word using the same IFW method as described above. 4. Continuing, the Sonko value is calculated.
During the search, 5t1's fonko/I squirrel found the corresponding loop, and then the above searched word 111 belonged to the corresponding group. Data 1! Match or mismatch with 'T is checked sequentially.3. As a result of the matching, the pig in the search 11'i word matches the data in the corresponding group's 1. A1. On the other hand, if the data of the search word ljj does not match any of the data of the corresponding group, write a message to the storage device 2 that it is a misspelling.
, 3 is sent, 1, 3, illegally (J each!? Lube one?
-(=,
In particular, in the case of a misspelling, the data of the search word is smaller than any of the data in the corresponding 1゛loop (, i) - +' riji (J)',), and I don't agree with the previous data. In this case, Nnonino-3 (the same M-edict) occurred, and 3,5-goteno-no-2.

``No'' means 2 or more words] ``Hason'' in the first word of -: +, (iI
9. While I was thinking,
Hayo ") na Jino. Nino is dictionary 11, - storage 4
- Roko, 1ha (j) 32 out of 72,000 words, ;7r
())♂) -Q I's i), erroneous recognition is 1 to 2250 words. I ;+! -8 and F tome-(-rarely 1-5
ka/l l-: Because there is no, :ljj Itl,,i-
There is no obstacle to ','(r)J.

(Effect of the invention) As explained below, it is clear from 31, ') that according to the present invention, A1, jli word group), and own I-b (1 (7)). 11 (σ record Wro (per J, f of the above J\tsunono value
Increasing or decreasing the number (t) of the number 31-1 to calculate the number 1.7i', converting it to -1-4, then the number 7 is 7°. ) 1i irises Il, '=1 Adjust the number -4-ro, 2 and basket, \ro4, spot, rnde, 2nd layer Il, set the number of digits large enough 1-\ku 4 By the way, if the initial letter and number of letters are 3L and 9 loops (), then i
), increase the number of horn loops, then 1 L'7" l) (during the middle, decrease the number of ia, 1 east loop and Δro.j
L-i, Spellch”! Shorten the time required for ``-4゛゛゛゛ word or something,゛゛゛゛゛゛゛゛゚゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛゛)゛゛゛゛゛stomach]
As long as the total number of ljj words in the present invention is constant,
Number of loops, -9゛1su'no゛runo゛I door)
I) It is not possible to adjust the number of words.

Previously, each of the 111 words was registered in the dictionary by changing it to '-1-' with one letter in the middle.
, In the present invention (J badge, 1 method is adopted, 31-0 is registered as 211 in the middle of Chinese) (with 0 per word)
Make the storage capacity sufficiently small. , it also reduces the dictionary appearance. .

Furthermore, the data + ty belonging to the old rubbish are arranged 1-1 in the order of A'I respectively (j, if by mistake, if a spelled word is entered, it will be manually written). Compare the data of the word with all the data of the corresponding group in the dictionary. At this point, judge the misspelling by I;
Because of this, every effort to shorten the processing time is 7' arbitrary. .

(1・゛margin) Table 1 initials Chiku 9 Ii Nko Otsuji □−1−−〇− Table 3 Table 4 Table 7 Table 5 Table 6 Table 8

[Brief explanation of drawings]

The drawing is a block diagram showing the general configuration of a language processing device to which the method of the present invention can be applied. 4...Dictionary.

Claims (2)

[Claims] (1) When storing multiple words in an electronic dictionary consisting of a memory device and performing a search, a character weight is determined in advance for each alphabetic character, a position weight is determined for each position within the word, and the initial character of each word is determined. Divide the value obtained by multiplying the character weight and position weight by a predetermined prime number to find the remainder,
For each subsequent character, the value obtained by adding the previous remainder to the multiplication value of the character weight and position weight is divided by the above predetermined prime number to obtain the remainder, and the operation is repeated for each character up to the last character of the word. The hash value obtained as the final remainder is divided into an index with a certain number of upper digits and data with a certain number of lower digits, and the data for each word is registered in the electronic dictionary for each group with a common index, and then searched. The electronic device is characterized in that the hash value is calculated for the word to be searched, and the match or mismatch between the data of the search word and the data group of the group corresponding to the index of the search word in the electronic dictionary is checked. How to search a dictionary. (2) The method according to claim 1, in which data groups belonging to each group are registered in order of decreasing numerical value.