JPH036657A - Word dictionary registering device - Google Patents

Abstract

PURPOSE:To attain the batch conversion of the stem and the inflectional ending of a word for each paragraph by registering the proper types of parts of speech to the words registered by users. CONSTITUTION:An input control part 11 inputs the KANJI (Chinese character) description and the KANA (Japanese syllabary) reading of a word to be registered via a character input device 2. A dictionary retrieving part 12 reads the corresponding KANA reading and type of the part of speech out of a word dictionary 3 for the KANJI description shown by a part-of-speech analyzing part 13. The part 13 obtains the stem and the type of the part of speed for the KANJI description and the KANA reading to be registered based on the KANJI description and the KANA reading of the input word and the information obtained via the part 12. A part-of-speech register part 14 registers the KANJI description, the KANA reading, and the type of the part of speech of the produced word into the word dictionary 3. Thus it is possible to convert en bloc the stems and the inflectional endings of those words having the proper parts of speech and given previously for each paragraph.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention is a word dictionary registration device that registers words in a word dictionary used in a kana-kanji conversion device that converts kana readings of words into kanji. Regarding.

[Conventional technology]

Conventionally, when registering a word in a word dictionary, the word dictionary registration device receives the kanji transcription and kana reading of the word from the user, and then receives the original kanji transcription, kana reading, and the word's original reading. Fixedly assigning a noun as a part of speech typeK
I'm doing more.

[Problem to be solved by the invention]

In the conventional word dictionary registration device described above, nouns are registered as the part-of-speech type of a word, so when an attempt is made to register a word with a part-of-speech type other than a noun, such as a verb or adjective, the word's original part-of-speech type is not registered. When a word that is not registered and is converted using a kana-kanji conversion device is converted into a word that cannot be treated as a verb or adjective because a noun is registered as the part of speech type, the word with the appropriate part of speech provided in advance It has the disadvantage that the stem and conjugated ending of a word cannot be converted in units of clauses, as in the case of ``-''.

[Means to solve the problem]

The word dictionary registration device according to the present invention includes an input control unit for inputting the kanji notation and kana pronunciation of the word to be registered, a dictionary search unit for reading word information from the word dictionary, and a word dictionary registration device for inputting words using the input control unit. A part-of-speech analysis section that analyzes the part of speech of a word based on information obtained using the kanji notation, kana reading, and dictionary search section, and a part-of-speech type, kanji notation, and kana reading of the word determined by the part-of-speech analysis section. It has a dictionary registration section for registering words in a dictionary.

〔Example〕

Next, one embodiment of the present invention will be explained in detail with reference to the drawings.

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

The word dictionary registration device 1 of this embodiment includes an input control section 11,
Dictionary search unit 12, part of speech analysis unit 13, and dictionary registration unit 14
A character input device 2 and a word dictionary 3 are connected.

The input control unit 11 uses the character input device 2 to input the kanji notation and kana reading of the word to be registered.

The dictionary search unit 12 reads the kana reading and part of speech type corresponding to the kanji notation indicated by the part of speech analysis unit 13 from the word dictionary 3. The part-of-speech analysis unit 13 uses the input control unit 11 to determine the kanji notation and kana reading of the input word, and the dictionary search unit 1
2. Determine the stem and part of speech of the kanji notation and kana reading to be registered from the information obtained using 2. The dictionary registration unit 14 registers the kanji notation, kana reading, and part of speech type of the word created by the part of speech analysis unit 13 in the word dictionary 3.

FIG. 2 is a flowchart showing the operation manual page of the word dictionary registration device 1.

The operation of the word dictionary registration device 1 will be explained using FIG.

First, the input control unit 11 uses the character input device 2 to input the kanji notation and kana reading of the word to be registered (STE
P2-01). Next, the part-of-speech analysis unit 13 analyzes the part-of-speech type based on the input kanji notation and kana reading (STEP 2-
11-5TEP2-42).

The part-of-speech analysis unit 13 first checks whether the 4L word is a noun or not, and whether or not the kanji notation includes okigana (STEP 2-11). If the kanji notation does not include okukana, the part of speech type should be determined as a noun. 5TEP2-
21), if the kanji notation includes okurikana, the part of speech type is determined to be other than noun.

For example, if the kanji notation is "application" and the kana reading is "shutsugan", the last part of the kanji notation is not a kana, so it is determined that there is no okigana, and the part of speech type is set to noun. Therefore, the kanji notation of the word to be registered is "application," the kana reading is "shiyutsugan," and the part of speech is noun. Furthermore, if the kanji notation is ``read'', the ending is the kana ``mu'', so it is determined that there is an okrigana, and the part of speech type is considered to be other than a noun, and the analysis proceeds.

If the word to be registered is a compound word whose part-of-speech type is determined to be other than a noun, it is determined by referring to the part-of-speech type of the word already held in the word dictionary 3. If the word to be registered is not a compound word, the part of speech type is determined from the word ending included in the okurikana.

Check whether the word you are trying to register is a compound word that has a word already included in the word dictionary 3 at the end of the word, that is, a word whose kanji notation and kana reading match the ending of the kanji notation and kana reading you are trying to register. is searched from the word dictionary 3 using the dictionary search unit 12, and if it can be searched, it is a compound word,
If it cannot be searched, it is determined that it is not a compound word (STE
P2-12).

If it is a compound word, the part of speech type is registered (STEP 2-31) to register the part of speech type of the searched word, and the conjugated ending is determined based on the kanji notation and kana reading of the searched word. Remove and leave only the stem (STE)
P2-32).

If it is not a compound word, extract those included in the kanji-written okurikana from common conjugated endings that connect to each pre-memorized part of speech type, and register the part of speech type that the extracted conjugated endings connect to. 5TEP2-41)
, remove the conjugated endings from the kanji notation and kana reading to be registered based on the extracted conjugated endings, leaving only the stem (S
TEP2-42).

For example, the kanji notation [read] of the word you are trying to register,
If the kana reading is ``yomikomu'', the word dictionary 3 shows ``kumu'' which corresponds to the end of the word (the dictionary contains the root of ``kumu'', the kanji table 1 is ``kumi'', and the kana reading is ``ko''). , the part-of-speech type is registered as a verb. "Kumu" is a verb, so "Kumu"
If you are able to search for ``read'' (which matches the ending of ``read''), then ``read'' is considered a compound word, and the kanji notation of the word to be registered is ``read'', the kana reading is ``yomiko'', and the part of speech is The type is verb.

Finally, the kanji notation, the stem of the kana reading, and the part of speech type obtained by the part-of-speech analysis unit 13 are registered into the word dictionary 3 using the dictionary registration unit 14 (STEP 2-02).

〔Effect of the invention〕

As explained above, the present invention enables the user to register appropriate part-of-speech types for words registered by the user.
When registered words are converted using a kana-kanji conversion device, the stems and conjugated endings of the words can be converted at once for each clause, similar to the words with the appropriate part of speech type provided in advance, increasing the efficiency of conversion. It has the effect of improving

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, and FIG. 2 is a flowchart showing the operating procedure of the word dictionary registration device 1. As shown in FIG. 1...Word dictionary registration device, 2...Character input device, 3...Word dictionary, 11...-...
Input control unit, 12... Dictionary search unit, 13...
...Part of speech analysis department, 14...Dictionary registration department.

Claims (1)

[Claims] an input control unit that inputs the kanji notation and kana reading of the word to be registered in the word dictionary; a dictionary search unit that reads word information from the word dictionary; the kanji notation and kana reading of the input word; a part-of-speech analysis unit that analyzes the part-of-speech type of a word based on information about the word, and a dictionary registration unit that registers the part-of-speech type, kanji notation, and kana reading of the word determined by the part-of-speech analysis unit to the word dictionary. A word dictionary registration device that analyzes and assigns an appropriate part of speech type of a word to be registered from kanji notation and kana reading.