JPH06187328A - Japanese syllabary-to-chinese character converting device - Google Patents

Abstract

PURPOSE:To improve the KANA(Japanese syllabary)-KANJI(Chinese character) conversion efficiency of homonyms. CONSTITUTION:The KANA-KANJI converting device 3 recognizes the reading of KANA characters inputted from an input device 2, reads candidates for KANJI information corresponding to the reading out of a KANJI dictionary storage part 5, and displays them on a display device 4, when KANJI information corresponding to the reading is determined with a determination indication from the input device 2, the reading and correspondence relation are learnt and stored in a specific memory area in a learning dictionary storage part 6 and when the reading of KANA characters including plural homonyms is inputted from the input device 2, a priority candidate for KANJI information corresponding to the reading including even the reading of one following character of the reading to be converted is learnt and stored in the specific memory formed in the learning dictionary storage part 6 successively.

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 device, and more particularly to a kana-kanji conversion device for learning the priority of kanji information determined from a plurality of conversion candidates in kana-kanji conversion of homonyms and synonyms.

[0002]

2. Description of the Related Art Conventionally, in a kana-kanji conversion device installed in a word processor, an electronic notebook, etc., when kana-kanji conversion of a homonym object is performed and the conversion result is learned, conversion candidates are selected from a plurality of conversion candidates according to frequency of use. The learning is performed by a frequency learning method for determining the priority order for displaying or a method called a latest learning method for determining the priority order for displaying the conversion candidates according to the order of latest use.

[0003]

However, both the frequency learning method and the latest learning method in such a conventional kana-kanji conversion device store the priority information of kanji candidates corresponding to the reading of homonyms. Because it was only
When a kana character string followed by a plurality of homonyms is input, there were the following problems.

That is, as shown in FIG. 5 (a), there are four kanji candidates "Your company, reporter, train, and returning company" for the reading "Kisha", and in the conventional frequency learning method or the latest learning method. If the prioritization is performed as shown in FIG. 5 (b), as shown in FIG. 6 (a),
For example, when a kana character such as "kisha no kisha kisha shisha shisha" is input, as shown in FIG. 6B,
As the same Kanji candidate for all homonym opposition words, "your company" with the highest priority is selected, and the Kanji candidate has to be determined for each homonym opposition word, which causes a problem that conversion efficiency decreases. there were.

An object of the present invention is to improve the kana-kanji conversion efficiency of homonym opposition words in a kana-kanji conversion device.

[0006]

The means of the present invention are as follows.

The input means for inputting the reading of kana characters, the dictionary storage means for storing kanji information corresponding to the reading of kana characters input by the input means, and the reading of kana characters input by the input means The kana-kanji conversion means for recognizing and reading the corresponding kanji information from the dictionary storage means and executing kana-kanji conversion, and learning and storing the correspondence between the kana-kanji conversion determined by the kana-kanji conversion executed by the kana-kanji conversion means and the kanji information. At the same time, when there are a plurality of conversion candidates of kanji information to be confirmed during the kana-kanji conversion, the priority candidates of the corresponding kanji information including the reading to be converted and the next one character of the reading to be converted are sequentially selected. And a learning and storing means for learning and storing.

Further, in this case, as in the invention described in claim 2, the learning storage means has a memory area for storing the correspondence between the reading of the input kana character to be converted and the kanji information, and the conversion target. It is also possible to provide a memory area for storing a priority candidate of Kanji information corresponding to the reading of one kana character next to the reading.

[0009]

The operation of the means of the present invention is as follows.

According to the present invention, the kana-kanji conversion means recognizes the reading of kana characters input by the input means, reads the corresponding kanji information from the dictionary storage means, and executes kana-kanji conversion. By means
When the kana-kanji conversion means executes the kana-kanji conversion, the correspondence between the reading and the kanji information is learned and stored, and when the kana-kanji conversion has a plurality of kanji information conversion candidates, the kana-kanji conversion target is converted. Along with the reading, the priority candidates of the corresponding Kanji information including the reading of one character next to the reading of the conversion target are sequentially learned and stored in a predetermined memory area. Therefore, when a plurality of kana-characters of the same-sound dissent word are input, it is possible to learn and memorize the priority candidate of the kanji information corresponding to the reading of the next one character together with each reading, and improve the kana-kanji conversion efficiency in the same-sound opposite word. It can be greatly improved.

[0011]

EXAMPLES Examples will be described below with reference to FIGS.

1 to 4 are diagrams showing an embodiment of a text processing device to which the kana-kanji conversion device of the present invention is applied.

First, the structure will be described. FIG. 1 is a block configuration diagram of the text processing device 1. In this figure, the sentence processing device 1 is composed of an input device 2, a kana-kanji conversion device 3, a display device 4, a kanji dictionary storage unit 5 and a learning dictionary storage unit 6.

The input device 2 is composed of a keyboard in which kana character keys and an instruction key for instructing a candidate of kanji information corresponding to reading of input kana characters displayed on the display device 4 by the kana-kanji conversion device 3 are arranged. , And outputs a series of kana character string reading and confirmation instructions to the kana-kanji conversion device 3.

The kana-kanji conversion device 3 has a CPU (Central
Processing Unit) and a ROM (Read Only Memory), etc., and executes Kana-Kanji conversion processing based on the Kana-Kanji conversion processing program stored in the ROM, and recognizes the reading of Kana characters input from the input device 2 and The Kanji dictionary storage unit 5 provides candidates for Kanji information corresponding to the reading.
When the kanji information corresponding to the reading is confirmed by the confirmation instruction from the input device 2, the reading and the correspondence are formed in the learning dictionary storage unit 6 described later. It is learned and stored in a predetermined memory area.

Further, the kana-kanji conversion device 3 includes an input device 2
When a kana character reading including a plurality of homonyms is input from, a priority candidate of the corresponding kanji information including the reading of one character next to the reading to be converted is formed in the learning dictionary storage unit 6. Further, the learning data is sequentially stored in a predetermined memory area.

The display device 4 is a CRT (Cathode Ray Tub).
e) and the like, and displays readings of kana characters input from the kana-kanji conversion device 3 and kanji information candidates. The kanji dictionary storage unit 5 is composed of a ROM or the like and stores kanji information corresponding to reading of kana characters.

The learning dictionary storage unit 6 is a RAM (Random Acc
esss Memory) and the like, and forms a memory area for storing kanji information candidates corresponding to the kana character reading input from the kana-kanji conversion device 3. For homophones, A memory area for storing priority candidates for Kanji information including reading of one character is formed.

Next, the operation of this embodiment will be described.

The kana-kanji conversion processing executed by the kana-kanji conversion device 3 will be described with reference to the flowchart shown in FIG.

First, when the kana character string reading is input by the operator's key operation on the input device 2 (step S1), the kana-kanji conversion device 3 analyzes the reading character string and recognizes a phrase break or the like (step S1). S2), the reading of the conversion target for each clause is taken out, and the reading of one character next to the reading is taken out (steps S3, S4). Next, it is searched whether or not the conversion object and the priority candidate of the kanji information corresponding to the reading of the next one character are stored in the learning dictionary storage unit 6 (step S5), and the priority candidate of the corresponding kanji information is determined. If it is stored, the Kanji dictionary storage unit 5 is searched and the corresponding Kanji information candidate is read (step S
6).

On the other hand, when the priority candidate of the corresponding Kanji information is not stored, it is judged whether or not the reading character string is the final phrase (step S7). When the reading character string is not the final phrase, the process returns to the reading process for reading the conversion target for each phrase in step S3, and the learning dictionary and the kanji dictionary search process is repeatedly executed. When the reading phrase is the final phrase, the searched kanji information is obtained. Is output to the display device 4 and displayed (step S
8). Next, the input device 2 is used to confirm the kanji candidate for each phrase.
Performed according to the confirmation instruction input from (step S
9) Confirm whether the conversion result is "OK" (step S
10) If the conversion result is not "OK", the next candidate is searched and displayed (step S11), and the conversion result is "OK".
If it is, it is confirmed whether the conversion is completed up to the final phrase (step S12). If it is not completed, the Kanji conversion candidate confirmation process from step S9 is repeatedly executed. The contents stored in the learning dictionary storage unit 6 are updated according to the conversion result, and the present process is terminated (step S1).
3).

In the above flow chart, the process of changing the segment breaks and obtaining the next and previous candidates of the Kanji information are the parts that are not directly affected by the conversion process of this embodiment. And the description is omitted.

However, it is needless to say that the kana-kanji conversion device 3 of this embodiment can also have a function of changing the punctuation of a phrase and obtaining the next and previous candidates of kanji information.

A specific example of a reading character string learned by the above kana-kanji conversion processing will be described with reference to FIGS. 3 and 4.

FIG. 3 shows an example of the learning contents stored in the memory area formed in the learning dictionary storage unit 5. In FIG. 3, the reading column stores "Kissha", and In the reading column, "no, ga, de, su" is stored as the one-character reading following the phrase "kisha", and in the candidate priority column, the next reading "no, ga, Then, as a priority order of candidates for each, "1: your company, 2: reporter, ...", "1: reporter,
"2: Train, ...", "1: Train, 2: Your company, ...", "1:
Returned to the office ... "is stored.

Here, as shown in FIG. 4 (a), if the reading character string is input from the input device 2, for example, "Kissha no kisha kisha kisha kisha."
The reading "kisha" for each phrase and the one-character reading "no, ga, desu" following each phrase are identified, the learning result in FIG. 3 is searched, and conversion is performed for each phrase. Then, Fig. 4
As shown in (b), the conversion result "Your reporter returns to the train" can be immediately output.

Therefore, when a series of reading character strings including a plurality of homonyms are input, the reading of the next one character is learned together with the reading of each clause, so that the corresponding kanji information candidate has priority. It is possible to surely study the kanji candidates that correspond to the reading without giving priority to the learning result of reading each bunsetsu, and to greatly improve the kanji conversion efficiency of homonyms. You can

Further, since the memory area formed in the learning dictionary storage unit 6 only needs to secure an extra memory capacity for storing the next reading of one character, the memory capacity conventionally used must be increased. In addition, the kana-kanji conversion device of the present invention can improve the kana-kanji conversion efficiency of the kana-kanji conversion device 3 and can be applied to a portable document processing device such as an electronic notebook or a portable word processor that cannot significantly increase the memory capacity. Therefore, it is possible to improve the kana-kanji conversion efficiency of the portable document processing device.

[0030]

According to the present invention, when the input kana character reading is recognized, the corresponding kanji information is read from the dictionary and kana-kanji conversion is executed, the kana-kanji conversion confirms the reading and kanji information. In addition to learning and memorizing the correspondence relation of, when there are a plurality of conversion candidates of kanji information for the kana-kanji conversion, the reading to be converted and the reading of one character next to the reading of the conversion target are handled. Since the priority candidates of the kanji information are learned and stored in a predetermined memory area in sequence, when multiple kana characters of the same phonetic object are input, each kanji character of the kanji information that corresponds to the reading of the next kanji character is surely read. The priority candidates can be learned and stored, and the kana-kanji conversion efficiency in homonyms and opposition words can be greatly improved.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a document processing device to which a kana-kanji conversion device of the present invention is applied.

FIG. 2 is a flowchart of a kana-kanji conversion process executed by the kana-kanji conversion device of FIG.

FIG. 3 is a diagram showing an example of learning contents stored in a learning dictionary storage unit of FIG. 1.

FIG. 4 is a diagram showing an example of a phonetic character string converted into kana-kanji based on the learning content of FIG. 3;

FIG. 5 is a diagram showing an example of reading and kanji candidates output by the conventional kana-kanji conversion processing.

6 is a diagram showing an example of a reading character string that is subjected to kana-kanji conversion processing based on the kanji candidate of FIG.

[Explanation of symbols]

 1 Document processing device 2 Input device 3 Kana-Kanji conversion device 4 Display device 5 Kanji dictionary storage unit 6 Learning dictionary storage unit

Claims (2)

[Claims] 1. An input means for inputting a kana character reading, a dictionary storage means for storing kanji information corresponding to the kana character reading input by the input means, and a kana character input by the input means. A kana-kanji conversion means for recognizing a reading and reading the corresponding kanji information from the dictionary storage means to perform kana-kanji conversion, and a correspondence relationship between the reading and kanji information determined by the kana-kanji conversion executed by the kana-kanji conversion means is learned. When there are a plurality of conversion candidates of kanji information to be stored and confirmed during the kana-kanji conversion, the corresponding candidate for kanji information including the reading to be converted and the reading of the next one character of the reading to be converted. A kana-kanji conversion device, comprising: a learning storage means for sequentially learning and storing. 2. The learning storage means includes a memory area for storing the correspondence between the reading of the input kana character to be converted and the kanji information, and the reading of one kana character next to the reading to be converted. The kana-kanji conversion device according to claim 1, further comprising a memory area for storing priority candidates of corresponding kanji information.