JPH0130173B2 - - Google Patents

Description

[Detailed Description of the Invention] <Technical Field> The present invention relates to an improvement of a kana-kanji conversion device, and more particularly to a kana-kanji conversion device that includes a katakana word registration dictionary that registers and stores independent words written in katakana in an independent word dictionary. It is related to the device.

<Prior art> In general, in a clause-separated kana-kanji conversion device, it is possible to know in advance that clauses of independent words written in katakana do not need to be converted, so the dictionary that is referred to during the conversion process is There is no need to prepare independent words written in katakana.

On the other hand, a method (hereinafter referred to as the solid input method) that automatically converts and outputs a sentence containing kana and kanji by inputting the pronunciation including punctuation marks without requiring the input of additional information such as bunsetsu delimiter marks. In this kana-kanji conversion device, since a phrase is determined by checking with a dictionary, it is necessary to store and prepare independent words written in katakana in advance in the dictionary.

Independent words written in katakana include product names and abbreviations (e.g. "pocket computer", "pocket computer")
etc.) etc. are created one after another. In addition, foreign words are being incorporated into Japanese sentences one after another, and their notations are also diverse (for example, ``binary'', ``binaryi'', ``binary'', etc.). Therefore, from this point of view, it is almost impossible to register and memorize all independent words written in katakana.
Furthermore, if only frequently used independent words written in katakana were stored in a dictionary, unregistered words would appear quite frequently in the kana-kanji conversion process.

Therefore, in a kana-kanji conversion device using a solid input method, how to deal with unregistered independent words written in katakana is an important issue.

Conventionally, devices have been proposed that allow an operator to register unregistered words in advance, but in such devices, the katakana words required by the operator are registered before performing the kana-kanji conversion process. This has become a burden on the operator.

<Purpose> The present invention has been made in view of the above points, and in order to reduce the burden on the operator, a conversion device automatically registers words that are deemed to be unregistered katakana independent words. The purpose is to provide a conversion device.

In general, most independent words written in katakana are nouns, followed by verbs and adjectives, and rarely have other parts of speech. Therefore, according to an embodiment of the present invention to be described later, automatically registered katakana words are given a part of speech so that they can be treated as nouns, verbs, and adjectives (in addition, they can also be treated as single-level verbs). is also possible).

In this way, the fact that the operator does not have to specify the type of part of speech is one of the factors that makes it possible to automatically register independent words in katakana notation. This was realized by effectively utilizing the characteristics of words.

<Example> Hereinafter, the present invention will be described in detail with reference to the drawings.

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

In Figure 1, 1 is a control device including a microcomputer (CPU), 2 is an independent word dictionary, and 3 is a control device including a microcomputer (CPU).
is an independent word temporary registration dictionary (katakana word registration dictionary), 4
5 is a storage device for dictionaries and tables other than the independent word dictionary 2 and registered dictionary 3, 5 is a storage device for converted kana strings, 6 is a storage device for katakana strings, and 7 is a storage device for conversion results.

In the kana-kanji conversion device configured as described above, after inputting the kana character string to be converted into the converted kana string storage device 5, by starting the control device 1, the independent word dictionary 2, katakana word registration Dictionary 3
and other dictionaries/tables 4, kana-kanji conversion processing is performed, and the conversion results are stored in the conversion result storage device 7.

The kana character string stored in the storage device 5 for converted kana strings is composed of hiragana, katakana (including long sound symbols), and symbols, and as shown in FIG. part 51 and katakana flag storage part 52, and the katakana part of the kana character string set in the storage device 5 (identified by the flag set state of the storage part 52)
The control device 1 is configured to operate so that the character is not written in anything other than katakana after conversion. The independent word dictionary 2 also includes, for example, frequently used words written in katakana.

If katakana is included in the kana character string set in the storage device 5 for converted kana strings, the control device 1 inputs the kana character string set in the storage device 5 and converts it into the above-mentioned general format. Before performing the kana-kanji conversion process, the following preprocessing according to the present invention is performed.

1 From the kana character string set in the storage device 5,
The katakana strings separated by hiragana and symbols are taken out one by one and stored in the katakana string storage device 6.

2. Perform kana-kanji conversion processing using the katakana string input and stored in the katakana string storage device 6 as input. In this conversion process, it is set in advance so that the conversion result is not allowed to be in anything other than katakana notation, so this conversion process is performed between independent words in katakana notation recorded in the independent word dictionary 2,
A determination operation is executed to check whether or not the katakana string input to the storage device 6 can be processed using the independent words in katakana notation currently registered in the registered dictionary 3.

3 If the kana-kanji conversion process in 2) above fails, the katakana string stored in the storage device 6 is regarded as one word and registered in the registered dictionary.

4 For all katakana strings included in the kana character strings input to the storage device 5, perform the above 1) to 3).
Repeat this action.

The operations 1) to 4) above are executed under the control of the control device 1.

The above operation will be explained using an example in which the word "school" is recorded in the storage device 2, but the word "Pokecom" is not recorded.

Now, it is assumed that the following character strings ○I to ○C are sequentially set in the storage device 5 and converted.

○B “…This is how this Pokemon school works…” ○B “…PokeCon school is like this…” ○C “…PokeCon book is…” First, preprocessing for the converted kana string in ○B above. As shown in FIG. 2a, first the storage device 5 is
The katakana string within is copied to the storage device 6, and then a check of the katakana string is executed between the storage device 2 and the storage device 6 as shown in FIG. After the determination, the "Pocket Computer" is registered in the storage device 3 as shown in FIG. 2c. At this time, as shown in FIG.
is memorized.

If the conversion process is performed on the character string marked ◯ after such pre-processing, the conversion process will not occur at the "Pocket Computer" and the conversion will not be possible.

Next, in the preprocessing for the character string to be converted in ○○, since "Pokecom" has already been registered in the storage device 3, the conversion for "Pokecom School" is successful as shown in Figure 3; "School" is never registered in the storage device 3.
Therefore, in the conversion process for the character string to be converted, "Pokécom School" is not processed as a single word, and "school" matches "school" stored in the storage device 2, so the memory By adding compound word information or semantic information to the memory contents of the device 2, it is possible to use the homophones ``exercise'', ``public and private'', ``minister'', ``lattice'', etc. ” can be expected to be selected as the first candidate.

In addition, even if the character string to be converted in ○c is entered in hiragana instead of katakana due to an operator error, "Pocket Computer" is registered in the storage device 3, so it is not shown in Figure 4. As shown, "Pokekon" is converted to katakana.

Note that, instead of registering independent words written in katakana as preprocessing, temporary registration may be performed when conversion becomes impossible during the conversion process.

However, when converting a large number of character strings at once, a more reliable effect can be expected by sequentially checking short katakana strings in preprocessing.

For example, if the above-mentioned character strings ○I to ○C are converted at once and "Pokecom School" and "Pokecon" appear before "Pokecon", they will be registered during the conversion process. In this method, the following process will proceed:

That is, when the process progresses to "Pokécom School", the conversion process becomes temporarily impossible because "Pokécom" is not registered in the storage device 3. Here, as shown by line A in FIG. 5, "Pokécom School" is registered as one word in the storage device 3, and the next process is restarted. However, "Pokecom School"
is processed as one word as shown by the line B in FIG. 5, so the information on "school" stored in the storage device 2 becomes unavailable. Furthermore, as shown in FIG. 6, "Pokecon" is either erroneously analyzed by the words stored in the storage device 2, or becomes unconvertible and has to restart processing from the next punctuation mark.

Therefore, great effects can be expected by registering independent words written in katakana as preprocessing and sequentially checking short katakana strings as preprocessing.

Note that the contents registered in the storage device 3 may be configured to be erased by the operator, or may be configured to be automatically cleared when the conversion of a series of character strings (for example, sentences) is completed. .

Furthermore, in the case where a series of conversion results are to be transferred to an external file, the contents of the storage device 3 may be stored together.

Further, as shown in FIG. 7, the storage device 5 in FIG. 1 is used as a buffer for an input device 8 such as a keyboard, and the storage device 7 is used as a buffer for output devices 9 and 10 such as a display or printer for real-time processing. It goes without saying that it may be done in such a way that it can be carried out.

<Effects> As described above, according to the present invention, as a preprocessing of conversion, it is checked whether a katakana string in a kana string to be converted can be analyzed using an independent word dictionary, and if it cannot be analyzed,
Since the entire katakana string is automatically registered in the katakana word registration dictionary, it is possible to efficiently convert kana character strings that include katakana strings without increasing the burden on the operator. I can do it.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an apparatus according to an embodiment of the present invention, FIG. 2 is a state diagram of each storage device for explaining the preprocessing operation in the present invention, and FIGS. 3 to 6 are also in accordance with the present invention. FIG. 7 is a block diagram showing the configuration of a modified embodiment of the present invention. 1...Control device, 2...Independent word dictionary, 3...Independent word temporary registration dictionary (Katakana word registration dictionary), 5...
Storage device for converted kana strings, 6... Storage device for katakana strings, 7... Storage device for conversion results.

Claims (1)

[Scope of Claims] 1. In a kana-kanji conversion device having a katakana word registration dictionary that registers and stores independent words expressed in katakana as an independent word dictionary, as a preprocessing for conversion, a katakana string in a kana string to be converted is converted into an independent word dictionary. means for determining whether or not the content stored in the katakana word registration dictionary can be analyzed; A kana-kanji conversion device comprising: means for registering.