CN1006333B - Chinese input method - Google Patents

Abstract

The invention discloses a method for processing homophones when the keyed-in Chinese phonetic alphabet sequence is transformed into a Chinese character sequence. This method has a list of homophones. For several Chinese homophones, the word with the highest frequency of use is regarded as class A, the m words with moderate frequency of use are regarded as class B, and the word with lower frequency of use is regarded as class B. as class C. When a character is selected from n homophones, a character of class A is first displayed, and when the character is not the required character, m words of class B are retrieved according to a specific method, and when characters of class A and class B When none of the required characters are found, then a character of class C is retrieved according to the original determined order, thereby realizing high-speed Chinese input.

Description

The present invention relates to an input method of Chinese language and characters (hereinafter referred to as the Chinese input method), in particular to a method for processing homophones in the case of using phonetic alphabets to input Chinese.

In an information processing system that handles Chinese, a Chinese keyboard is essential, and as such a keyboard input method, a method of encoding the shape of Chinese characters, the reading method, or a combination of the two is generally used.

The phonetic alphabet representing Chinese pronunciation includes Pinyin formulated by the Chinese government and Zhuyin that was used before this formulation. Now China is dominated by pinyin, and the phonetic transcription is mainly used by the elderly and in some areas.

The reading method of Chinese has the advantage that one Chinese character corresponds to one syllable in principle, and there are no other reading methods except for some characters. However, there are many homophones with the same pronunciation, that is, the same syllable, especially in some of the frequently used characters written in Japanese kana, there are many single-character words composed of one character, and most of them have many homophones , and words with more than two characters also have homophones, but their number is very small.

Therefore, when inputting the above-mentioned homophones according to typing phonetic letters (such as pinyin), generally in operation, a desired word is selected from a plurality of homophones represented in sequence.

However, with this method, there is a disadvantage that the above-mentioned selection operation must be repeated every time the same character is repeatedly input. Therefore, in general, when learning and processing homophones, if the same method is used to process homophones with very different frequencies of use, this will instead become a reason for making the input complicated.

About the frequency of use of Chinese characters According to a recent survey, in general books and newspapers, about 6,000 Chinese characters are used, but the 60 characters with high frequency of use account for more than 30%, and the most used The number of occurrences of the word "的" is more than 4%. Conventional Chinese input methods have not adopted a method of learning homophones that takes these frequency of use into consideration.

The object of the present invention is to provide a kind of Chinese input method that classifies each homonym according to frequency of use and actual situation, and improves above-mentioned shortcoming.

The Chinese input method of the present invention, in the Chinese input method that the keyed-in phonetic letter sequence is transformed into the Chinese character sequence, has a homonym table, and it regards the n homonym characters of Chinese, uses the most frequently used word as In category A, m words with moderate frequency of use (m≤n-1) are classified as category B, and l words with low frequency of use (l≤n-m-1) are classified as category C. When a word is selected from the m words of the aforementioned B category, the first means of displaying sequentially with the word that has just been selected is arranged, and when a word is selected from the l words of the above-mentioned C category, it has been determined according to the previous The second means of displaying sequentially and sequentially. When a character is selected from the above n homophones, a character of the above-mentioned A category is initially displayed, and then, when the character of the A category does not match, m characters of the B category are retrieved according to the above-mentioned first means. When each word of above-mentioned A category and above-mentioned B category does not match, realize according to above-mentioned 1 word of above-mentioned C category retrieval of second means.

According to the Chinese input method of the present invention, a word with the highest frequency of use is initially displayed, and when it is not a required word, then from the words with moderate frequency of use, it is displayed sequentially from the word just selected, because this is done This operation is judged, so it is not necessary to repeat the same selection operation whenever the same word is input as in the past. That is to say, even if the first displayed character is not the required character, the degree of the required character must be very high for the character displayed after the second time. Due to the effect of this learning function, high-speed Chinese can be expected to greatly save complicated operations. enter.

Fig. 1 is the block diagram according to an embodiment of Chinese input method of the present invention; Fig. 2 is the flow chart of action step; Fig. 3 is the explanatory figure of homophone table example; Fig. 4 is the explanatory diagram of classifying homophone.

Realize the best scheme of the present invention

Below, the Chinese input method of the present invention will be described with reference to the design drawing.

Fig. 1 is the block diagram of an embodiment of the present invention, the Chinese input method in this figure, by having pinyin key 1a, and can input the input section 1 of Chinese phonetic letter sequence (pinyin letter sequence), temporary storage contains various The input buffer part 2 of the input signal 100 of the control signal, the transformation part 3 that converts the sequence of pinyin letters into the sequence of Chinese characters, the dictionary part 4 with the correspondence table between the sequence of pinyin letters and the sequence of Chinese characters, and the sequence of control pinyin letters and sequences of Chinese characters The display control unit 5 for displaying and the display unit 6 for displaying them are constituted.

The input unit 1 includes a pinyin key 1a, and outputs an input signal 100 including various control signals coexisting with pinyin letters.

The buffer unit 2 receives an input signal 100, and outputs input data 101 that is recognized and displayed as pinyin and a conversion control signal 106 that instructs conversion from pinyin to Chinese characters.

The conversion unit 3 includes dictionary access means 3a, read data buffer 3b, and conversion control means 3c. The dictionary access means 3 a receives input data 101 and outputs a search signal 102 according to a control signal 106 . Also, the conversion control means 3C outputs a conversion instruction signal 107 and a display instruction signal 108 based on the conversion control signal 106 . The read data buffer 3b receives and temporarily stores the read data 103, and selects one word from the read data 107 to output as the converted data 104 (the selection method will be described later).

The dictionary unit 4 includes a homophone table 4 a including a correspondence table between pinyin letters and Chinese characters, and outputs the Chinese characters corresponding to the retrieval signal 102 as read data 103 .

The display control unit 5 includes an input data display buffer 5a, a converted data display buffer 5b, and a display buffer 5c. Furthermore, the input data display buffer 5a and the converted data display buffer 5b respectively receive the input data 101 and the converted data 104 and temporarily store them. The display buffer 5C saves the input data 101 and the converted data 104 as described above. An input is made according to a display instruction signal 108 and output as a display signal 105 .

The display unit 6 receives the above-mentioned display signals 105, displays them on the display, and can recognize the operation process.

Fig. 2 is a flow chart showing the action steps of the above-mentioned Chinese input method. In this figure, processing 21 is the operation of typing in the phonetic alphabet sequence, and then processing 22 is to judge whether the phonetic alphabet sequence that has been keyed in in processing 21 is converted into a Chinese character, and returns to processing 21 when not converting, and continues typing The phonetic alphabet sequence is transferred to processing 23 when the conversion is performed.

Processing 23 displays the characters of type A, and processing 24 judges whether the displayed type A characters are desired characters. If it is the desired word, it transfers to processing 25, and when it is not the desired word, it transfers to processing 27.

Processing 25 writes the displayed desired word into the file, and according to this processing, the phonetic alphabet is replaced into Chinese characters.

Processing 27 displays the words of Class B according to the above-mentioned order, processing 28 judges whether the displayed Class B characters are required characters, and transfers to processing 29 when required characters, and transfers to processing 30 when not required characters .

Processing 29 rewrites the above-mentioned reference mark, that is, rewrites the desired word determined in processing 28 to the reference mark displayed first at the time of the next search, and then transfers to the above-mentioned processing 25 .

Process 30 judges whether all the words of class B have been displayed, and if there are still undisplayed characters, transfer to above-mentioned processing 27, and when all have been displayed, transfer to processing 31.

Process 31 the word of C class is displayed according to the order that has been stipulated originally, process 32 judge whether the word of C class that is displayed is required word, when it is required word, transfer to processing 25, when not required word When , transfer to processing 33 .

Processing 33 judges whether the word of C class has all been shown, when there is the word that does not show When , it transfers to the above-mentioned processing 31, and when all of them have been displayed, it transfers to the processing 34.

Processing 34 indicates that the required word does not exist in the homophone table 4a (for example, indicating that [?] is written to the file at the same time).

Processing 26 is for judging whether to continue Chinese input, when continuing, return to processing 21 and repeat the above operations.

In the above-mentioned Chinese input method, orderly homophones are sequentially displayed as candidate required characters, and Chinese input is performed while judging whether they are required characters.

According to the search signal 102 output by the dictionary access means 3a, the read data 103 read includes homophones. For the read data buffer 3b, the required homophone table 4a is temporarily stored.

FIG. 3 is an explanatory diagram of a part of the above-mentioned homophone table 4a. In this figure, homophone table 4a is made up of the phonetic letter field 10 of depositing index during retrieval, the field 11 of depositing A class Chinese Chinese characters, the field 12 of depositing B class Chinese Chinese characters and the field 13 of depositing C class Chinese Chinese characters, and Each homonym in field 12 contains its reference mark field 12a (learning field) in the display order in the B-type retrieval.

The word stored in class A is the most frequently used word among homophones. When typing an index in field 10, that word will be displayed first. The characters stored in category B are some characters with medium usage frequency among homophones, and when searching for the characters in category B, they are displayed according to the order indicated by the reference marks in field 12a. That is, the characters are displayed sequentially from the character just selected. The reference marks need to be rewritten each time the desired word is selected by the operation. The words stored in category C are some words with lower frequency of use among the homophones. When these words are retrieved, they are displayed according to the order determined originally, as in the usual dictionary.

Next, the data buffer 3b is read out, and the content of the above-mentioned homonym table is temporarily stored, according to the conversion instruction signal 107 output by the conversion control means 3C, as the conversion data 104 sequential output. In addition, the conversion control means 3C outputs a display instruction signal 108, and the above-mentioned conversion data 104 is displayed on the display unit 6 in sequence by the display control unit 5, that is, referring to the processing module 41 of FIG. The homophones of the class and the processing module 43 respectively sequentially represent the homophones of the class C, and perform actions according to the judgment of the operation.

Figure 4 is an example of classifying Chinese homophones (especially single-character words). In this figure, the characters represented in category A are characters written in kana that are partially equivalent to Japanese, and they are all frequently used the word.

In this example, some Chinese single-character words equivalent to Japanese kana are mainly described, and the same method as above can be used to realize Chinese input for words with more than two characters.

Claims (1)

1. A kind of Chinese phonetic alphabet string with key entry is transformed into the Chinese character input method of Chinese character string, has following feature:

   N phonetically similar word for Chinese is provided with homophonic character table, this table the highest word of usage frequency as category-A, m moderate word of usage frequency (m≤n-1) as category-B, low usage frequency

   

   Individual word (

   

   ≤ n-m-1) as the C class;

   When having from m word of above-mentioned category-B the word of selection, be beginning with the word of just having selected, first means that order shows and from above-mentioned C class

   

   When selecting a word in the individual word, according to original order of having arranged, the second means that shows in turn with a certain rule;

   When from a said n phonetically similar word, selecting a word, an initial word that shows above-mentioned category-A when not conforming to this word of above-mentioned category-A, is retrieved m word of aforementioned category-B according to aforementioned first means, when each word of aforementioned category-A and category-B does not conform to, retrieve aforementioned C class according to aforementioned second means

   

   Individual word.

Images