Optimized phonetic and sequential code simplified and traditional Chinese character hybrid input method

Abstract

The invention discloses an optimized phonetic and sequential code simplified and traditional Chinese character input method which is a computer Chinese character input method with phonetic codes as main codes and shape codes as auxiliary codes. In the input method, the common coding form of Chinese characters is phonetic codes and shape codes, the shape codes are used for eliminating repeated codes in combination with character frequency, syllable representative characters only have phonetic codes and do not have shape codes, and the shape codes of other Chinese characters are gradually increased along with reduction of the character frequency. The Pinyin input method comprises intelligent Pinyin, has many repeated codes, and is low in efficiency due to character selection. A five-stroke character input method and other phonetic and morphological code input methods are high in learning cost and difficult to grasp. The input method is simple and easy to learn, can be learned immediately, and has the characteristics that word-by-word input is emphasized, word input is not rejected, Chinese characters, words and punctuation marks can be typed in a touch manner, and the input experience is smooth; traditional Chinese characters also have sound and shape codes, so that the requirements of different people can be met, and Chinese character culture transmission is facilitated; the method is close to the reality of Chinese characters, and when the manuscripts are drawn out through a computer, interference on thoughts is little.

Description

TECHNICAL FIELD

[0001] The present application optimizes the phonetic and semantic code mixed input method of Chinese characters, which is a computer input method of Chinese characters with phonetic code as the main form and semantic code as the auxiliary form. BACKGROUND

[0002] There are three kinds of computer input methods of Chinese characters, including phonetic code input method, semantic code input method and phonetic-semantic code input method. Various phonetic code input methods, including intelligent pinyin, have not solved the problem of multiple codes when inputting word by word, and need to frequently use the page key and number key to select the word, which affects the efficiency. The five-stroke input method needs to learn a complex system of Chinese character disassembly and coding rules, only focuses on the character shape, and is not close to language, and the input needs to disassemble and code the characters at the same time, which greatly interferes with thinking. The phonetic code of the natural code input method uses double pinyin, and the semantic code is taken from the pronunciation of the component name, which has certain difficulty in learning and using. SUMMARY

[0003] The general coding form of Chinese characters in the present application is: phonetic code + semantic code, with phonetic code as the main form, gradually increasing the number of semantic code bits through word frequency participation, and following certain rules and steps to completely eliminate the code, that is, to realize that a letter string has a unique Chinese character corresponding to it. The semantic code is the code of six strokes, which is added after the phonetic code according to the stroke order, thereby forming a large number of letter strings matched with Chinese characters. The six strokes refer to: dot, horizontal, vertical, hook, nodule, and fold, and the first letter of their Chinese pinyin is used as the stroke code: d, h, s, p, n, and w (exception). The basic strokes of Chinese characters are usually divided into five kinds (adopted by the five-stroke input method), and the dot and nodule are combined into one kind, but in the present input method, they are distinguished as six kinds in order to facilitate the reference. The realization of Chinese character without code also uses the same method to realize the word without code, and the semantic code of the word uses the semantic code of the last word. This method takes clear and simple code, has less ambiguity, and has no memory burden. Therefore, the present input method is simple and easy to learn, that is, to learn and to do. The input of punctuation marks in the present input method is also improved.

[0004] The feasibility of this invention is based on two points: the frequency distribution characteristics of Chinese characters; the permutations and combinations of phonetic codes and shape codes can generate a sufficient number of letter strings corresponding to Chinese characters. Zipf's Law can be expressed as follows: in the corpus of natural language, the frequency of a word's appearance is inversely proportional to its ranking in the frequency table. Observing the frequency distribution of Chinese characters in this way, it can be seen that although the total number of Chinese characters is very large, the most active ones are only a minority. The Chinese character frequency statistics data of Tsinghua University shows that when the most commonly used character counts are 500, 1000, 1500, 2000, 2500, 3000, and 3500 respectively, the coverage rates are: 78.53%, 91.91%, 96.47%, 98.38%, 99.24%, 99.63%, and 99.82% (the number of Chinese characters used is 6763, national standard character set, and the total number of words in the sample text is 86405823). Assuming the number of syllables in Chinese is 400, when the shape code is increased to two digits, the phonetic-shape code letter strings can reach tens of thousands (the formula is: 400 + 400×5 + 400×5×6), and theoretically, tens of thousands of Chinese characters can be encoded without duplicate codes (in practice, the commonly used simplified Chinese characters are about 3500, and the shape codes of a few commonly used characters may have three or four digits).

[0005] Chinese character input is essentially a retrieval. In this input method, the deep meaning of the shape code is: what position this character ranks in terms of the character frequency in this syllable, and what position this character ranks among the characters starting with a certain point (horizontal, vertical...) in this syllable. This is also a kind of retrieval, but different from the general shape-based retrieval. It is a retrieval from another channel. It focuses on the character frequency, that is, the activity level of the character, and does not delve deeply into more internal structure information in the character shape. Therefore, it does not need to consider problems such as "the last root" and "the last stroke" like the Wubi font. When there are differences in the stroke order, such as for the characters "wan" and "fang", several encoding methods are designed according to the different stroke orders.

[0006] The technical solution of this invention:

[0007] First, there are or can be added six categories of coding in the word library. I. Each syllable is given a code, and each Chinese character is given a code. Through the length of the Chinese character code, the layer of word frequency is involved to realize the non-redundant code of Chinese characters. The principle followed is: high frequency code short, low frequency code long. The method and steps to determine the length of Chinese character code are: first, determine the syllable representative word (the one with high frequency), let it have only sound code but no shape code, second, when two Chinese characters are in the state of redundant code, let the one with low frequency add one shape code, and the shape code after it should be omitted, like this until the implementation of the syllable code word library without redundant Chinese characters. The coding of simplified characters can be completed first. II. Determine 26 one-key words. One-key word generally uses the first letter of pinyin as the code, i, u, v (reflecting the sound code first) are the input codes of the words "no", "on", and "at". One-key word also has its complete code form. Zh, ch, sh are determined as the short codes of the words "middle", "cheng", and "up". III. The coding of words is full pinyin; when there are homophones, add shape code to eliminate redundancy. The shape code of a word is taken from the shape code of the last word. The method and steps to determine the length of the shape code of homophones are the same as those of Chinese characters: the one with high frequency has sound code but no shape code, the one with low frequency adds one shape code, until there is no redundant code for homophones. IV. Simplified and traditional Chinese characters are coded in the same way. The word frequency of traditional Chinese characters is considered lower than that of simplified Chinese characters, so as to determine the shape code of traditional Chinese characters. In this way, it can adapt to the actual application of traditional Chinese characters by different groups of people. V. Chinese punctuation marks, in addition to still being able to follow the existing usual input method: comma, period, semicolon have separate key positions, exclamation mark, question mark, etc. are input by combination key, can also be coded and input: colon, question mark, exclamation mark, space, dash, ellipsis, input by full pinyin or short pinyin code; quotation marks, book title marks, parentheses, square brackets, single quotation marks, single book title marks can be coded and input according to the actual writing order or whole coding input, avoiding or reducing the use of combination keys, avoiding or reducing the movement of the cursor. The above-mentioned Chinese characters, words, and Chinese punctuation marks are generally fixed in the first position of the selected character window when inputting, so as to facilitate blind typing. VI. There are or can be added common symbols in the word library, which can include: ① Chinese pinyin letters with tone marks and the letter ü, ② capital Latin letters, ③ capital and small Greek letters, ④ numbers with circles and brackets, ⑤ four arithmetic symbols, ⑥ other commonly used special symbols: ○, △, ☆, @, ※, √, ×, etc. These symbols are generally fixed in the second to fifth positions of the selected character window for selection.

[0008] Secondly, many designs and personalization methods in the input method can be realized through the "customized words" function. For example, adding or replacing a one-key word in the word library, adding traditional or variant characters in the word library, adding new words in the word library, adding new symbols in the word library, adjusting the front and back positions of symbols in the character selection window, etc. Therefore, in the input method, the space for customizing words should be large enough to display the total space (or total number), used space (or used number), and remaining space (or remaining number).

[0009] Thirdly, the phonetic code of the input method does not use double pinyin, because double pinyin requires memorizing the key positions of many vowels, which requires a lot of effort. The input method uses the method of omitting one or two vowels as a means to reduce the number of keystrokes. The vowels to be omitted refer to vowels containing three or four letters, such as ang, eng, ing, ong, ian, iang, iao, iong, uai, uan, uang, and üan. When encoding, one of the vowels is omitted, which will not cause confusion and result in multiple codes, just like the positive sign of a positive number can be omitted, which is obvious. If two vowels are omitted at the same time, the tone and vowel combination needs to be considered, and eng and üan, or uan and üan, or uang and iang, or uang and iong can be omitted. The ng in the vowel ong can be omitted. The vowel ang is not omitted, and the vowel of a zero initial character is not omitted. Omitting vowels is a personalization method, and users can choose an omission method through the custom word function.

[0010] Advantages of the input method: The phonetic code of the input method uses full pinyin, and the shape code is mostly one or two stroke codes, with no memory burden, so it is simple and easy to learn, and it can be learned and used immediately; The input method values character-by-character input and does not exclude word input, and the input of characters, words, and punctuation symbols can all appear at the top of the character selection window, so the blind typing rate is high, the input speed is fast, the input experience is good, the encoding method is close to the actual situation of Chinese characters, and the interference to thinking is small; The traditional characters in the input method also have phonetic and shape codes, which meet the needs of various groups at home and abroad, and are conducive to the spread of Chinese character culture. DETAILED DESCRIPTION

[0011] Example 1 takes the da syllable as an example to explain the input method of Chinese characters, and also explains the method of determining the length of the shape code of Chinese characters and the editing method of the da syllable Chinese character encoding table.

[0012]

[0013] Explanation of Example 1:

[0014] The representative word of da syllable is "big", because the frequency of "big" is high, the code of "big" is da, only the phonetic code without the shape code, is the shortest code in this syllable, other words must add the shape code, can not be with the code of "big" code. The code of "hit" is dah, because "hit" is the highest frequency of da syllable horizontal pen (note: the code of horizontal pen after the stroke is omitted). "Da" is also horizontal pen, the frequency is lower than "hit", so add a shape code p, the code of "da" is dahp. Similarly, the shape code of "Lao" is one more than "da", the code is dahpn. Da and da both use the second code, because the second stroke is different and does not constitute a code. "Da" is a traditional Chinese character, the input method is the same as the simplified Chinese character, the mixed coding of simplified and traditional Chinese characters in the word library is feasible, which can avoid the switching of simplified and traditional Chinese characters. The input method of "answer" is dap, which is the highest frequency of da syllable stroke in the word.

[0015] The coding table of each syllable of Chinese characters needs to be modified and adjusted repeatedly to gradually improve it. The tedious work can be done with the help of full spelling or Sogou input method.

[0016] Example 2, one key word input method.

[0017]

[0018] Explanation of example 2:

[0019] One key word, here refers to the Chinese character displayed in the first position of the word selection window by pressing a letter key, and then pressing the space bar to select and enter the text.

[0020] One key word, the shallow understanding is to place Chinese characters on A to Z letter keys. It can also be said that 26 characters have simple codes. One key word takes the first letter of Chinese pinyin as the code, and the simple codes of one, no and on are i, u and v. Pressing a letter key and then pressing the space bar can input Chinese characters, which is very economical and efficient, and has good input experience. One key word has only 26, but the coverage reaches more than 20% of the total input. One key word also has its complete code form, such as "de" for "of". In this way, one key word can be replaced, such as j-just, which can be replaced by j-add; z-at, which can be replaced by z-this, without affecting other matters.

[0021] One-key word is a concept that is extracted on the basis of several factors, such as phonetic code priority, rational use of letter string resources, and efficient input (one, no, and three words are coded as i, u, and v, which are also related to pronunciation), and by analogy, the letter strings zh, ch, and sh can also be equipped with simple codes: zhong, cheng, and shang. In this input method, one-key word is a relatively isolated concept, because it no longer needs to introduce two-key words and three-key words, and there is no "key word root" for subsequent use. In the five-stroke font, there is a discussion and design of first-level simple codes (25 words), second-level simple codes (up to 25 x 25 words), and third-level simple codes (several thousand words).

[0022] Example 3, input method of words.

[0023]

[0024] Explanation of Example 3:

[0025] Input of words, directly input the phonetic code of each word, if there are homophonic words, the shape code of the last word needs to be input. In this example, the phonetic codes of the four words are the same, only input jiaoshi, "teacher" is displayed, if "classroom" is needed, the shape code d of the "room" word needs to be added, "classroom" appears in the first position of the word selection window, and is selected into the text by the space key. If "clergy" is needed, the first shape code h of the "soldier" word needs to be added, "reef stone" appears, and the second shape code s of the "soldier" word needs to be added, "clergy" appears in the first position of the word selection window, and is selected into the text by the space key.

[0026] When the intelligent pinyin inputs by sentence, the word that appears in the word selection window is often not the word that is intended in advance, and needs to be repeatedly selected and operated, the input is not smooth, the efficiency is not high, and the input experience is poor. The "emphasis on word-by-word input, and not excluding word input" proposed by the present input method is in line with the actual situation. Emphasis on word-by-word input, the word library file can include fewer unnecessary words, the input method can be slimmed down, and computer system resources can be saved.

[0027] Example 4, input method of Chinese punctuation marks.

[0028]

[0029] Explanation of Example 4:

[0030] In Chinese character input, the input of Chinese punctuation is very frequent, accounting for about one-tenth of the total input, which is a large proportion. In order to make the input smooth and try to type blindly, it is necessary to carefully design the input method of punctuation symbols. Therefore, the use of combination keys to input punctuation symbols should be avoided or reduced, and the left and right adjustment of the cursor should be avoided or reduced. In this input method, the usual method of inputting Chinese punctuation is continued, such as comma, period, semicolon, and colon, which are directly input by keys. Question marks and the like are input by combination keys. The input of emphasis marks and superscripts is set in the "font". In addition, this input method also uses coding to input Chinese punctuation. As with Chinese character input, Chinese punctuation is fixed to appear in the first position of the character selection window for blind typing.

[0031] The coding input of quotation marks can be input separately to the left and right or as a whole, both of which can be typed blindly. The former is the natural writing order, and the latter requires moving the cursor once. The coding input method of parentheses, book titles, and single book titles is the same as that of quotation marks. The coding of these symbols in this example is a personalized method, which can be selected by the user or can be reprogrammed according to personal habits.

[0032] Example 5, input method of other commonly used symbols.

[0033]

[0034] Explanation of Example 5:

[0035] In the current commonly used input method of symbols, there are some inconvenient situations: capital Latin letters require the use of shift keys or enter keys; Chinese pinyin letters with tones and the letter ü require the use of "insert" operations or copying and pasting from elsewhere; Greek letters require the use of "insert" operations; % requires the use of shift keys; serial numbers ①, ②, ③ require the use of "insert" operations, etc. These operations interfere with the continuity of input and affect efficiency. Therefore, in addition to Chinese punctuation symbols, which can be coded, many commonly used symbols can also be coded, which are generally fixed to appear in positions 2 to 5 of the character selection window for selection. The coding of these symbols in this example is a personalized method, which can be selected by the user or can be reprogrammed according to personal habits.

Claims

1. An optimized phonetic code simplified-traditional mixed encoding Chinese character input method, characterized in that: The general coding form of Chinese characters is: phonetic code + form code. The phonetic code is generally in full spelling, and the form code is composed of stroke codes, which are consistent with stroke order. The strokes are divided into six types: dot, horizontal, vertical, hook, nodule, and fold. The first letter of the name of each type is used as the code: d, h, s, p, n, and w (exception). The method and steps to determine the length of the form code of Chinese characters are as follows: first, determine the syllable representative word (the one with high frequency), which only has phonetic code without form code; second, when two Chinese characters are in the state of homonym, the one with low frequency is added with one bit of form code, and the form code after it should be omitted; and third, the above process is repeated until there is no homonym in the phonetic code word library. Chinese characters, words, and punctuation marks are fixed in the first position of the selected character window for blind typing input. In order to reduce the number of keystrokes, part of the phonetic code can be omitted without causing homonym. There are or can be added 26 one-key words in the word library, whose coding is the first letter of their Chinese pinyin, as well as the letters i, u, and v. Zh, ch, and sh are also the simplified codes of Chinese characters, which are generally high-frequency words, and they also have relatively complete phonetic and form codes. There are or can be added traditional Chinese characters in the word library, whose frequency is considered lower than that of simplified Chinese characters, so as to determine the length of the form code. There are or can be added words in the word library. When the phonetic code of a word is the same, the form code is used to eliminate homonym. The form code of a word is the form code of the last character. The method and steps to determine the length of the form code of a word are the same as those of Chinese characters. First, the high-frequency words only have phonetic code without form code. Then, the low-frequency words are added with one bit of form code until there is no homonym in the word. Simplified Chinese characters, traditional Chinese characters, words, and one-key words without homonym are fixed in the first position of the selected character window for blind typing input.

2. The input method of claim 1, wherein: There are or can be added the following situations of omitting the final consonant in the word library: for the final consonant with three or four letters, one type of final consonant can be omitted in the coding of Chinese characters; when two types of final consonant are omitted at the same time, they can be eng and üan, or uan and üan, or uang and iang, or uang and iong; the ng in the final consonant ong can be omitted (the final consonant ang cannot be omitted, and the zero initial consonant word cannot omit the phonetic code); the above methods of omitting the final consonant can also be applied to the coding of words.

3. The input method of claim 1, wherein: There are or can be added the following symbol elements in the word library: ① Chinese punctuation marks, ② Chinese pinyin letters with tone marks and the letter ü, ③ capital Latin letters, ④ capital and small Greek letters, ⑤ numbers with circles and parentheses, ⑥ four arithmetic symbols, and ⑦ other commonly used special symbols such as ○, △, ☆, @, ※, √, and ×. Chinese punctuation marks are generally fixed in the first position of the selected character window for blind typing, and other commonly used symbols are generally fixed in the second to fifth positions of the selected character window for selection.

4. The input method of claim 1, wherein: In addition to the current common input method, punctuation marks can also be input by their full spelling or simplified spelling code, such as colon, question mark, exclamation mark, space, dash, and ellipsis.

5. The input method according to claim 1 and 4, characterized in that: ​ Quotation marks, book names, brackets, square brackets, single quotation marks, single book names can be coded and input according to the actual writing order or whole coding input, avoid or reduce the use of combination keys, avoid or reduce the cursor movement operation.

Images