JPH0581241A - Homonym processing system - Google Patents

Abstract

PURPOSE:To improve the operability of 'KANA' (Japanese syllabary)/'KANJI' (Chinese character) conversion and to simplify the maintenance of an example dictionary (a semantics combining example dictionary is adopted instead of a conventional word combining example dictionary). CONSTITUTION:When plural homonyms exist as changing candidates in 'KANA'/'KANJI' conversion by a conversion part 3, a homonym restricting means 5 restricts the most optimum word. A 'KANA'/'KANJI' conversion dictionary 4 stores the 'KANJI' information, speech part information and semantic information of words for 'KANA' character strings. A semantics combining example dictionary 7 stores combining examples of plural semantic information stored in the dictionary 4. A semantics combination processing means 6 restricts the most optimum word in the processing of the means 5 based upon the contents of the dictionaries 4, 7.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a kana-kanji conversion system in a Japanese word processor or the like, and particularly when a plurality of homophone meaning words exist as conversion candidates for kana-kanji conversion, the most appropriate word (conversion candidate) is limited. Regarding the meaning word processing method. It should be noted that the expression "most appropriate" is used to mean "considered to be the most appropriate" regardless of whether or not it is actually "most appropriate".

[0002]

2. Description of the Related Art Conventionally, as a homophone meaning word processing method of this kind, there has been a method using an example of a combination of words (one-to-one combination of words). In this method, a word combination example dictionary that stores examples of combinations of words is provided. Then, when there are a plurality of homonymous meaning words as conversion candidates for kana-kanji conversion, the word combination example dictionary is searched (one word in the combination of words is used as a key to search the other word), and Appropriate words (conversion candidates) are limited.

Specifically, in the conventional homophone meaning word processing method, the following processing is performed.

For example, a kana-kanji conversion dictionary in which a plurality of homophone meaning words such as "thick", "hot", and "serious" are registered for a kana character string such as "atsushi" is provided.

[0005] For example, one of "hot" and "black tea"
Example of one-to-one combination and 1 for "thick" and "book"
A word combination example dictionary in which examples of paired combinations are registered is provided.

[0006] In the kana-kanji conversion using the dictionary as described above, for "atatsu" of "atitsukoucha",
Based on the word combination example dictionary, "hot" is limited from conversion candidates such as "thick", "hot", and "serious". Detected). In addition, for “hot” of “hot”, “thick” is limited from conversion candidates such as “thick”, “hot”, and “serious” based on the word combination example dictionary ( Using the "book" as a key, "thick" which is the combination partner is detected).

As a combination partner of "hot", there are various words such as "black tea", "udon", "forehead" and "iron". Therefore, in the conventional homophone meaning word processing method, corresponding to all of them, a one-to-one combination of "hot" and "udon", a one-to-one combination of "hot" and "forehead", and " It was necessary to register each example such as a one-to-one combination of "hot" and "iron" in the word combination example dictionary.

[0008]

In the conventional homophone meaning word processing method described above, a word combination example dictionary that stores examples of combinations of words is used, and a plurality of homophone meaning words are used as conversion candidates for kana-kanji conversion. Since the conversion candidates are limited in the case where there exists, there are the following drawbacks. Since the number of examples of one-to-one combinations of words is large, the capacity of the word combination example dictionary is large. When a word is added to the Kana-Kanji conversion dictionary, it is necessary to list the examples related to that word and register new information in the word combination example dictionary, which makes maintenance of the word combination example dictionary complicated and makes it difficult to convert the Kana-Kanji conversion dictionary. The increase in the capacity of the word combination example dictionary due to the addition of one word in the dictionary becomes large.

In view of the above points, an object of the present invention is to efficiently limit the conversion candidates when there are a plurality of homophone meaning words as conversion candidates for the kana-kanji conversion, and to improve the operability in the kana-kanji conversion. It is possible to improve and improve the forward conversion rate, simplify the maintenance of the example dictionary (the meaning combination example dictionary is adopted as the example dictionary of the present invention in place of the conventional word combination example dictionary), and expand the example dictionary. The object of the present invention is to provide a homophone meaning word processing method capable of achieving improvement in sex.

[0010]

The homophone meaning word processing method of the present invention limits the most appropriate word when a plurality of homophone meaning words are present as change candidates in the kana-kanji conversion by the converter. In a kana-kanji conversion system having means, a kana-kanji conversion dictionary that stores kanji information, part-of-speech information, and meaning information of words for kana character strings, and a semantic combination that stores an example of a combination of meaning information in the kana-kanji conversion dictionary It has an example dictionary and a meaning combination processing means for limiting the most appropriate words based on the kana-kanji conversion dictionary and the meaning combination example dictionary in the processing of the homonym meaning word limiting means.

[0011]

In the homophone meaning word processing method of the present invention, the kana-kanji conversion dictionary stores kanji information, part-of-speech information, and semantic information of words for kana character strings, and the semantic combination example dictionary stores meaning information in the kana-kanji conversion dictionary. The meaning combination processing means limits the most appropriate word based on the kana-kanji conversion dictionary and the meaning combination example dictionary in the processing of the homophone meaning word limiting means.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the homonym meaning word processing system of the present invention. The homophone meaning word processing method of the present embodiment uses an input unit 1 for inputting a sentence (a set of kana character strings) using a keyboard, a sentence storage unit 2 for storing the input sentence, and an input kana character. A conversion unit 3 for converting a string into a word composed of a kana-kana mixed character string using a kana-kanji conversion dictionary 4, and kanji information, part-of-speech information, and semantic information ("shape", " Kana-Kanji conversion dictionary 4 storing information indicating semantic attributes such as “temperature”, “state”, “liquid”, “solid”, and “extraction concept”, and a combination of semantic information of words (connection relationship) An example dictionary 7 for meaning combinations in which the meaning information in the orika kana-kanji conversion dictionary 4 is searched as a key, an input kana character string, conversion candidates for kana-kanji conversion, conversion results of kana-kanji conversion, etc. are stored. screen A screen display unit 8 for displaying, or of each part and the control unit 9 (the following description the control unit 9 for controlling the operation of each unit
The processing is not mentioned) and is included. Since the input unit 1, the text storage unit 2 and the screen display unit 8 are realized by a known technique in a Japanese word processor or the like, detailed description will be omitted in the following description.

When there are a plurality of homonymous meaning words as the conversion candidates obtained by using the Kana-Kanji conversion dictionary 4, the converting unit 3 uses the grammatical information or the like to limit the most appropriate conversion candidates. The limiting means 5 has one function.

The homophone meaning word limiting means 5 has, as one function, a meaning combination processing means 6 for limiting conversion candidates by searching the meaning combination example dictionary 7 using the meaning information in the kana-kanji conversion dictionary 4 as a key. is doing.

2 and 3 are diagrams for explaining the specific operation of the homonym meaning word processing method of the present embodiment.

Next, the operation of the homonym significant word processing system of the present embodiment thus constructed will be described. The operation will be described with reference to the specific examples shown in FIGS. 2 and 3.

FIG. 2 is a diagram showing a correspondence between a part of the contents of the Kana-Kanji conversion dictionary 4 and a part of the contents of the meaning combination example dictionary 7. The expression "thick (adjective / shape)" in the kana-kanji conversion dictionary 4 indicates a combination of "kanji information (part-of-speech information / semantic information)".

When the Kana-Kanji conversion is performed, the input unit 1
Inputs a sentence that is a set of kana character strings (kana headings) (the sentence storage unit 2 stores the sentence). Here, as one specific example, it is assumed that the input unit 1 has input the sentence "Atsukokocha".

The conversion unit 3 sequentially searches the Kana-Kanji conversion dictionary 4 for each kana character string in the sentence input by the input unit 1 and stored in the sentence storage means 2, and "thick" for the kana character string "atsu". Detects conversion candidates such as "(adjective / shape)", "hot (adjective / temperature)", and "serious (adjective / state)", and replaces the kana character string "kocha" with "tea (noun / liquid)" Is detected.

As described above, since there are a plurality of homonym meaning words for "hot", the conversion unit 3 activates the homonym meaning word limiting means 5, and the homophone meaning word limiting means 5 further means the meaning combination processing means 6. To start.

The meaning combination processing means 6 performs the meaning combination based on the meaning information ("shape", "temperature", "state", "liquid", etc.) attached to the conversion candidates detected as described above. The example dictionary 7 (as shown in FIG. 2), examples of combinations of meaning information such as “liquid-temperature”, “solid-shape”, and “abstract concept-state” are registered in the meaning-combination example dictionary 7. Search). That is, the meaning combination processing means 6 detects (applies) the combination of "solid-shape" to the conversion candidate "thick (adjective / shape)" using the meaning information "shape" as a key, and "hot". For the conversion candidate “(adjective / temperature)”, the combination of “liquid-temperature” is detected using the semantic information “temperature” as a key, and for the conversion candidate “serious (adjective / state)”, the “state” is detected. A combination of "abstract concept-state" is detected using the semantic information "" as a key. Further, the meaning combination processing means 6 detects a combination of "liquid-temperature" for the conversion candidate "black tea (noun / liquid)" using the meaning information "liquid" as a key. As a result, the meaning combination processing means 6
For “Atsuko” of “Atsukocha”, “hot” is selected (limited) from a plurality of homophone meaning words (see FIG. 3, which shows the situation where homophone meaning words are limited by an example of a combination of meaning information). That).

The conversion means 3 receives the information regarding the above-mentioned limitation through the homonym meaning term limiting means 5 and outputs "hot black tea" as the conversion result to the screen display means 8 and the like.

In the process of the above processing, the screen display means 8
The input kana character string, conversion candidates, conversion results, etc. are displayed on the screen.

Next, as another specific example, let us consider a case where the input section 1 inputs a sentence "Atitsuhon".

In this case, the conversion unit 3 selects "thick (adjective / shape)", "hot (adjective / temperature)" and "serious (adjective / state)" for the kana character string "hot". The conversion candidate is detected, and the conversion candidate "book (noun / solid)" is detected for the kana character string "hon".

As in the previous specific example, since there are a plurality of homonym meaning words for "hot", the converting section 3 activates the homonym meaning word limiting means 5, and the homophone meaning word limiting means 5 further means. The combination processing means 6 is activated.

The meaning combination processing means 6 searches the meaning combination example dictionary 7 based on the meaning information attached to the conversion candidates detected as described above. That is, the meaning combination processing means 6 is similar to the above specific example for each conversion candidate of “thick (adjective / shape)”, “hot (adjective / temperature)” and “serious (adjective / state)”. Then, the combination of "solid-shape", the combination of "liquid-temperature" and the combination of "abstract concept-state" are detected. Further, the meaning combination processing means 6 uses the meaning information "solid matter" as a key for the conversion candidate "book (noun / solid matter)" as "solid matter-".
The combination of "shapes" is detected. As a result, the meaning combination processing means 6 selects "thick" from a plurality of homonymous meaning words for "atatsu" of "atitsuhon" (limited).
(See FIG. 3).

The converting means 3 receives the information regarding the above-mentioned limitation through the homonym meaning limiting means 5, and outputs the "thick book" of the conversion result to the screen display means 8 and the like.

Further, as a specific example other than the above, the conversion result of "severe virtue" is output with respect to the input of the sentence "toku ga tsu" (see FIGS. 2 and 3).

[0031]

As described above, according to the present invention, by providing the kana-kanji conversion dictionary storing meaning information in addition to kanji information and part-of-speech information, a meaning combination example dictionary, and meaning combination processing means, kana-kanji conversion is performed. When there are multiple homophones as conversion candidates, the conversion candidates can be appropriately limited efficiently, and the operability and the correct conversion rate in kana-kanji conversion can be improved. In the present invention, it is possible to simplify the maintenance of the meaning combination example dictionary (no need to add an example of a combination of words when adding words to the Kana-Kanji conversion dictionary) and improve the extensibility of the example dictionary. There is an effect that you can.

[Brief description of drawings]

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

FIG. 2 is a diagram showing a part of the contents of a kana-kanji conversion dictionary and a meaning combination example dictionary in FIG. 1 and their correspondence.

3 is a diagram for explaining a specific operation of the homonym meaning word processing method shown in FIG. 1 (a diagram showing a situation where homonym meaning words are limited by an example of a combination of meaning information).

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 input section 2 text storage means 3 conversion section 4 kana-kanji conversion dictionary 5 homophone meaning word limiting means 6 meaning combination processing means 7 meaning combination example dictionary 8 screen display means 9 control section

Claims (1)

[Claims] 1. A kana-kanji conversion system having a homophone meaning word limiting means for limiting the most appropriate word when a plurality of homophone meaning words are present as change candidates in the kana-kanji conversion by the converting portion, for a kana character string. A kana-kanji conversion dictionary that stores kanji information, part-of-speech information, and meaning information of a word, a meaning-combination example dictionary that stores an example of a combination of meaning information in the kana-kanji conversion dictionary, and a process of the homonym meaning word limiting means, A homophone meaning word processing method, comprising: a meaning combination processing means for limiting the most appropriate words based on the kana-kanji conversion dictionary and the meaning combination example dictionary.