Chinese character self-assembling input method and etymon keyboard thereof

Abstract

In order to solve the problem that an existing input method cannot have both easy learning and high efficiency, the invention creates a Chinese character self-assembling Chinese input method and a character root keyboard thereof which can freely select roots, autonomously split characters and self-automatically encode, establish Chinese character shape codes and efficiently typewrite on the basis of general common knowledge of the public. The method is suitable for the public cognitive common knowledge, more than 770,000 character splitting and coding schemes and related 3196 kinds of etymon and keyboard key mapping relations are created for more than 20,000 GBK Chinese characters, a user does not need to learn any etymon table, etymon songs, etymon key mapping maps, key names, fonts, character splitting and coding specifications or special rules, and the method is simple and convenient. A user can easily print wanted characters and words only by thought of the existing cognition of Chinese character structures, stroke orders, etymons and pronunciation of the Chinese characters and thought of splitting and coding; through a quantitative analysis algorithm, a character splitting and coding scheme of Chinese characters is preferentially graded and sequenced, and a user can input all GBK Chinese characters and 600,000 phrases in a touch typing manner by applying a simple coding skill without turning a screen.

Description

[0001] Part 1: Technical Field

[0002] "Self-Grouping Chinese Input Method and Its Root Keyboard" (hereinafter referred to as "the present invention") relates to the Chinese character decomposition and encoding technology required for inputting Chinese into a computer system using a keyboard and its supporting Chinese character root keyboard.

[0003] Part 2: Background Art

[0004] The present invention belongs to the field of Chinese keyboard input methods, which mainly includes "pinyin input methods" and "shape code input methods".

[0005] Section 1: Inherent Shortcomings of Pinyin Input Methods

[0006] 1. The comprehensive efficiency of inputting Chinese using pinyin input methods lags far behind that of using shape code input methods (especially for professional words and terms).

[0007] 2. Due to a large number of homophonic words and terms, it is impossible to type most words and terms blindly, and the operation of flipping through the screen to select characters is frequent and inefficient.

[0008] 3. Pinyin input methods lead to a serious degradation of users' ability to recognize and write Chinese characters, which is not conducive to the inheritance and development of Chinese characters.

[0009] 4. The popularity of input methods such as Cangjie and Sucheng in Hong Kong, Macao and Taiwan regions shows that pinyin input methods cannot replace shape code input methods.

[0010] Section 2: Thinking Constraints of Traditional Shape Code Input Methods

[0011] At present, there are many patent applications for shape code input methods. However, before the present invention came out, there had never been a "shape code input method that is easy to learn and use and can be typed blindly in general". This is mainly because the technical personnel in this field are generally restricted by the following thinking constraints:

[0012] 1. Most traditional shape code input methods (hereinafter referred to as "traditional shape codes") focus on pursuing mathematical statistics performance, resulting in relatively complex and rigid decomposition and encoding methods. One must master a large number of rare root characters and complex and strict decomposition and encoding rules, and the entry cost is very high.

[0013] Second, most traditional shape codes focus on saving machine resources, resulting in a stingy number of character codes provided for each Chinese character or word (Note: "character code" refers to the keyboard symbol code that can be recognized and processed by a computer after converting Chinese characters through specific rules, so that users can input Chinese information into the computer through the keyboard. Its specific forms include "stroke code", "full pinyin code", "pure shape character code" in Section 1, Point 4 of Part 4, and "phonetic shape word code" in Section 2, Point 2, etc.). Most Chinese characters and words only correspond to a unique character code. Users must precisely master the cumbersome norms of relevant radicals and strictly abide by many rigid regulations regarding the rules of character splitting and coding in order to give the only correct character code that precisely conforms to the relevant rules for each Chinese character or word, and then they can type out the Chinese characters or words they want. This ignores the vastly different cognitive situations of the vast majority of users and is very unfriendly.

[0014] Third, due to the above two reasons, users must re-learn a whole set of special and proprietary knowledge forcibly stipulated by the inventor before they can start using traditional shape codes. This keeps most ordinary users who do not have a strong learning need, insufficient conditions, and lack of ability out of the door. Moreover, it cannot adapt to the accelerating pace of people's work and life. This is the main reason for the gradual decline of traditional shape codes.

[0015] Fourth, the vast majority of technical personnel in this technical field have a thinking misunderstanding, that is, the number of radicals cannot be too many (usually deliberately controlling the number of radicals to around three hundred). But in fact, the number of radicals is not important. What is important is whether the selected radicals can adapt to the existing cognition of the vast majority of users. If as long as the characters, radicals, and strokes that users themselves recognize are "legal" radicals, then no matter how many radicals there are, it will not increase any additional learning cost for users. Instead, if certain characters, radicals are deliberately and rigidly stipulated as "legal radicals" while the rest are not, it will greatly increase the learning cost for users.

[0016] Section 3. The shape code input method that fully respects the general cognitive common sense of the general public is the best shape code input method

[0017] In the process of promoting the previous invention "Character Grouping Input Method", the inventor communicated with professionals in the input method departments of leading domestic IT companies. They believed that current pinyin input methods have significantly improved the typing experience and efficiency of users through means such as artificial intelligence word selection, cloud dictionaries, and providing a large number of pinyin error tolerance codes, and are also accessible to ordinary people without the need for learning. Therefore, for a new generation of shape code input methods to have the opportunity to compete and give full play to their advantages, they must also achieve "accessible without learning". To this end, the inventor attempted to solve the problem of "accessible without learning" by simplifying the root character recognition and character decomposition and encoding rules, and adding compatible word and character encodings. After repeated improvements in multiple aspects, starting over many times, and hundreds of iterations of optimization, the inventor created an original root character selection and character decomposition encoding concept centered around the user's existing knowledge, and finally, through a qualitative change from quantitative change, unexpected disruptive innovation results emerged.

[0018] Section 4: The rapid development of computer technology provides the possibility for the leapfrog innovation of Chinese keyboard input methods

[0019] After decades of rapid development and popularization, computer technology has become a common tool for the public. Compared with the era when the "Wubi Input Method" was popular, today's personal computers have increased their memory, hard disk capacity, and CPU computing power by a factor of ten thousand compared to early personal computers. People no longer need to be as stingy in using computer resources as they used to be. With the continuous appreciation of human resources, people are increasingly reluctant to spend a large amount of human cost to specifically learn and adapt to those rare and strange root characters and complex and cumbersome character decomposition and encoding rules. The rapid development of computer technology provides fertile ground for the leapfrog innovation of Chinese shape code input methods.

[0020] Part Three: Summary of the Invention

[0021] Section 1: The purpose of the present invention and the technical problems to be solved

[0022] The innovative shape code input method and its keyboard overcome the defects of difficult root character selection, difficult character decomposition, and difficult mapping in traditional shape codes ("mapping" refers to the process and corresponding relationship of assigning root characters to keyboard keys according to rules, the same below), enabling users to form word and character encodings (self - composed characters) by "freely selecting root characters, independently decomposing characters, and self - encoding" based on general knowledge without the need for specialized learning, and inputting Chinese easily and efficiently.

[0023] Section 2: The systematic innovative content of the present invention

[0024] 1. The present invention has achieved a subversive innovation in the technical concept of Chinese root character keyboard input

[0025] (1). The present invention has pioneered a brand - new concept centered around the user's existing cognitive common sense and oriented towards the user's usage experience.

[0026] Traditional shape codes adhere to the outdated technical concept centered on saving machine resources and oriented towards the mathematical statistics performance of coding. They dominate the selection of root characters and the formulation of character splitting and coding rules, optimize the coding performance through mathematical statistics methods, reduce the duplicate code rate of coding, and improve the simplicity of coding, aiming to achieve higher typing efficiency while saving machine resources as much as possible.

[0027] The present invention fully respects, deeply understands, and deeply develops and utilizes the cognitive laws and cognitive resources of the public. It dominates the formulation of Chinese character splitting and coding rules, the selection of root characters, the priority grading of root characters, and the priority grading of coding according to the cognitive psychological laws of users. Combining information coding means, it optimizes the mathematical statistics performance of the coding system and takes into account the user requirements of easy learning and high efficiency.

[0028] (2) The present invention has pioneered a brand-new concept of "not relying on specialized knowledge and not transforming the user's subjective cognition".

[0029] The inventors of the old shape codes did not deeply understand and attach importance to the existing cognitive resources and inherent cognitive laws of the vast number of users, nor the important significance of the practicality issue of Chinese character shape code input technology. They did not even fully consider the great difficulty of changing people's existing subjective cognition. Instead, they required users to change their existing subjective cognition to adapt to the specialized knowledge of the coding system rigidly stipulated by the inventors with the goal of saving computer resources and optimizing mathematical statistics performance. This move has led to the root character selection and character splitting and coding rules of traditional shape codes not conforming to the subjective cognitive laws of the vast number of users, bringing high learning costs and poor usage experiences to users.

[0030] The present invention has pioneered a brand-new concept of "not relying on specialized knowledge and not transforming the user's subjective cognition". It no longer attempts to change the user's cognition, but instead adapts and accommodates the user's existing cognition by transforming the coding rules and coding system, thereby minimizing the user's learning cost. Combining coding means, it develops, utilizes, and optimizes the configuration of the user's cognitive resources, aiming to achieve universal blind typing of words and phrases while optimizing the user's typing experience. Users do not need to specifically learn. According to general common sense and extremely simple rules, following their own existing subjective cognition and thinking habits, they can easily type the Chinese words and phrases they want by splitting characters and coding as a matter of course.

[0031] The present invention has pioneered an extreme compatibility innovation concept of "multiple character splits for one character", "multiple keys for one root", "multiple codes for one character", "multiple codes for one word", "massive compatibility", and "varying character splitting and coding methods for different people" that adapts to the diversity of user subjective cognition, subverting the backward concepts in traditional shape codes, such as each root character can only have one mapping method, each word and phrase has only one word and phrase code, different users must adopt exactly the same word and phrase code, it cannot be compatible with the diversity of user subjective cognition, it must rely on specialized knowledge to use, and it must transform the user's subjective cognition to adapt to the rigid character splitting and coding rules formulated by the inventors to save computer resources, etc.

[0032] Second, the present invention fully respects, optimally allocates, and deeply develops and utilizes the general cognitive laws and cognitive resources of the public.

[0033] (1). Advocate taking the existing cognitive common sense of users as the criterion, and abandon the rigid regulations and specialized knowledge beyond the public's cognitive common sense.

[0034] Traditional shape codes often rigidly stipulate that a series of "specialized knowledge" must be learned before users can use them, such as: more than a dozen character decomposition rules, multiple font structures, multiple keyboard partitions, more than twenty key names, multiple special case coding rules; stipulate that some structures inside Chinese characters are radicals while others are not; stipulate the keyboard mapping letters for two or three hundred radicals, and it is difficult to complete Chinese input operations without learning any one of them; and these rigidly stipulated "specialized knowledge" is useless in people's daily lives.

[0035] The present invention advocates that: taking the existing cognitive common sense of users as the criterion, what is a "correct" radical, character decomposition and coding method, and word coding should not be stipulated by "authoritative persons", but should be determined by the public's cognitive common sense; the writing structures that ordinary people recognize and repeatedly appear inside Chinese characters are all "correct" radicals, the character decomposition and coding methods that ordinary people take for granted are "correct", and the radical mapping and coding methods that conform to the existing cognitive common sense of ordinary people are all "correct" ( "mapping coding" refers to the process and result of replacing radicals with specified key characters according to rules; the same below), and the word codes obtained therefrom are all feasible.

[0036] (2). Prioritize using radicals with more strokes: Chinese people have a strong cognitive ability for Chinese characters. They can remember unfamiliar Chinese characters for a long time after looking them up one or two times. For the general public, "characters within characters" are easier to remember than "radicals", and the number of strokes of "characters within characters" is generally more than that of "radicals"; in addition, the fewer the number of strokes of the radicals split out, the more likely they are to be affected by users' writing habits and cognitive deviations in stroke order and result in differences (for example: the "word coding" with single strokes as radicals is most likely to be affected by users' writing habits and cognitive deviations in stroke order and cause "stroke order errors", and it is most likely to cause users to be unable to type out the words they want). Therefore, the present invention advocates that users preferentially split out the radicals with more strokes among the radicals they recognize during the process of character decomposition.

[0037] (3). There is no need to distinguish between radicals and characters within characters: It is very difficult for ordinary people to distinguish whether each radical is a "radical" or a "character within a character", but it is very easy to identify which one of the two radicals has more strokes. Therefore, the present invention treats "radicals" and "characters within characters" equally. Users do not need to distinguish between "radicals" and "characters within characters", nor do they need to distinguish between classifications such as "basic radical characters", "basic-level characters", "dot-added characters", "key names", and "special characters", and only need to select the one with more strokes among the radicals they recognize.

[0038] (4) Massive coding with all-round coverage to accommodate the cognitive differences of the public: The cognitive levels of the general public regarding the characters within characters and radicals vary greatly. Therefore, the present invention aims to comprehensively cover and accommodate various root characters and character decomposition coding methods that conform to the general cognitive common sense of the public. The range of character decomposition and root selection extends from large characters within characters, small characters within characters, radicals to strokes, with seamless connection, fully adapting to the cognitive differences of the general public with widely varying cognitive levels. Through the strategy of "massive coding with all-round coverage to accommodate the cognitive differences of the public" that comprehensively decomposes and exhausts all reasonable character decomposition methods, users with different cognitive levels and preferences can obtain a smooth typing experience.

[0039] (5) Placing the codes of commonly used words and phrases at the front position within the code table: People can hardly tolerate even one additional operation of flipping the screen to select words during the input of commonly used words and phrases; however, they don't mind a few more operations of flipping the screen to select characters during the input of low-frequency words and phrases. Therefore, the present invention sorts the words and their codes in the code table according to the composite index of the code string and the character frequency or word frequency, so that the words and phrases with higher character frequency / word frequency are presented at the front position on the character selection screen (see item (8) of item (2) in section 9 of part 4 for details).

[0040] III. The present invention takes the following measures to fully utilize and develop the cognitive laws and cognitive resources of the public

[0041] (1) Greatly simplified the root selection rule: During the process of character decomposition, what root characters the user can decompose depends on the user's cognitive level and varies from person to person. There is no single correct standard. The user can decompose any root character that they recognize without any additional learning or deliberate identification and selection. When the user encounters any root character they don't recognize, they can simply decompose it into root characters that they know.

[0042] (2) Greatly simplified the character decomposition rule: During the process of character decomposition, the specific character decomposition method depends on the user's cognitive level and varies from person to person. There is no single "correct" standard. The user can decompose the characters according to the stroke order and writing structure they understand, and can decompose any root character that they recognize. As long as they try to select the one with a relatively larger number of strokes among the root characters they know (even if the user selects a root character they know that does not have the largest number of strokes or even just a single stroke, after entering the corresponding code, the character can still be typed, but it may be necessary to flip through a few more pages of the character selection screen), so the user can easily decompose Chinese characters without special learning.

[0043] (III) The basic mapping rules of radicals are greatly simplified: All radicals in the present invention can use their first pinyin letters as coding letters (see the first point of the third section of the fourth part [Table 1] for details); for radicals with multiple pronunciations (hereinafter referred to as "polyphonic radicals"), if they have a "publicly known first pronunciation" (*Note ①), their publicly known first pronunciation is used, otherwise the first pinyin letters of any pronunciation commonly used in society can be used as coding letters, and the specific pronunciation to be used is up to the user; if the user does not know the pronunciation and pinyin of a certain radical, the question mark "?" can be used as the coding symbol of the unknown pronunciation, so that the user will not be stumped by radicals that he cannot read. (*Note ①: If the first pronunciation of a "polyphonetic root" is widely known and very commonly used, while its other pronunciations are rarely used, its first pronunciation is called the "publicly known first pronunciation", and the same definition applies to the entire text. For example: Everyone knows that the word "大" is pronounced "da", and very few people pronounce it as "tai", so "da" is the "publicly known first pronunciation" of the word "大". Another example: " 从" is usually pronounced "cong", and very few people pronounce it as "zong", so "cong" is the "publicly known first pronunciation" of the word " 从")

[0044] (IV) Expanded the mapping of radicals on the key positions of radical-like letters: Through simple expansion mapping clues, “彐EF讠辶丨口囗○ The 18 characters "凵匚]冂冂丷乂" can be assigned to the key positions of their pinyin initial letters, and can also be additionally assigned to the corresponding letter keys with similar shapes, which significantly increases the amount of blind typing encoding (see Part 4, Section 3, Point 2 [Table 2] for details).

[0045] (V) The mapping of radicals on punctuation key positions has been expanded: By simply expanding the mapping clues, the six high-frequency radicals "木月扌氵钅釒" can be assigned to the key positions of their first pinyin letters, and can also be additionally assigned to the five punctuation key positions of ",. / '[", which significantly increases the number of blind typing codes (see Part 4, Section 3, Point 3 [Table 3] for details).

[0046] (VI) The mapping of radicals on numeric keys has been expanded: By simply expanding the mapping clues, the radicals and the phonetic letters or punctuation marks of their corresponding keys can be further assigned to the corresponding keys of the numeric keypad, so that the present invention can type on the letter punctuation keyboard and the numeric keypad respectively according to the same set of character decomposition encoding rules (see Part IV, Section 3, Point 5 [Table 5] for details).

[0047] (VII) Massive coding covers all aspects and is compatible with differences in public cognition: Taking the common knowledge of ordinary users as the core foundation of the coding system, by exhaustively enumerating the word-splitting coding methods, it strives to fully cover and be compatible with various word-splitting coding methods that may be in line with the common sense of ordinary users. By using massive coding to tirelessly cover and be compatible, the compatibility of the coding system is maximized.

[0048] (8). In aspects such as root selection, character decomposition, and root mapping encoding, comprehensively cover and be compatible with the existing common sense of ordinary users: The root set should comprehensively cover and include as many radicals and characters within characters contained in GBK Chinese characters as possible. The character and word encoding should comprehensively cover and be compatible with various character decomposition encoding methods that conform to the common sense of ordinary users. For poly-radical characters, if there is a "well-known first pronunciation", use its well-known first pronunciation; otherwise, any commonly used pronunciation in society can be used, making the roots and their mapping methods highly compatible.

[0049] (9). Fully respect the individual cognitive differences of users:

[0050] The "root set" of the present invention comprehensively covers the roots within GBK Chinese characters. The number of roots actually used by users during the process of using the present invention varies from person to person and with time. Which specific roots in the "root set" are used by each user during the process of typing with the present invention is completely determined by the user's cognitive situation of the roots during the typing operation, and there is no absolute "correct" standard. Users can select the roots they need without conflict and can easily type out the words they want;

[0051] During the mapping encoding process of poly-radical characters in the present invention, which specific pronunciation's first letter, shape-similar letter, or punctuation mark of the poly-radical character will be used as the key mapping letter can vary from person to person for different users and also vary with time for the same user at different learning stages. It is completely determined by the user's cognitive situation of the roots during the typing operation, and there is no absolute "correct" standard. Each user can select what they know and type out the words they want.

[0052] (10). Create a function to dynamically adjust the priority of "character and word encoding" according to the actual usage situation of users

[0053] In the present invention, the usage situation of users for character and word encoding will change the ranking of the used character and word encoding in the total code table. The more frequently used a character and word encoding is by users, the closer it will move to the top of the character selection screen (see Section 2, Point 9 of Part 4). This will make it easier for users to perform blind typing for the more frequently used words and give users a good typing experience of "the more you use, the smoother it gets".

[0054] In summary, the present invention uses extremely simplified root selection methods, character decomposition methods, and root mapping encoding rules to perform decomposition and exhaustive expansion of character and word encoding operations on tens of thousands of Chinese characters, creating a large number of encodings that comprehensively cover and are compatible with various character decomposition encoding methods that adapt to the cognitive laws of the public, achieving the ease of learning and use and dynamic code arrangement of the shape code input method, making it smoother for users to use.

[0055] 4. Through the simple "root supplement and character decomposition method", expand and optimize the single-character encoding, and achieve the general blind typing of GBK Chinese characters

[0056] "Code conflict" refers to the phenomenon that different Chinese words and phrases have the same "word and phrase code"; "code conflict rate" refers to the number of "code conflicts" that occur in every 100 word and phrase codes on average. "Blind typing" refers to the phenomenon that after typing the code of a word or phrase, the desired word or phrase can be directly entered into the text ("single-code blind typing"), or the word or phrase appears at the top of the character selection screen, and the user can press the space bar to enter the word or phrase into the text ("first-code blind typing"), without any operation of flipping the screen or pressing the number keys to select Chinese words and phrases.

[0057] Universal "blind typing" of words and phrases is the key to high-speed typing. The key to achieving the former is to reduce the code conflict rate: increase the dispersion of coding elements (diversification of radicals) and appropriately increase the coding length (ensuring that each Chinese character has a "word and phrase code" of no less than 4 letters).

[0058] Among the key letters, the other 23 letters except "ivu" can be used as the initials of Chinese pinyin. The information capacity of a 1-letter initial coding of Chinese pinyin = 23; the information capacity of a 2-letter coding = 23^2 = 529; the information capacity of a 3-letter coding = 23^3 = 12167; the information capacity of a 4-letter coding = 23^4 = 279841. With word and phrase codes composed of no more than 3 initials of Chinese pinyin, only 23 + 529 + 12167 = 12719 non-repeating codes can be provided, which cannot meet the requirement that each of the 21003 Chinese characters in the GBK character set can be assigned a non-repeating exclusive code. By using word and phrase codes composed of no more than 4 initials of Chinese pinyin, 23 + 529 + 12167 + 279841 = 292560 non-repeating codes can be provided. In theory, it can meet the requirement that each of the 21003 Chinese characters in the GBK character set and 87887 Chinese characters in the GB18030 character set can be assigned at least 1 non-repeating exclusive code. Therefore, the present invention uses 4 letters as the maximum coding length for a single Chinese character.

[0059] During the process of splitting Chinese characters into radicals, if larger radicals are used ("larger radicals" refer to radicals containing more strokes inside, the same below), the number of radicals obtained by splitting the Chinese characters will be relatively small, and the resulting codes will be relatively short, which is more likely to cause a high code conflict rate where a large number of different Chinese characters use the same short codes; if smaller radicals are used ("smaller radicals" refer to radicals containing fewer strokes inside, the same below), on the one hand, because the number of radicals split from each Chinese character is too large, the resulting codes will be too long, reducing the typing efficiency, and on the other hand, it will lead to insufficient diversity of radicals and insufficient dispersion of coding elements, which will also cause a significant increase in the code conflict rate of the first 4-bit codes. Typically, for example, there are only five coding elements of "dot, horizontal, vertical, left-falling stroke, and folding stroke" for "strokes", resulting in the codes of the "stroke input method" being not only long, but also having a very high code conflict rate for the first 4-bit codes, and it is more likely to have differences in stroke order.

[0060] To solve the above problems, the present invention allows the user to first extract the "sub-root character" with the largest number of strokes from the Chinese characters to be input. If the number of "sub-root characters" extracted is less than 4, then further extract the "grandson root characters" from the already extracted "sub-root characters" to supplement the number of root characters until the total number of extracted root characters is supplemented to 4 (see Section 1, Subsection 2, Point (2) of Part Four, "Root Character Supplement and Character Decomposition Method"). This measure can ensure that each Chinese character can easily give a word code containing 4 letters. Theoretically, on average, each Chinese character in the GBK character library can be allocated more than 13 non-repeating coding spaces (23^4 = 279841; 279841 / 21003 = 13.32). This provides abundant coding space for the present invention, thus ultimately achieving the excellent effect that most modern Chinese words and phrases can be easily "typed blindly" (for technical effects, refer to Section 9 of Part Four).

[0061] V. Achieving Universal and Efficient Blind Typing of Common Phrases through the Method of Combining Phonetic and Shape Encoding

[0062] To meet the needs of professional typists for efficient input of phrases, the present invention has created an efficient encoding input method of "combining phonetic and shape encoding" for phrases (see Section 4, Subsection 2, Point 2 of Part Four), enabling most common phrases to be typed blindly, and nearly 600,000 practical phrases can be input without screen flipping.

[0063] VI. Achieving the Function of High-Speed Typing on the Numeric Keypad

[0064] The present invention expands the mapping range of root characters through the "root character → letter punctuation → numeric key mapping method" (see Section 1, Subsection 3, Point (2), Sub-point 4 of Part Four), enabling root characters to be further specified from the letter punctuation key positions to the numeric key positions through simple extended mapping clues, so that the present invention can easily and quickly type on both the large letter keyboard and the small numeric keyboard.

[0065] VII. Reasonably Arranging the Priority Ranking of Word Codes to Achieve the Optimization of the Coding System

[0066] The present invention greatly optimizes the matching and development and utilization of the coding, its root character keyboard, and the public's cognitive resources through technical means such as analysis, comparison, classification, selection, compatibility, organization, and ranking (refer to the initial priority ranking method of "pure shape character codes" in Section 2, Subsection 9, Point (8) of Part Four), achieving "ready to use without learning" for ordinary users and universal blind typing of daily words and phrases.

[0067] Part Four Specific Implementation Modes

[0068] Predicate conventions: ① The keyboard symbol "x" is always expressed as "{x}"; ② The word code "xxxx" is always expressed as "{xxxx}"; ③ A string of characters consisting of at least one radical is called a "radical string"; ③ The "radical string" obtained by splitting Chinese characters (only the first four radicals are taken) is called a "character splitting scheme"; ④ When giving an example of a "character splitting scheme", "丶一丨丿 " corresponds to "dot, horizontal, vertical, left-falling stroke and right-falling fold". ⑤ The "nth section and nth point" quoted in the text, unless otherwise specified, are all from the fourth part of this article.

[0069] For example, the word code of the word "编" is {sb}, which means that by typing the two letters "sb" (without the angle brackets {}, the same below), the word "编" will appear in the list of words to be selected on the word selection screen; another example: the key symbol "a" on the keyboard is expressed as {a}, the key symbol b is expressed as "{b}", and the key symbol semicolon is expressed as "{;}"; another example: the word decomposition scheme of the word "马" is " one".

[0070] Section 1: Create a "single Chinese character input method" that adapts to the public's common sense

[0071] 1. Radicals: They are the structural components that make up Chinese characters, and are divided into three types: "strokes", "radicals", and "characters within characters";

[0072] (I) Stroke: It is the most basic radical. In the decomposition encoding process of the present invention, all "radicals" composed of single "strokes" are preferentially regarded as "strokes" (for example, but not limited to "、一丨丿乙"), and it is not recommended to regard them as "characters in characters" radicals; "strokes" can be divided into five types: "dots, horizontals, verticals, left strokes, and right strokes": ① all short strokes without corners, all strokes extending to the lower right without corners (na), are classified as "dots" (such as "丶"); ② all strokes extending to the right horizontally or extending to the right diagonally upward, and 1. All strokes without corners (such as the "提" in the characters "一" and "七") are classified as "horizontal", 2. All vertical strokes without corners (such as "丨") and all strokes that extend vertically downward and then turn to the upper left (such as the "竖钩" in "水") are classified as "vertical", 3. All strokes that extend to the lower left without corners are classified as "撇" (such as "丿"), 4. All strokes except "竖钩" that have a corner or are rounded in the middle of the extension are classified as "折" (such as: "乙、、、乛、乚、○", etc.).

[0073] (ii) Radicals: They are composed of no less than two strokes and do not have complete meanings. They cannot be used as complete and independent Chinese characters in modern Chinese sentences, but can be used as components of Chinese writing structures within Chinese characters (for example, but not limited to: “艹、氵、扌、钅、疒、勹、、亻、久、、癶、灬…”, etc.).

[0074] (3) Chinese characters within Chinese characters: They are composed of no less than two "strokes", have complete literal meanings, can be used as complete and independent Chinese characters in modern Chinese sentences, and can also serve as structural components within Chinese characters (such as but not limited to: "wood, fire, earth, metal, water, sun, moon, sky, people, mouth, field, mountain, grain, white, black...", etc.).

[0075] II. Chinese character splitting rules

[0076] (1) Simple Chinese character splitting method: A simple method of splitting a single Chinese character into a root string

[0077] Based on the user's understanding of the writing structure and stroke order of Chinese characters (where "Chinese characters" include Chinese characters and the roots within Chinese characters), starting from the beginning of the remaining part of the Chinese character to be split, the user looks for a "Chinese character within a Chinese character" or "radical" that they recognize and splits it out. When no "Chinese character within a Chinese character" or "radical" that they recognize is found at that position, then the first "stroke" of the remaining part is split out. For a Chinese character composed of a single stroke, the single stroke is directly taken as the "root" within the Chinese character.

[0078] Repeat the above operation until the first 4 roots in the writing order within the Chinese character are split out (when the number of roots contained in the Chinese character is less than 4, split until it is finished), and then, referring to the writing order, arrange them into a "root string".

[0079] The above method of splitting Chinese characters is called the "simple Chinese character splitting method", and the resulting "Chinese character splitting plan" is called the "simple Chinese character splitting plan". Users with different cognitive levels and preferences may have different splitting methods for Chinese characters that are not single strokes. Users should adopt a specific "simple Chinese character splitting plan" that suits their cognitive level and preferences. For example: one = horizontal stroke; two = horizontal stroke + horizontal stroke; three = two + horizontal stroke; ten = horizontal stroke + vertical stroke; people = left-falling stroke + dot; wide = dot + factory radical; four = mouth + son radical; Zhang = bow radical + long radical; Li = wood radical + son radical; Wang = dry radical + horizontal stroke; sheng = cow radical + horizontal stroke.

[0080] (2) Root supplementing Chinese character splitting method: A method of supplementing roots to the root string

[0081] When the number of "roots" contained in the "root string" in the "Chinese character splitting plan" is less than 4, the following method can be used to append "roots" to the "root string" to increase the number of "roots" contained in the "root string" to 4:

[0082] 1. The "radical string" obtained by splitting Chinese characters according to the "simple character splitting method" is called a "sub-radical string", and the "radical" in the "sub-radical string" is called a "sub-radical"; the more subdivided radical obtained by splitting the "sub-radical" according to the "simple character splitting method" is called a "grandson radical" (when the "sub-radical" is a single stroke, its "grandson radical" is also the single stroke); the first "sub-radical" in the "sub-radical string" is split according to the "simple character splitting method", and the first "grandson radical" obtained is called the "first grandson radical", and the second "sub-radical" in the "sub-radical string" is split according to the "simple character splitting method", and the first "grandson radical" obtained is called the "second grandson radical";

[0083] 2. Append the "first grandson radical" to the end of the "radical string". When the "radical string" has enough 4 radicals, stop appending. Otherwise, append the "second grandson radical" to the end of the "radical string". Repeat the stroke of a single-stroke Chinese character until it has four radicals.

[0084] The above method of supplementing the character root of Sun is called "root-supplementing character splitting method". The splitting scheme obtained by splitting Chinese characters according to the "root-supplementing character splitting method" is called "root-supplementing character splitting scheme". For example: the "root-supplementing character splitting scheme" of the character "Li" is "木子十了"; the "root-supplementing character splitting scheme" of the character "Li" is "王里干甲"; the "root-supplementing character splitting scheme" of the character "二" is "一一一一".

[0085] In the process of implementing the "root-supplementing and character-splitting method", when users with different cognitive levels and preferences have different ways of splitting the first "sub-radical" and the second "sub-radical" of the Chinese character they want to input, correspondingly, users with different cognitive levels and preferences may have different implementation methods for adding the "first grandson radical" and the "second grandson radical" to the Chinese character. For example, the "root-supplementing and character-splitting method" for the character "理" can also be "王里".干日" ("里" can be split into "甲二" or "日土"); the sub-radical of "的" = 白+勺, the sub-radical "白" can be split into 2 ways: ① "丿+日" or ② "亻+彐", the sub-radical "勺" can be split into 2 ways: ① "勹+丶" or ② "丿++丶"; so the "root-supplementing and character-splitting scheme" of "的" can have 2*2=4 ways: ①白勺丿勹, ②白勺丿丿, ③白勺亻勹, ④白勺亻丿.

[0086] 3. Root mapping encoding rules

[0087] Each radical in the radical string obtained by splitting the Chinese character according to the character splitting rules is converted into a corresponding key position character according to the mapping relationship between the radical and the keyboard key position stated in [Table 1], [Table 2], [Table 3], [Table 4], and [Table 5] in Section 3. The "word code" consisting of no more than 4 key position characters obtained is the "pure form code" of the Chinese character.

[0088] (1). Basic mapping method: Ordinary users should assign each root in the "root string" to the corresponding initial letter key of the pinyin according to [Table 1]. When a root has multiple pronunciations, users should, according to their own cognitive preferences, select the initial letter of one specific pronunciation as its key mapping letter: Users can input all GBK Chinese characters only by using this basic mapping method.

[0089] (2). Extended mapping method: The following several extended mapping methods are available for users with relevant personalized needs to choose independently:

[0090] 1. For those who prefer shape - similar associations, they can assign some or all relevant roots in the root string to the shape - similar letter keys according to [Table 2];

[0091] 2. For those who pursue fast touch - typing, they can assign some or all relevant roots in the root string to the punctuation keys according to [Table 3];

[0092] 3. For those with a strong southern accent, they can assign some or all relevant roots in the root string to the corresponding "southern - accent initial letter" keys in the way stated in [Table 4] (The word codes using the "southern - accent initial letter" encoding are collectively called "southern - accent compatible codes");

[0093] 4. For those using the numeric keypad, first map the root string to letter and punctuation encodings according to [Table 1], [Table 2], [Table 3], and [Table 4], and then further assign all the letter and punctuation encoding characters obtained in the previous step to the corresponding numeric keys according to the mapping relationship described in [Table 5] (This mapping method is called "root → letter - punctuation → numeric - key mapping method" throughout the text);

[0094] 5. When a user knows that a writing structure is a root but doesn't know which specific key it should be assigned to, the question mark "?" can be used as a "universal root substitute" to assign the root to the "?" key (it is the " / " key on the numeric root keypad);

[0095] (3). Diversified compatibility method for character - splitting encoding methods:

[0096] Users with different cognitive levels and preferences may have different specific character - splitting methods for the same Chinese character and different specific mapping methods for the same root (different mapping methods of roots are shown in [Table 1], [Table 2], [Table 3], and [Table 4]). As a result, the "pure - shape character codes" obtained by users with different cognitive levels and preferences for character - splitting encoding of the same Chinese character may have different results.

[0097] (4). Examples of root mapping codes: one = horizontal stroke = h; two = two horizontal strokes = hh; three = two horizontal strokes = eh; ten = horizontal and vertical strokes = hs; person = left-falling and dot strokes = pd; wide = dot and factory = dc; four = mouth and son = ke; Zhang = bow and long = gc; Li = wood and son = mz; Wang = dry and horizontal stroke = gh; sheng = ox and horizontal stroke = nh.

[0098] 3. Classification of root mapping codes:

[0099] For the convenience of statement, the following terms are agreed upon: ① According to the "root mapping coding rule", the "pure form character code" obtained by mapping and coding the root string of the "simple character splitting scheme" is called the "simple character code"; ② The "simple character code" with less than 4 characters is called the "simple short code"; ③ According to the "root mapping coding rule", the "pure form character code" obtained by mapping and coding the root string in the "root supplement character splitting scheme" is called the "root supplement character code"; ④ The "simple character code" and "root supplement character code" with no less than 4 characters are collectively called the "four-digit character code".

[0100] 4. Input method for a single Chinese character:

[0101] (1). Simple typing method: After the user finishes typing the "simple character code" on the root keyboard (when typing the "numeric character code" described in Section 2, Point 10, the dot "." must be used as the start and end symbol), corresponding measures shall be taken according to the situation:

[0102] 1. When the Chinese character the user wants does not appear on the character selection screen, the user should use the page turning key to flip through the content on the character selection screen until it appears on the character selection screen or until all the content on the character selection screen has been flipped through; 2. When the Chinese character the user wants appears on the character selection screen, the user should enter the serial number corresponding to the Chinese character on the character selection screen to select and input the Chinese character. If it appears at the top of the character selection screen, the user can press the space bar to input it; 3. When the Chinese character the user wants appears on the character selection screen in the touch screen, the user should click on the Chinese character on the character selection screen to select and input it; 4. When the user still cannot find the Chinese character they want after flipping through all the content in the candidate word list on the character selection screen, it means that the code entered by the user contains a special character splitting coding method that is not compatible with the present invention. The user should use the built-in code query dictionary of the present invention to query the preferred character splitting coding method for this Chinese character (for the usage method of the code query dictionary, please refer to the statement in Section 8, Point 6 (1)).

[0103] (2). Root supplement typing method

[0104] After typing the "simple short code" of a Chinese character, when the desired Chinese character does not appear on the character selection screen, if you do not want to scroll through the screen to find the character, you can use the "root supplement and character decomposition method" and the "root mapping coding rule" (see the second and third points in the first section), supplement the root of the "simple short code" and map and code it into a "root-supplemented character code", and then type its "root-supplemented character code", then most daily Chinese characters can be typed blindly, and all GBK Chinese characters can be input without scrolling through the screen (see the ninth section for the relevant technical effects).

[0105] (3). One-letter blind typing method

[0106] In addition to the above blind typing methods, the present invention provides an additional "one-letter blind typing" privilege for 26 ultra-high-frequency Chinese characters, where typing a single letter followed by a space allows for blind typing: An=a, Bu=b, Chang=c, De=d, Er=e, Fa=f, Ge=g, He=h, Yi=i, Jing=j, Ke=k, Le=l, Men=m, Ni=n, Kou=o, Ping=p, Quan=q, Ren=r, Shi=s, Ta=t, Yao=u, Wei=v, Wo=w, Xiao=x, Yao=y, Zai=z.

[0107] (4). Character decomposition and coding techniques that help improve the blind typing rate

[0108] Each Chinese character can have multiple different character decomposition and coding methods. Users can type the desired words and characters by simply decomposing and coding the characters according to common sense. However, in the general code table of the present invention, different character decomposition and coding of the same Chinese character are arranged from high to low according to its "priority classification" (see the initial priority classification and sorting method of "pure shape character codes" in the eighth point of the ninth point in the second section). Therefore, the "pure shape character codes" obtained by the character decomposition and coding methods that meet the following operation rules occupy a more priority position in the code table. Using these codes with priority positions can more easily achieve the technical effects of blind typing and inputting Chinese characters without scrolling through the screen. Therefore, if users want to more easily type words and characters blindly, they can understand and get used to the following "character decomposition and coding techniques" and follow the clues:

[0109] 1. All recognizable roots can be used, and decompose characters according to common sense: Users do not need to distinguish which characters within characters and radicals are available roots and which are not. Whether it is a character within a character, a radical, or a stroke, as long as it is a root that the user recognizes, it is an available root, and any character decomposition method that conforms to the user's general common sense is an available character decomposition method. This is the most basic principle of the character decomposition and coding method of the present invention.

[0110] 2. Do not decompose recognizable roots in detail, and preferentially select large roots:

[0111] ① Do not decompose recognizable roots in detail: When the user extracts the "sub-root" from the Chinese character to be input, any character within a character or radical that the user recognizes can be used as a root, and it should be taken out as a whole without decomposition.

[0112] ② Give priority to large roots: In the process of decomposing characters, users do not need to distinguish between "characters within a character" and "radicals". They only need to take out the "radicals" with more strokes that they know. If the character within a character or the radical can be taken out, the single stroke will not be taken out.

[0113] 3. Unrecognizable radicals, group them together or break them down: When users encounter a radical they don’t recognize during the character decomposition process, if they can include it in a larger “radical” they recognize, they will include it in the larger “radical” and decompose it together. Otherwise, they will break it up into smaller “radicals” they recognize. If the user cannot find a character or radical they recognize, they will decompose it into the first stroke at that location.

[0114] 4. The monomer structure is written in stroke order to simulate writing and remove the big root: When there are no obvious discrete components inside the split Chinese character, the remaining part of the Chinese character should be simulated to write each stroke from the beginning. The largest root recognized by the user ("largest root" refers to the root with the largest number of strokes among the optional roots at the current position being split, the same for the whole text) is the root to be removed, such as: "木" = "十人", "大" = "丶", "王" = "干一", "男" = "田力".

[0115] 5. Two roots share the same stroke, and priority is given to splitting it: If two roots share the same stroke, priority is given to splitting it and splitting it into two related roots; if the user does not split the shared stroke but splits one of the related roots, he can still type the desired Chinese character, but he cannot enjoy the priority blind typing treatment. For example: "戠" is split into "音戈" (it can also be split into "音 丿丶" or "立日戈" or "立日 "丿" (heavy) is divided into "千里" (it can also be divided into "千日二" or "丿一天土" or "丿一申二" or "丿一里"); "薰" is divided into "千黑" (it can also be divided into "丿一黑" or "丿一囗丷" or "千囗丷二").

[0116] 6. For "polyphone radicals", users should use the pronunciation they are most familiar with to implement mapping coding: the present invention puts the coding of the pinyin initials of the most popular pronunciations among the general public of the Chinese characters containing polyphone radicals in the highest priority position in the code table. Users use the pronunciation they are most familiar with to map the polyphone radicals, which will make it easier to obtain priority blind typing convenience when typing.

[0117] 7. Use the quick mapping method: Users who want to improve their blind typing rate can understand and use the similar letters iouv in the "radical→similar letter mapping method" described in [Table 2] and the "radical→punctuation mark mapping method" described in [Table 3] to map and encode the radicals. Since these similar letters and punctuation marks are not used in a cluster with other radicals, they have a higher blind typing rate.

[0118] 8. Appropriate splitting of Chinese characters can improve blind typing: Splitting Chinese characters completely into 4 to 5 radicals can result in a higher blind typing rate. For example, if “挑” is split into “扌佥” and the code is {sq}, type {sq}, and “挑” appears at the 20th position on the character selection screen; split “挑” into “扌人一” and the code is {sq}. " and get {srhx}. Type {srhx}, and "挑" appears in the first position, then you can type blindly.

[0119] 9. Use it or lose it: The more times a user uses a certain "pure character code" to input a Chinese character, the greater the chance that the user will be able to blindly type the character using the "pure character code" (see Section 2, Point 9 (ix) for details).

[0120] Section 2: Based on the "single Chinese character input method", create a "pure character code" system that adapts to the public's cognitive rules

[0121] There are many users of shape code input methods, and different users have different subjective perceptions of issues related to Chinese character decomposition encoding:

[0122] A) Different preferences for the separation of the internal structure of Chinese characters: For example, for the character "戊", different users may adopt different splitting methods such as "厂戈", "厂乚丿丶", etc. based on different preferences for the separation of the root characters; for the character "咸", different users may adopt different splitting methods such as "戌口", "戊口口", etc. based on different preferences for the separation of the root characters;

[0123] B) Different levels of recognition and usage preferences for character roots: For example, based on different literacy levels and usage preferences, different users can use a variety of different splitting methods such as "艹濮", "氵蒪", "氵浦寸", "氵艹尃", "氵艹甫寸" and so on;

[0124] C) Different preferences for writing strokes: For example, the standard stroke order of “力” is “丿”, but some people use the stroke order of “丿”; the standard stroke order of “女” is “折丿一”, but some people use the stroke order of “折一丿”, “一折丿” or “一丿折”;

[0125] D) Some users prefer to use letters that are similar to certain radicals to map radicals (especially those who have learned the "graph code input method" and "character group input method"): for example, "口" and "囗" are similar to the letter "o"; "丷" is similar to "v"; "凵冂匚]" is similar to "u"; "冂" is similar to "n"; "讠辶" is similar to "i"; "彐" is similar to "E"; "乂" is similar to "x"; "丅" is similar to "T", etc.;

[0126] E) Different pronunciation preferences for radicals: Many radicals have multiple pronunciations, such as: 艹{cao (grass), ao}; 阝{er (ear side), fu (fu)}; 疒{bing (disease), ne}; 冖{gai (cover head), mi (power), tu (bald treasure cover)}; 犭{quan (reverse dog), fan (reverse)}; 乡{shan (shirt right side, three strokes), xian}; 夂{zhong (end), do ng (the radical of winter)}; 攵{pu(扑), fan(反文), wen(文)}...etc.; different users may adopt different pronunciations for the same polyphonic root; for example, people in the south pronounce the pinyin initial consonant "n" as "l" (pronounce "女" as "lv"), and the pinyin initial consonant of the character beginning with "hu" as "f" (people in Hunan tend to pronounce "虎" as "fu"), or as "w" (people in Guangdong tend to pronounce "胡" as "wu").

[0127] F) Other cognitive biases in writing structures that are difficult to automatically process by computers: For example, “敖” can be split into “丿” 攵”, but some users split it into “キ万攵”; “熬” can be split into “敖灬”, “丿” There are similar situations in the “ ” or “ ” of “ ”; “ ” of “ ” and “ ”.

[0128] To sum up, due to reasons like these, there are 100,000 different preferences for the way of splitting characters in the eyes of 100,000 shape code users. Only the input method that fully respects and adapts to the actual situation of the mainstream public's cognition is the most suitable input method for the public. If there is a shape code input method that can be used without learning, which can be used easily for typing and can use simple rules to achieve high-speed blind typing of Chinese characters, who would be willing to use the pinyin input method with lengthy encoding and a high rate of repeated code flipping and character selection?

[0129] To this end, the present invention has created the following multiple "character splitting scheme" databases with comprehensive compatibility:

[0130] 1. Create a "theoretical large root decomposition scheme database" (scheme library ①) that is suitable for the theoretical largest root

[0131] (I) First, we established a database of more than 20,000 Chinese characters and their stroke order in the GBK character library (hereinafter referred to as the "GBK Chinese character stroke order database"), and used various related materials and dictionary tools combined with manual review to repeatedly cross-check and revise the stroke order data of each Chinese character;

[0132] (ii) Based on the "GBK Chinese Character Stroke Order Database", a computer algorithm is used to search for a "potential radical list" of radicals, radicals, and strokes that may be contained in each Chinese character (arranged from most to least in terms of the number of strokes contained in the radical);

[0133] (3), because the same stroke order of different radicals has a lot of "same strokes but different structures" phenomenon, for example, in terms of stroke order: 湇=湆、汢=江、沞=沛、泪=郁、捏=泹、汎=汷、宂=它、冫=亠、文=六、=关、粙=畨、籹=娄、炎=炏、之=辶……, the computer can not distinguish these objects only by stroke order information, and a large amount of stroke order data of more than 20,000 GBK Chinese characters from public information channels such as the Internet also inevitably have some errors, so the inventor can only adopt the method combining computer database cross-checking and manual checking, The above-mentioned "potential radical list" was repeatedly reviewed to remove the mismatched radicals of "same strokes but different structures" that did not conform to the actual situation of Chinese characters, and solved some Chinese character splitting problems that could not be handled by computer technology alone, such as: "命" should be split into "亼敲" instead of "合卩"; "是" should be split into "日丅人在個人" instead of "日工工" ("丅+一"#"工" in "是"); "在" should be split into "丨土" instead of "廾土" (ナ+丨#廾); "得" should be split into "彳日于丶" instead of "彳旴丶" (日+于#旴); "发" should be split into " 丿又丶" instead of " 丿叉" (also +, #叉);

[0134] (IV) The present invention also supplements some radicals that are not included in the GBK character library (they are not Chinese characters in the GBK character library, such as: =niao niaowu灬; also =xing xing kai san jiu ... =ba water right eight / right open eight; =gou 冓 has no Ran; Li =li 婯 has no woman; =xue has no children; =man滿无氵……), only the theoretical largest radical that meets the actual situation of each Chinese character and conforms to the "simple character decomposition method" is retained, and finally the "simple character decomposition scheme" composed of the theoretical largest radicals of each GBK Chinese character is obtained, which is called the "theoretical large root character decomposition scheme" and saved in the "theoretical large root character decomposition scheme database" (scheme library ①).

[0135] 2. Create a "popular root character solution database" (solution database ②) that is suitable for the mainstream public's level of cognition

[0136] The aforementioned "theoretical big root character splitting scheme" is based on selecting the largest character root in theory each time the character is split, and the premise is that the user must recognize all the largest character roots in theory inside the Chinese character, but in fact very few people can fully reach this cognitive level, even the inventors, there is no guarantee that the "theoretical big root character splitting scheme" can be quickly and accurately given to each Chinese character during the typing process. In order to select a "simple character splitting scheme" (called "popular big root character splitting scheme") that is easily recognized by the mainstream public for each GBK Chinese character, so that ordinary users can more easily obtain easy-to-understand and better character splitting schemes, the present invention follows the following three principles of the general cognitive law of the user to select the character roots that the "popular big root character splitting scheme" is applicable to:

[0137] (I) The principle of giving priority to easily recognizable large roots: In the "popular large root decomposition scheme", priority is given to splitting Chinese characters into roots that are easier to recognize and have more strokes, and avoiding selecting uncommon roots or roots with fewer strokes.

[0138] For example, "夸" is split into "大损" instead of "夳". ” (Because the root obtained by the latter method is relatively rare and difficult to be recognized by the public, it is not as easy to recognize as the former method), nor is it "大二" ” (Because “Sophomore " is easy to recognize, but the first two radicals "大" and "二" have fewer strokes than "损"); "骗" is split into "马扁" instead of "驴冊" (because the radical "冊" obtained by the latter splitting method is relatively rare and not easily recognized by the public, so it is not as easy to recognize as the former splitting method); "查" is split into "木旦" instead of "杳一" (because the radical "杳" obtained by the latter splitting method is less used than the two radicals of "木旦", so it is not as easy to recognize as the former splitting method); and so on: "棚" is recommended to be split into "木朋" instead of "枂月"; "脚" is recommended to be split into "月却" instead of "胠卩"; "奔" is recommended to be split into "大卉" instead of "夲廾"; "嫩" is recommended to be split into "女委" instead of "娕攵"; "轰" is recommended to be split into "口孔" instead of "吇乚".

[0139] (II) Principle of decomposing and familiarizing uncommon radicals: Decompose the uncommon radicals in the theoretical big root decomposition scheme into more common radicals.

[0140] If the "theoretical root decomposition scheme" of common Chinese characters contains low-frequency and uncommon root characters that are simply composed of root characters that are more familiar to the public, such as but not limited to: "亽亼戓咅胠徍夲兟吅氿玨尗叒夃吂纩扏圤娕玟二茾吇屵枞叻厎厉殸陃唭嶏陏蛄啫淯氶团围敕棥畐暃壴啚卂丵厷弚尶巸豙夗亶四髟叚獸戉氰夅冘迿犮夋缺白冋曷苋憡乑鬲臽荅夌戀瀒洴洿卬厃乊 ..." and so on. Since these radicals are only composed of common and easily recognizable radicals, in order to reduce the difficulty of character decomposition for general users, the present invention will break down these radicals in the process of character decomposition and root extraction of the "popular large-root character decomposition scheme", and decompose them into common large radicals that are more familiar to the mainstream public.

[0141] For example, professional writers may know the character "亽", but most ordinary people do not know it. Therefore, it is easy to have differences in the way of splitting Chinese characters such as "今" and "令" that contain the character root "亽". Therefore, according to the "theoretical big root splitting scheme", the character "今" should be split into "亽". ”, splitting the character “令” into “亽”; but according to the mainstream cognitive preference of ordinary users, the character “今” is usually split into “人丶 ”, and split the character “令” into “人丶”; and so on: theoretically, split the character “苦” into “吅犬”, and commonly split the character “苦” into “口口犬”; theoretically, split the character “琴” into “玨今”, and commonly split the character “琴” into “王王今”; theoretically, split the character “合” into “亼口”, and commonly split the character “合” into “人口”; theoretically, split the character “俞” into “亼刖”, and commonly split the character “虞” into “人一月刂”, and so on.

[0142] (III) Principle of high degree of integration

[0143] 1. "Discrete composite radical" means a larger radical composed of no less than two separate smaller radicals;

[0144] 2. Structural classification of “discrete combined radicals”

[0145] ① Fully surrounded structure: at least one separate radical is completely surrounded by another radical, such as: 国回困囚;

[0146] ② Semi-enclosed structure: at least one discrete radical is incompletely surrounded by another radical, such as: 区岗唐风凤向冋乔;

[0147] ③ Up-down structure: no less than two separate radicals are arranged vertically, such as: 字李圭亖;

[0148] ④ Left-right structure: no less than two separate radicals are arranged horizontally, such as: 打慕川;

[0149] ⑤Mixed structure: no less than three separate parts present no less than two of the listed structural characteristics, such as: enamel and medicine.

[0150] 3. Grading of the degree of integration of “discrete combined radicals” that adapt to the public’s cognitive rules

[0151] Ordinary people are most likely to classify the fully enclosed structure as an independent Chinese character, followed by the semi-enclosed structure, then the mixed structure, then the up-and-down structure, and finally the left-right structure (because the written text of modern Chinese generally adopts a horizontal extension from left to right, so people are generally accustomed to preferentially recognize the two separate writing structures arranged left and right as two independent Chinese characters, and are accustomed to preferentially recognize the two adjacent separate writing structures as a whole Chinese character). From this, we can get the ranking of the degree of combination of different root combinations: fully enclosed structure > semi-enclosed structure > mixed structure > up-and-down structure > left-and-right structure.

[0152] 4. High binding priority principle

[0153] In the "Popular Large Root Character Decomposition Scheme", without violating the above-mentioned Principles 1 and 2, when a separate smaller radical can be simultaneously incorporated into two different larger radicals before and after the stroke order, it will be preferentially incorporated into the larger radical with a higher degree of integration.

[0154] For example: "Tang" is split into "扌唐" instead of "宽肀口" ("广" in "唐" is a semi-enclosed structure, and its degree of integration is higher than the left and right structures of "广" in "宽"); "塷" is split into "土鹵" instead of "坫乂丶丶丶丶" (because "占" in "鹵" is a fully enclosed structure, and its degree of integration is higher than the left and right structures of "占" in "坫"); "骗" is split into "马扁" instead of "驴冊" ("户" in "扁" is a semi-enclosed structure, and its degree of integration is higher than the left and right structures of "户" in "驴"); "簥" is split into "喬" instead of "笑口冋" ("夭" in "喬" is a semi-enclosed structure, and its degree of integration is higher than the upper and lower structures of "夭" in "笑"); "濓" is split into "氵亷" instead of "浐彐 八” (“產” in “亷” is a semi-enclosed structure, and its degree of integration is higher than the left-right structure of “產” in “浐”); “游” is split into “氵斿” instead of “汸子” (“方” in “斿” is a compound structure, and its degree of integration is higher than the left-right structure of “方” in “汸”); “籝” is split into “盈” instead of “笀口肗凡” (“亡” in “赢” is a compound structure, and its degree of integration is higher than the upper and lower structure of “亡” in “笀”); “蕘” is split into “艹"堯" instead of being split into "芏土土兀" ("土" in "堯" is a compound structure, and its degree of combination is higher than the upper and lower structures of "土" in "芏"); "药" is split into "艹乐" instead of being split into "苩姚尧木" ("白" in "乐" is a compound structure, and its degree of combination is higher than the upper and lower structures of "白" in "苩"); "崖" is split into "山厓" instead of being split into "屵圭" ("厂" in "厓" is a semi-enclosed structure, and its degree of combination is higher than the upper and lower structures of "厂" in "屵").

[0155] The above three principles are only used by the inventor to select the preferred decomposition method among different decomposition methods of the same Chinese character.Before using the present invention to input Chinese, the user does not need to understand or learn these principles at all , you can type the desired Chinese characters by using your own decomposition and mapping coding methods; however, the "pure character codes" corresponding to the "simple decomposition scheme" ("Scheme Library ②") that meet the above three principles are arranged in the first echelon of the codes in the total code table. Users who use these codes can get priority in blind typing opportunities. The user can learn relevant "techniques" to obtain higher typing efficiency For the convenience of description, this paper refers to the database containing all GBK Chinese characters' "popular big root character decomposition schemes" as the "popular big root character decomposition scheme database" (scheme library ②).

[0156] In fact, among the 21,003 GBK Chinese characters, only the "theoretical big root decomposition scheme" and the "popular big root decomposition scheme" of 1,986 Chinese characters are different, and the two decomposition schemes of the remaining Chinese characters are the same;.

[0157] 3. Create a "personalized root selection and character decomposition solution database" (solution database ③) to adapt to the public's differentiated cognitive level

[0158] A large number of actual tests of the present invention have found that different users have different levels of cognition and preferences for character roots, and the specific "largest character root" to be taken out of the same Chinese character often varies from person to person. The specific implementation methods of the "simple character decomposition scheme" for the same Chinese character are often diverse. For example, different users may adopt a variety of different methods for the character "薄", such as "蒲寸、艹普、艹浦寸、艹氵尃、艹氵甫寸".

[0159] In order to enable different users with different cognitive levels to successfully complete the character splitting operation based on their existing cognition, in addition to creating the aforementioned "theoretical large root character splitting scheme" and "popular large root character splitting scheme", the present invention also starts from these two schemes to perform multi-layer decomposition, recursive and iterative splitting and refinement processing on the "character root string" of each Chinese character. The specific method is as follows:

[0160] (I) Select the easily ambiguous radicals that need to be broken down

[0161] According to the cognitive level of the general public, the present invention divides the radicals into three categories and performs different treatments on them respectively:

[0162] 1. Single-stroke radicals (no need to break them down): These are basic common sense that are well known to the general public and do not need to be broken down and cannot be broken down.

[0163] 2. Common sense radicals (no need to be broken down): very common radicals in life such as "釒口钅牜..." and so on; very common Chinese characters in life such as "口木土虫女月日山火王石目人禾米心又八十七田广力几皿巾耳白丫大方鱼革立雨鸟舟午丰小厂子工弓..."; they are widely known and not easily misidentified by ordinary people, and have no cognitive ambiguity, so the present invention does not break down this type of common sense radicals to avoid adding unnecessary redundant encoding and repeated codes.

[0164] 3. Easy-to-ambiguous character roots (need to be broken down): Easy-to-ambiguous character roots are character roots that are easy to cause cognitive ambiguity in people. It is necessary to create a compatible breaking method for them. Easy-to-ambiguous character roots include the following two categories:

[0165] 1) Roots that are rarely used directly in daily life: "勹" and other radicals, "乂隹镸亼帀巛冎见" and other characters in Chinese characters. These characters are rarely used directly in daily life by ordinary people and are not easy to be recognized accurately. It is necessary to break them down so that different users can "use them without learning";

[0166] 2) Radicals composed of distinct common sense radicals: Some characters in words that are often used directly in daily life are easily split into more detailed radicals by users based on their first impression because they are composed of distinct common sense radicals, such as: 全恒酉穴貝欠羽 从旦比住合示貝... Such radicals also need to create a compatible splitting method for splitting;

[0167] In summary, the present invention determines whether a radical needs to be broken down based on factors such as the recognition popularity of the radical and the internal structure of the radical, and marks it in the radical table (see Section 3, Point 1 [Table 1] and its standards and instructions for details).

[0168] (II) Formulate a plan for splitting characters with easily ambiguous roots

[0169] Formulate various main decomposition schemes for each easily ambiguous root in order to perform the following step (3). For example, the main "decomposition scheme" is listed in the brackets after each easily ambiguous root. Some roots can have multiple decomposition methods. Different decomposition schemes for the same root are separated by commas, and different roots are separated by semicolons: 艹(十丨);冂(| );亻(丿丨);丵(丶嫫,丶丨一);亠(丶一);扌(Eleven);丷(丶丿);嫫(丶一);(丿 );(丿一);宀(丶冖,丶丶 );冖(丶 );糹(壹丶丶丶);忄(丶丶丨,丶丨丶);訁(亠二口);彐( 2,] 1); 1 (1); 2 (2); 丨); 辶(丶 丶); fish (砜田灬); 厶 ( 丶); 疒(广冫, 冫广); 匚(一 );貝(目八);灬(丶丶丶);]( 一); bird (亻彐一 , 灬);乂(丿丶); (丨丨); 衤(礻丿,丶 丨 ); car (亘丨, one for one, ten for one day); 幹(丿 );隹(住一,亻丶二土);…and so on.

[0170] (III) Multi-level recursive and iterative processing of radicals

[0171] 1. Take the "radical string" of each Chinese character in "Solution Library ①" and "Solution Library ②" as the "parent radical string", and follow the second step below to split and modify it into the "child radical string";

[0172] 2. According to the "simple character splitting method", among the no more than 4 radicals in the front of the "parent radical string" of the Chinese character, the "easy radical" with the largest number of strokes is split according to its "character splitting scheme", and the "radical string" consisting of no more than 4 radicals in the front of the new radical combination obtained is called the "child radical string", and each "child radical string" is stored together with the Chinese character in the "scheme library ③" (if there is more than one easy radical with the largest number of strokes, they are split and processed independently, and a plurality of corresponding different child radical strings are obtained, which are stored in the "scheme library ③" respectively);

[0173] 3. Substitute the "offspring radical string" obtained in the aforementioned step 2 into the aforementioned step 2 as the "parent radical string" of the new generation, obtain the "offspring radical string" of the next generation and store it together with the Chinese character in the "scheme library ③";

[0174] 4. Repeat steps 2 and 3 for the same Chinese character recursively until no "easy-to-distinguish radical" is found in the first four radicals of the last "descendant radical string". Thus, all radical combinations of the Chinese character that meet the "simple character decomposition method" rules can be obtained, and all decomposition results of each generation have been stored in the "solution library ③";

[0175] 5. If a Chinese character in "Scheme Library ①" and "Scheme Library ②" has multiple different "Simple Character Decomposition Schemes", each of its "Simple Character Decomposition Schemes" will be processed from steps 1 to 4 to obtain various possible character decomposition schemes and store them in "Scheme Library ③".

[0176] 4. Manually create a "stroke order error-tolerant character splitting solution database" (solution library ④) that is compatible with stroke order deviations

[0177] Through a large number of actual tests, it is found that the public's understanding of the stroke order of Chinese characters varies greatly. For example, the standard stroke order of "力" should be "丿", but some people split it into "丿"; the standard stroke order of "女" should be " 丿一", but some people split it into " "One" or "One "丿" or "一丿 "; "爨" should be split into "亻二同彐", but some people split it into "同亻二彐"; "夒" should be split into "一自止巳", but some people split it into "一止自巳"; "兠" should be split into "白北儿", but some people split it into "丨二白匕"; "兜" should be split into "白丿 ]", but some people split it into "丿 Bai]"; "燕" should be split into "二十口北灬", but some people split it into "二十丨二口"... and so on; for this reason, each GBK Chinese character is manually reviewed, and the GBK Chinese characters that are prone to stroke ambiguity similar to the above are created one by one with compatible splitting schemes corresponding to their common non-standard stroke orders, and the results are stored in the "Scheme Library④".

[0178] 5. Create a "database of compatible character splitting schemes for character splitting with similar character radicals" (Scheme Library ⑤)

[0179] Through a large number of actual tests, we found that some Chinese characters have some writing structures that are similar to some English letters. Some users (such as old users of "Character Code" and "Character Combination") like to use some English letters to represent the similar structures in Chinese characters, such as splitting "馬" into "F十" and "F十". "灬"; split "镸" into "F云"; split "仺" into "人E", and use the letter "o" to map "口" or "囗". In order to give full play to the image thinking advantage of this type of users, each GBK Chinese character is analyzed one by one through manual methods. For GBK Chinese characters with similar radicals and letters, compatible character splitting schemes are created one by one, which use additional similar letters to map corresponding radicals (for specific mapping methods, please refer to the second point of Section 3 [Table 2]), and the results are stored in the "Scheme Library⑤".

[0180] 6. Create a "database of other compatible word splitting schemes" that are compatible with other types of cognitive biases (Solution Library ⑥)

[0181] Through a large number of actual tests, it is found that the splitting process of some Chinese characters is prone to disagreement. For example, according to the standard recommendation, "敖" should be split into "丿" 攵", but some users split it into "キ万攵"; "熬" can also be split into "敖灬", "丿" 攵" or "キ万攵灬", other Chinese characters containing "敖" such as "傲殖敖遨嗷骜箍獒聱廒鼇鷔鰲驁隞贅謸蔜磝璈獓滶摮慠嶅嫯嗸厫鏊謷" and so on also have similar situations. In addition, there are many other Chinese characters that are prone to disagreements in character splitting. These special splitting methods cannot be easily solved by computer algorithms. The present invention reviews each GBK Chinese character one by one through a manual method. For this type of GBK Chinese characters, 2974 other additional compatible personalized character splitting schemes are created one by one, and the results are stored in the "Scheme Library ⑥". For example: 是(日下人、日F人、旦 people); Not (one person, one person, one); in (one person, one official, earth); this (辶文); above (丨二, 丄一, 一丄); middle (冂十); big (one person, one person); for (力丶丶), etc.

[0182] 7. Combined decomposition of the first four radicals to create a "radical combination decomposition plan" (plan library ⑦)

[0183] Users may recognize some radicals with more strokes, but not some radicals with fewer strokes. Therefore, the compatible decomposition scheme of "Scheme Library ③" which only decomposes the radicals with the most strokes cannot fully cover the diverse cognitive differences of users on radicals. For this reason, the "Radical Combination Decomposition Scheme" is further supplemented. The steps are as follows:

[0184] (I) Based on the root set (see Section 3, Point 1 [Table 1] and its explanation), create a root data table, select and mark the ambiguous roots that need to be split and their detailed splitting plans (see Section 2, Point 3, (I) and (II)).

[0185] (ii) Take the first four radicals of each Chinese character in "scheme library ①, ②, ③, ④, ⑤, ⑥" (if there are less than 4 radicals, take all radicals) as the first "parent radical string", and disassemble and modify it into a "child radical string" according to the following step (iii);

[0186] (III) According to the ambiguous radicals and their splitting methods marked in the above-mentioned point (I), the split fine radicals in the first four radicals in the "parent radical string" of each Chinese character are subjected to all possible combinations of non-split and splitting in various ways through a four-layer nested loop algorithm, thereby obtaining a series of new radical combinations, of which only the first four radicals are taken, which are called "child radical strings", and each "child radical string" is stored together with the Chinese character in the "solution library ⑦";

[0187] *Note: The differences between "Scheme Library ③" in the third point and "Scheme Library ⑦" in the seventh point of this section are as follows: The former only breaks down the most stroke-rich ambiguous root and recursively processes it repeatedly (excluding the method of not breaking down ambiguous roots), which is a straightforward unfolding process; the latter is a combination of different breakdown methods of each ambiguous root (including the method of not breaking down), which is a combination of various possible breakdown methods of different roots; the duplicate results of the former and the latter will be deleted in the subsequent steps.

[0188] VIII. Create a "Supplementary Root Character Decomposition Scheme Database" (Scheme Library ⑧) for the character decomposition schemes of Chinese characters with less than 4 roots

[0189] Read, copy, and store all Chinese characters with less than 4 roots and their character decomposition schemes in the above "Scheme Libraries ①, ②, ③, ④, ⑤, ⑥, ⑦" into "Scheme Library ⑧". Using database technology, according to the "Supplementary Root Character Decomposition Method" (see (2) in the second point of Section 1), all root strings in "Scheme Library ⑧" are supplemented to root strings of 4 roots. If there are more than one way to break down a related sub-root into grand-children roots, then implement multiple "Supplementary Root Character Decomposition Method" methods corresponding to the multiple breakdown methods of this root. Thus, more than 64,000 "Supplementary Root Character Decomposition Schemes" are generated and stored in "Scheme Library ⑧".

[0190] IX. Create a total database of single-character codes adapted to the alphabetic punctuation keyboard ("Total Alphabetic Character Code Library")

[0191] (1). Create a "Total Character Decomposition Scheme Library" based on all GBK Chinese characters

[0192] Merge the above "Scheme Library ②, Scheme Library ①, ③, ④, ⑤, ⑥, ⑦, ⑧" (put "Scheme Library ②" as the "preferred character decomposition scheme" at the front), and after deleting duplicate character decomposition schemes, the resulting character decomposition schemes form the "Total Character Decomposition Scheme Library".

[0193] (2). Create a "First Root Sound Character Code Library" (Character Code Library A) based on the mapping encoding of the first pronunciation of roots

[0194] According to the first pronunciation of each root in [Table 1] and the "Root Mapping Encoding Rule", convert each root in the "Total Character Decomposition Scheme Library" into the initial letter of the corresponding pinyin of its first pronunciation, and then the "Character Code Library A" is generated.

[0195] (3). Create a "Secondary Root Sound Character Code Library" (Character Code Library B) based on the mapping encoding of other pronunciations of roots

[0196] Some radicals in [Table 1] have more than one pronunciation. Since different users have different pronunciation preferences for the radicals, the mapping encoding methods for the same polyphonic radicals will vary from user to user. Therefore, according to the pronunciations other than the first pronunciation of the polyphonic radicals in [Table 1] and the "radical mapping encoding rules", the polyphonic radicals in the "total library of character splitting schemes" are converted into the first pinyin letters corresponding to the pronunciations, thus generating the "character code library B" based on the mapping of non-first pronunciations.

[0197] (IV) Creating a "Character Code Library" (Character Code Library C) based on the mapping encoding of character root-shaped letters

[0198] Some radicals can be assigned to key letters by additional clues similar to letters, for example: the stroke "折" is similar to the letter "v" (v = fold); the stroke "竖" (丨) is similar to the letter I (i = vertical); "口" and "囗" are both similar to the letter "o" ("口" and "囗" = kou, o); "匕" is similar to "七" (compatible with each other); "丷" is similar to "v"; "凵冂匚]" is similar to "u"; "讠辶" is similar to "i"; "彐" is similar to "E"; "乂" is similar to "x"; "丅" is similar to "t"; this mapping method can improve the utilization rate of low-frequency letters such as "iovu", and can significantly increase the probability of Chinese characters obtaining blind typing codes, and create additional letter-like mapping methods for some radicals that are clustered on high-frequency keys; according to the mapping relationship between radicals and "radical-like letters" in [Table 2], the radicals in the "total library of character decomposition schemes" are converted into corresponding "radical-like letters", and the "character code library C" is generated.

[0199] (V) Create a "punctuation letter code library" (code library D) based on the root punctuation mapping code

[0200] {,. / '[} and other keys are adjacent to {jkl;} and other right-hand basic finger keys in the keyboard, but these punctuation keys are not involved in the input of Chinese in the "traditional shape code", which is a huge waste of key resources; the present invention uses these keys as additional shortcut mapping keys for 6 high-frequency radicals: {,} = 木, {.} = 月, { / } = 扌, {'} = 氵, {[} = 釒, 钅. This does not affect the normal input of these 5 punctuation marks, and they can be directly entered by typing in the English input state; in the Chinese input state, press these punctuation keys, and the corresponding full-width symbols and digital serial numbers will appear on the character selection screen. Press the serial number to enter the corresponding full-width punctuation mark. If you press the Enter key after pressing the punctuation key, you can enter the half-width punctuation mark corresponding to the key position;

[0201] According to the mapping relationship between radicals and "punctuation marks" in [Table 3], the character decomposition schemes containing relevant radicals in the "character decomposition scheme library" are mapped and encoded into corresponding "punctuation marks" to generate the "character code library D".

[0202] (VI) Create a "Nanfang Root Sound Code Library" (Code Library E) based on the mapping code of the root "Nanyin First Letter"

[0203] According to the method described in [Table 4], the relevant radicals in the "total library of character splitting schemes" are mapped to the corresponding "Nanyin initial letters", and the "character code library E" is generated. When a radical has more than one Nanyin initial letter, the present invention creates a compatible "pure form character code" for each Nanyin initial letter corresponding to the radical for the Chinese character containing the radical; when the character splitting scheme of a Chinese character contains more than one radical with Nanyin initial letter, the present invention simultaneously creates a combination compatible code of each standard pinyin initial letter and each Nanyin initial letter corresponding to each radical of the Chinese character; the user should choose whether to use the "Nanyin initial letter" of some or all of the relevant radicals of the Chinese character for mapping code according to his or her own cognitive level and preference. For example, the character "胡" is pronounced fu in Xiangyin and wu in Cantonese; the present invention provides the first root code (Cao Cao + Hu Hu) = {ch} for "胡", and provides (Cao Cao + Hu Fu) = {cf} compatible with Xiangyin and (Cao Cao + Hu Wu) = {cw} compatible with Cantonese.

[0204] (VII) Create a “letter code library”

[0205] By merging the five character code libraries "Character Code Library A, Character Code Library B, Character Code Library C, Character Code Library D, and Character Code Library E" ("Character Code Library E" is placed at the end as an optional character code library, only for users with southern accents), and deleting duplicate Chinese character codes, it becomes a total database of single-character letter codes ("letter character code total library").

[0206] (VIII) Initial Priority Classification and Sorting Method of “Pure Font Code”

[0207] 1. The echelon division of "pure character codes" and their initial priority classification and sorting method: the "pure character codes" using the "popular character root" decomposition scheme (scheme library ②) and the "first pronunciation" of the character root (see [Table 1] for description) mapping encoding are "priority character codes", and the "pure character codes" using other decomposition encoding methods are "compatible character codes"; each GBK Chinese character has and has only one "priority character code", and can have multiple "compatible character codes"; all "pure character codes" are divided into three echelons, and the "priority character codes" are placed at the front of the Chinese character code table as the first echelon; the "compatible character codes" (character code library E) using the "Nanyin initials" encoding are placed at the end of the Chinese character code table as the optional third echelon; the remaining codes are placed between the first and third echelons as the second echelon;

[0208] 2. The sorting method of "pure character codes" within each coding echelon

[0209] 1) Calculation formula for the recognition priority of radicals: R = SIGN(INT(0.7 * F))

[0210] Explanation of the formula (applicable throughout the text): ① R = radical recognition priority; ② SIGN(X) = sign function (returns -1, 0, 1 when X < 0, X = 0, X > 0 respectively); ③ INT() = integer function; ④ F = character frequency, where "character frequency" refers to the number of times the Chinese character appears per 100,000 words in ordinary text; ⑤ 0.7 is the first-level character frequency coefficient = 1 / 1.43, and the lowest character frequency of the top 3500 Chinese characters in terms of character frequency is 1.43, so a character with a character frequency lower than 1.43 is regarded as a low-recognition radical; ⑥ Value range of R: R = 1 when F >= 1.43, and R = 0 when F < 1.43.

[0211] 2) Initial priority ranking method for the encodings of different character decomposition and coding methods of the same Chinese character within the same character code echelon

[0212] Based on the "initial echelon ranking" in point 1 above, when a Chinese character has multiple character decomposition schemes and contains "multi-mapping radicals", the rankings of its various "pure shape character codes" within the same coding echelon are arranged in descending order according to the following "U" value:

[0213] U = (R1 * B1 * (10^8)) + (R2 * B2 * (10^6)) + (R3 * B3 * (10^4)) + M1 * 100 + M2 * 10 + M3

[0214] Explanation of the formula (applicable throughout the text): ① U is the "priority of the character decomposition and coding scheme"; ② R1 / R2 / R3 are the recognition priorities of the 1st / 2nd / 3rd radicals in the radical string respectively, B1 / B2 / B3 are the stroke counts of the 1st / 2nd / 3rd radicals in the radical string respectively, ⑤ M1 / M2 / M3 are the "priority of the mapping implementation method (M)" of the 1st / 2nd / 3rd radicals in the radical string respectively, ⑧ Method for obtaining the value of the "priority of the mapping implementation method (M)" of the radical: M = 6 when the radical is assigned to the first letter key of the pronunciation, M = 5 when assigned to the second pronunciation key, M = 4 when assigned to the third pronunciation key, M = 3 when assigned to the fourth pronunciation key, M = 2 when assigned to the key of the similar-shaped letter, M = 1 when assigned to the punctuation key, M = 0 when assigned to the first letter key of the southern pronunciation; ⑨ "+" is the plus sign, "*" is the multiplication sign, "^" is the exponentiation sign (e.g., 10^4 = 10000); ⑩ When there is no nth radical in the radical string, Rn = 0, Bn = 0, Mn = 0.

[0215] 3) Initial priority ranking method for homographs: Based on the two initial rankings described in points 1 and 2) above, Chinese characters and their "pure form character codes" are sorted in the coding echelon according to the composite index of "C + STR((10^6) - INT(0.5 + 100 * character frequency))" (where "C" is the "pure form character code" string, "-" is the minus sign, "^" is the exponentiation sign, and "STR()" is the numerical conversion to character function, which returns the corresponding Arabic numeral characters for the input numerical value). When a "pure form character code" corresponding to multiple Chinese characters is typed on the root keyboard, the homograph Chinese characters with higher character frequencies in the same character code echelon are presented in the front positions on the character selection screen.

[0216] (IX). Dynamic ranking method of "word and character coding"

[0217] Based on the above three initial rankings, as the user types in the "word and character coding" and selects the corresponding words and characters, the rankings of the words and characters that have been input on the character selection screen will gradually move closer to the position of the first place in the candidate word and character list. If the ranking of a certain word and character in the candidate word and character list is N when the user uses a certain "word and character coding" to input that word and character this time, then when the user types in that "word and character coding" next time, the position of that word and character in the candidate word and character list will become INT(1 + K ** N); K is the movement coefficient, and the initial value of K is 0.5. The user can set the value of K between 0 and 1 according to their own preferences. The smaller the value of K, the faster the forward movement.

[0218] For example: When K = 0.5 is set, continuously type in {sq} multiple times and select "pick up". When {sq} is typed in for the first time, "pick up" appears in the 20th place, and the screen is flipped to select the character "pick up" for input; when {sq} is typed in for the second time, "pick up" appears in the 10th place, and 0 is typed in to input "pick up"; when the third operation is performed, "pick up" appears in the 5th place, when the fourth operation is performed, "pick up" is in the 3rd place, when the fifth operation is performed, "pick up" is in the 2nd place, when the sixth operation is performed, "pick up" is in the 1st place. Starting from the sixth time, every time {sq} is typed in, "pick up" appears in the 1st place and can be input blindly by pressing the space bar. And so on. The more times the user uses a certain coding to input a word and character, the more forward the rankings of that word and character and the coding will move. Eventually, the words and characters and their codings with the highest user usage rates can all achieve "first place blind typing".

[0219] X. Create a total digital character code database ("total digital character code library") adapted to the numeric keypad

[0220] According to the "root → letter punctuation → numeric key mapping method" (see item 4 in point (II) of section I), all Chinese characters and the root strings of their character splitting schemes in the "total letter character code library" (see item (VII) of section II) are mapped and converted into digital "pure form character codes" (referred to as "digital character codes" throughout the text), and the "total digital character code library" is obtained.

[0221] Section 3: Creating a "root set" that adapts to the public's cognitive resources and cognitive rules

[0222] "Character root collection" of the present invention is formed by summarizing and arranging 3196 kinds of character roots contained in the "Character root scheme" of each Chinese character in the "Character root scheme total library" (see Section 2, No. 9, (a) for details). The establishment of the two is not accomplished overnight, but is the result of complementing each other and finally forming after repeated mutual checking and correction, as shown in the following tables:

[0223] 1. The root table of the present invention is shown in [Table 1]:

[0224] [Table 1]: 3196 kinds of radicals, the classification and identification of radicals, and the method of assigning radicals to letter keys are described in the table below

[0225]

[0226]

[0227]

[0228]

[0229]

[0230]

[0231]

[0232]

[0233]

[0234]

[0235] [Table 1] Description:

[0236] 1) The root characters are arranged in a table frame of the corresponding letter key area according to the pinyin initials of their first pronunciations ("first pronunciation" refers to the only pronunciation of the root character or the most popular pronunciation of the root character in society), and different root characters are separated by semicolons;

[0237] 2) Each root is followed by a "[]" to indicate its pinyin. Different pinyins of polyphonic roots are separated by commas. Some pinyins are annotated with parentheses, such as "卬[ang(昂字底),yang(仰)]" in area a, which means that "卬" can be read as ang(昂字底) or yang(仰);

[0238] 3) Each root can be assigned to the corresponding letter key according to the pinyin initials of its first pronunciation, and polyphonic roots can also be assigned to the corresponding letter keys using the pinyin initials of other pronunciations listed in the square brackets behind them;

[0239] 4) The first pronunciation of some high-frequency polyphonic roots is well known and commonly used, while its other pronunciations are rarely used. For such roots, only its first pronunciation is used as the pinyin mapping clue for its assignment to the letter key position, and the pinyin abandoned as the mapping clue is marked with "(abandoned...)", for example, "柱[zhui, (abandoned chui)]" in the z key area indicates that the root "柱" only uses its first pronunciation "zhui" as the clue for its assignment to the letter key position, and abandons "chui" as the mapping clue;

[0240] 5) The radicals marked with an asterisk are the "basic radicals" commonly used by ordinary users. When users extract "sub-radicals" from Chinese characters (see Section 1, Point 2 (a) "Simple Method of Decomposing Characters"), it is recommended that users remove these "basic radicals" as a whole, and it is not recommended that users remove these "basic radicals" into more detailed radicals. For example, the "*concave [ao]" in the a key area indicates that when users extract "sub-radicals" from Chinese characters, "concave" should be removed as a whole radical, and "concave" should not be removed into more detailed radicals; on the contrary, the radicals without an asterisk are "easily ambiguous radicals" that users may have cognitive differences with. When users extract "sub-radicals" from Chinese characters, it is tacitly allowed for users to separate them.

[0241] 6) The radicals marked with % in front are all "uncommon radicals" that are rarely used by ordinary users. It is not recommended for ordinary users to extract "uncommon radicals" from Chinese characters. The present invention does not use these "uncommon radicals" in the "popular large radical decomposition scheme"; on the contrary, the radicals without % in front are all "popular radicals" that are easier for ordinary users to recognize. It is recommended for ordinary users to give priority to extracting "popular radicals" from Chinese characters. These radicals are given priority in the "popular large radical decomposition scheme" of the present invention.

[0242] 2. "Character root → letter-like key position" extended mapping method, as shown in [Table 2]:

[0243] [Table 2]: 18 additional radicals that can be assigned to corresponding letter keys based on morphological similarity clues and their key position expansion mapping methods:

[0244]

[0245] [Table 2] Description: The radicals in [Table 2] can be additionally assigned to corresponding letter-like key positions according to the clues described in [Table 2]. Among them, "冂" can be assigned to the u key or the n key. Users can choose one of the mapping methods according to their preferences.

[0246] 3. The extended mapping method of "radical → punctuation key position" is shown in [Table 3]:

[0247] [Table 3]: Additional key positions or shape clues assigned to punctuation key positions for 6 types of radical key expansion mapping scheme:

[0248]

[0249] [Table 3] Description: The radicals in [Table 3] can be additionally assigned to corresponding punctuation key positions according to the clues described in [Table 3].

[0250] 4. The error-tolerant mapping rules of “radical → standard pinyin first letter key position → Nanyin first letter key position” are shown in [Table 4]:

[0251] [Table 4]: Error-tolerant mapping rules for radical → standard pinyin first letter key position → Nanyin first letter key position

[0252]

[0253]

[0254] [Table 4] Description: When a root character should be assigned to the "standard pinyin first letter" key position in [Table 4] according to the mapping relationship stated in [Table 1], users with a heavy southern accent can assign the root character to the "Southern accent first letter" key position with the same sequence number as the "standard pinyin first letter" key position in [Table 4] ("Southern accent first letter" refers to the non-standard pinyin first letter affected by the southern accent); the root character beginning with "hu" in pinyin has the Southern accent first letter f or w (users can choose according to their own preferences), such as "胡" in Hunan is pronounced as fu, and in Guangdong it is pronounced as wu; For characters beginning with ji, the first letter of their Nanyin pronunciation is g, for example, "家" is pronounced as ga, and "街" is pronounced as gai; for characters beginning with wen in pinyin, the first letter of their Nanyin pronunciation is m or y (users can choose according to their own preferences), for example, "文" is pronounced as men and "翁" is pronounced as yong in Cantonese; for characters beginning with cheng in pinyin, the "first letter of their Nanyin pronunciation is s in Cantonese, for example, "成" is pronounced as seing in Cantonese; for characters beginning with c in pinyin but not cheng in pinyin, the "first letter of their Nanyin pronunciation is q in Cantonese, for example, "车" is pronounced as qe in Cantonese;

[0255] 5. "Character root → letter punctuation → number key position" extended mapping rule. After the character root is assigned to the letter, punctuation mark and space key positions on the letter punctuation keyboard, it can be further assigned to the number key positions, as shown in [Table 5]:

[0256] [Table 5]: Radical → letter punctuation → number key mapping table:

[0257]

[0258] [Table 5] Description: After mapping the radicals to letter punctuation key positions according to [Table 1], [Table 2], [Table 3], and [Table 4], the radicals can be further assigned to corresponding numeric key positions according to the mapping relationship described in [Table 5].

[0259] Section 4: Creating a phrase input method and a "phonetic-morphological-word code" system

[0260] 1. Create a simple phrase input method

[0261] (I) Simple input method for word association phrases: Using the "word association function" of the present invention, every time a Chinese character is input, a commonly used short phrase of 2 to 3 characters with the Chinese character as the first character will appear on the word selection screen for the user to select (the "word frequency" of a phrase refers to the number of times the phrase appears in every 10 million phrases in ordinary texts on average, the same below), and it can be used without any special learning. For example, after inputting the word "do", "done, do, make, do, do what, do one, do it, do it well, how to do it, do it, do not do it, done, do a, do an official, dream, ..." will be arranged on the word selection screen according to their word frequencies from high to low.

[0262] (II) Simple input method for phonetic initials phrases: Nearly 600,000 practical phrases in the present invention can all be input using the phonetic initials of the Chinese characters in the phrases. Take the phonetic initials of no more than 6 Chinese characters in the phrases and add a semicolon as the "phonetic initials phrase code" (if there are 6 letters, then stop. If there are less than 6 letters, add a semicolon ";" at the end). Type it and the corresponding phrase can be typed out, and it can be used without special learning. For example, type {cg;}, and the word selection screen will show "success, miss, exceed, pass, eat, sing, give, come out, Chu State..."; type {gmj;}, and the word selection screen will show "more dense, sell fine wine, mold, public sale bureau, coal supply machine..."; type {ybdt;}, and the word selection screen will show "one step to the sky, misty fishing, originally to him..."; type {nmgzzq} and you can type "Inner Mongolia Autonomous Region".

[0263] (III) Full spelling phrase input method: The present invention incorporates nearly 600,000 practical phrases, all of which can be input by typing the full spelling code of the phrase, and its usage is similar to the popular full spelling input method on the market.

[0264] 2. Create a phrase sound and shape joint coding input method

[0265] If users need to input phrases in a large-scale and universal way, they can use the "phonetic-graphic word code" compiled by the following "phrase phonetic-graphic joint coding rules" to enable most daily phrases to be typed on the screen in a blind manner, which can greatly improve the efficiency of phrase input and provide a universal blind typing experience:

[0266] (I) Phrase sound and shape joint coding rules:

[0267] 1. The encoding formula of the "phonetic and morphological code" of a phrase:

[0268] ① The phonetic code of the two-character phrase = W1p1+W2p1+; + W1s1+W2s1+W2s2

[0269] ② The phonetic and morphological code of the three-character phrase = W1p1+W2p1+W3p1+; + W1s1+W2s1

[0270] ③ The phonetic and morphological code of the four-character phrase = W1p1+W2p1+W3p1+W4p1+;+W1s1

[0271] ④ The phonetic and morphological code of the five-character phrase = W1p1+W2p1+W3p1+W4p1+W5p1+;

[0272] ⑤N-word phrase (N>=6) sound and shape code = W1p1+W2p1+W3p1+W4p1+W5p1+W6p1

[0273] For example: ① the phonetic and morphological code of "like" is {xh; snz} (like x + good h +; + scholar s + woman n + son z), ② the phonetic and morphological code of "organization department" is {zzb; ss} (organization z + organization z + department b +; + s + s); ③ the phonetic and morphological code of "everyone is happy" is {jdhx; b} (everyone is happy + big d + happy h + happy x +; + than b), ④ the phonetic and morphological code of "Krupp cannon" is {klbdp;} (krupp k + l + bob + big d + cannon p +;), ⑤ the phonetic and morphological code of "Inner Mongolia Autonomous Region" is {nmgzzq} (inner n + Mongolia m + ancient g + self z + z + district q)

[0274] 2. The encoding formula of "phonetic-morphological-word code" is explained as follows:

[0275] ①W1p1 / W2p1 / W3p1 / W4p1 / W5p1 / W6p1 are the pinyin initials of the 1st / 2nd / 3rd / 4th / 5th / 6th Chinese characters respectively;

[0276] ② In the phrase coding process, according to the "simple character splitting method" of the second point (a) of the first section and the "radical mapping coding rule" of the third point, the relevant Chinese characters in the phrase are split, and the split radicals are assigned to key symbols, W1s1 / W2s1 are the mapping key symbols of the first radical of the first / second Chinese character, respectively, and W2s2 is the mapping key symbol of the second radical of the second Chinese character;

[0277] ③ The "sound-shape word code" consists of two parts: "pinyin initial letter string" and "root mapping key string" (hereinafter referred to as "shape code part"). ";" is the separator between the two parts and appears in the "sound-shape word code"; "+" is only used in the formula and will not appear in the "sound-shape word code"; ④ The more characters a phrase has, the fewer "shape code parts" need to be appended to it, and the simpler the input operation is: 2 / 3 / 4 characters need to be appended with 3 / 2 / 1-bit shape codes respectively, and phrases with more than 4 characters do not need to append the shape code part.

[0278] (II) Compatibility rules of phonetic, morphological and word codes

[0279] In the "shape code part", the "popular big root character splitting scheme" is used to split the characters, and the pinyin initial letter of the first pronunciation of the root is used to map the code, and the resulting sound-shape word code is the "priority word code". In the "shape code part", other word splitting methods other than the "popular big root character splitting scheme" are used to split the characters, or other letters or punctuation key character encodings other than the pinyin initial letter of the first pronunciation of the root are used, and the resulting sound-shape word code is the "compatible word code". Each phrase has and has only one priority word code, and there can be multiple compatible word codes, such as: the priority word code of "like" is {xh;snz} (like x + good h +: + scholar z + female n + son z); its compatible word code is {xh;znz} (xh +; + 壴z + female n + son z) or {xh;jnz} (xh +; + lucky s + female n + son z).

[0280] Many radicals are "polyphonetic radicals", and some radicals can be assigned to keys that resemble letters or punctuation marks. These radicals that can be mapped to more than one key are collectively referred to as "multi-mapping radicals". Many Chinese characters have multiple ways of splitting. For example, if the relevant Chinese characters in a phrase have multiple ways of splitting, compatible encodings of the corresponding multiple ways of splitting are created for their phonetic and phonological codes.

[0281] "Compatibility Code" creates strong "compatibility" from the following aspects:

[0282] 1. In terms of character splitting methods, the "popular large root character splitting scheme" is given priority, while other compatible character splitting methods such as the "theoretical large root character splitting scheme" and the common "personalized root selection character splitting scheme" are also compatible.

[0283] For example, "今" is broken down into "人丶" according to the "common root splitting scheme". ", the priority code of "today" is "today {jt} + {;} + person {r} + two {e} + person {r}" = {jt; rer}; "today" is split into "亽" according to the "theoretical big root splitting scheme" ”, and the compatible code of “today” is “today{jt}+{;}+亽ji{j}+二{e}+人{r}”={jt;jer}.

[0284] 2. Compatible with "one root multiple mappings" in terms of root mapping letters

[0285] For example, if compatible word codes are created for all the different pronunciations of the "multi-mapping radicals" of the Chinese characters involved in the shape codes in a phrase, when there are multiple "multi-mapping radicals" in the shape code part of the phrase, due to the combination, the compatible word codes will increase multiplicatively to generate a huge number of compatible word codes, which will not only not help to improve efficiency and user experience, but will greatly increase the re-coding rate of the phrase.

[0286] Fortunately, through the actual test of 600,000 phrases, it was found that, in the case of including all compatible word codes, after typing the first 5 codes that can include semicolons in the phonetic word codes, more than 90% of the phrases can appear on the homepage of the word selection screen. For these phrases, it is only necessary to provide the multi-pronunciation first letter compatible word code of the "multi-mapped root" in the first 5 codes (only 4 letters except the semicolon) to meet the user's normal phrase input needs, without providing the compatible code of the 6th code (including the semicolon);

[0287] For the less than 10% of the phrases that fail to appear on the first screen of the word selection screen after the first 5-digit code (including the semicolon) of the phonetic-morphological word code is input, the present invention simultaneously creates "phonetic-morphological word codes" for these phrases that are compatible with all different pronunciations or different decomposition methods of the "multi-mapping radicals" therein, so that the user can easily type the phrase without turning the screen no matter which mapping method of the radical is used. For example, "self" can be encoded in many ways: ①{zj;pfz}(self=丿目=pie=p, self=]fang+折zhe=f+z); ②{zj;puz}(self=丿目=pie=p, self=]u+折zhe=u+z); ③{zj;pvh}(self=丿目=pie=p, self=折v+一heng=v+h); ④{zj;pzh}(self=丿目=pie=p, self=折zhe+一heng=z+h); ⑤{zj;xfz}(self=丿目=pie=p, self=折zhe+一heng=z+h); Three=xiang=x, one=]fang+fold zhe=f+z); ⑥{zj;xuz}(自ニ Three = xiang = x, one =]u+zhefold = u+z)…

[0288] For high-frequency 2-3-word phrases that fail to appear on the first screen of the word selection screen and are ranked in the top 10 in frequency, the present invention creates a priority privilege short code for them. As long as the first 4-5 digits of the code including the semicolon {;} are entered, they can appear on the first screen of the word selection screen. For example: typing {bb;b} can get: "not, mean, not, help, and, vigorous, north, running, not sick, half"; typing {bb;bb} can get: "mean, help, vigorous, ice, bibi, backpack, disease, defeat, half, poor gang".

[0289] 3. Create the "numeric word code" for a phrase: According to the "radical → letter punctuation → numeric key mapping method" (Item (2) of Point (3) in Section 1), convert all the letter punctuations in the phonetic shape word code into numbers, and the "numeric word code" of the phrase can be obtained.

[0290] 4. Automatically and compatibly adapt to the user's coding usage preferences:

[0291] With the "compatible word code", users can easily type a phrase without precisely giving the "preferred word code" of the phrase. Coupled with the "dynamic ranking function of word and character coding" (see Item (9) of Point (9) in Section 2), if a user often uses a certain "compatible word code" to input a certain phrase, after several repeated operations, this phrase can be typed blindly using this "compatible word code".

[0292] (3). The method for prioritizing and ranking the "phonetic shape word code"

[0293] 1. Echelon division of the "phonetic shape word code" and its initial prioritizing and ranking method: All phonetic shape word codes are divided into three word code echelons. The "preferred word code" (see Item (2) of Point (2) in Section 4) is used as the first echelon of word codes and is placed at the very front in the phrase code table; The phonetic shape word codes using the "southern accent initial letter" coding in the shape code part are used as the third echelon of word codes for a small number of users with southern accents to select, and are placed at the very end of the phrase code table; The phonetic shape word codes in other ways are used as the second echelon of word codes and are placed behind the first echelon of word codes and in front of the third echelon of word codes.

[0294] 2. The initial prioritizing and ranking method for homograph phrases in the same word code echelon: On the basis of the initial ranking described in Point 1 above, the phrase and its "phonetic shape word code" are sorted in the code table according to the composite index of "C + STR((10^6) - int(10*F))" (here, "C" is the "phonetic shape word code" string, and "F" is the word frequency of the phrase). When a "phonetic shape word code" corresponding to multiple phrases is typed, the homograph phrases with higher word frequencies in the same word code echelon are presented in the front position on the character selection screen;

[0295] 3. The dynamic ranking of the "phonetic shape word code": On the basis of the above two initial prioritizing and ranking methods of the phonetic shape word code, a dynamic ranking function of the phonetic shape word code is further provided. The specific method is described in Item (9) of Point (9) in Section 2.

[0296] (4). Typing "phonetic-shape word code" and selecting / inputting phrases: After the user types the aforementioned "phonetic-shape word code" on the root key keyboard (when inputting "numeric word code", the dot "." must be used as the start and end symbol of the "numeric word code"), for some high-frequency phrases, only the first 3 to 5 digits of the "phonetic-shape word code" need to be input, and the corresponding high-frequency phrases will appear on the character selection screen; when the phrase the user wants does not appear on the character selection screen, the user must use the page turning key to browse the content on the character selection screen; when the phrase the user wants appears on the character selection screen, the user keys in the corresponding serial number on the left side of the phrase on the character selection screen to select and input the phrase; when the phrase appears at the top of the character selection screen, the space bar can be clicked to input it; when the phrase appears on the character selection screen on the touch screen, the phrase can be clicked on the character selection screen on the touch screen to input it; when the user has typed the phrase code and still cannot find the desired phrase after browsing through the entire candidate word list on the character selection screen, the "code lookup dictionary" provided by the present invention can be used to query the preferred character splitting code method for relevant words (for the usage of the code lookup dictionary, see the statement in Section 8, Point 6 (1)).

[0297] III. Creating a "phonetic-shape word code" database for 600,000 practical phrases

[0298] (1). Inputting the phrases included in "Xinhua Dictionary", "Modern Chinese Dictionary", "Chinese-English Dictionary", "English-Chinese Dictionary", etc. through methods such as OCR, manual input, and manual proofreading; using database technology to extract phrases from more than two hundred e-books and a large number of online articles containing approximately 60 million Chinese characters in total, and conducting word frequency statistics to obtain a "database of 600,000 phrases and their word frequencies".

[0299] (2). Using database technology, according to the "joint coding rule of phrase phonetic-shape" (see Section 4, Point 2 (1)) and the Chinese character splitting coding scheme in the "total library of letter word codes" (see Section 2, Point 9 (7)), assigning "phonetic-shape word codes" to the aforementioned 600,000 phrases, and a "total database of phonetic-shape word codes" containing data such as 600,000 phrases and their word frequencies, phonetic-shape word codes, etc. is obtained. If a phrase contains multiple character splitting coding methods for the Chinese characters it contains, then the phrase also has multiple corresponding phonetic-shape word codes.

[0300] IV. Creating functions for adding new phrases and exporting / importing custom phrases

[0301] The 600,000 phrases provided by the present invention are sufficient to meet the typing needs of most users. However, users in different industries have different requirements for phrases. Therefore, the present invention provides a very simple "adding user phrases" function. When a user finds that a certain phrase that they need to frequently input is not in the system thesaurus (after typing the character-shape word code but cannot type out the desired phrase and cannot find the phrase in the code lookup dictionary of the present invention), the user can use the following method to add the new phrases they need to the system.

[0302] (1). Click the first icon (small gear) on the right side of the input method status bar to pop up the settings menu; (2). Click "Add User Word" in the settings menu to pop up the floating window of [Add Entry - Self-formed Character Input Method], and then you can start adding new phrases; (3). Enter the phrase you want to add in the "Entry" box in the floating window (you can use other input methods), and the system will automatically generate the "Phonetic and Shape Word Code" of this phrase according to the phrase coding rule; (4). After the user browses and checks that the code is correct (you can also manually modify the code according to your own idea), click [OK] to save, and click [Cancel] to give up; (5). The phrase only needs to be added once and can be used permanently.

[0303] The present invention can export the new word list text added by the user during the use process for the user to share and use on other computers, or be used to import the self-built word list to restore the word library added by the user when installing and upgrading this software. Once the user has mastered the powerful "Add New Word" function of this system, they can add private phrases to their own exclusive word library at will, and typing will become more and more handy.

[0304] Section 5 Create the text of the "Total Alphabet Punctuation Coding Table", the word association code table, and the code lookup dictionary text of the present invention

[0305] 1. Create the text of the "Total Code Table of Alphabet and Punctuation Encoding": Export the "Total Library of Alphabet and Character Codes" described in Section 2, Point 9 (7) and the "Total Database of Phonetic, Shape and Word Codes" described in Section 4, Point 3 into a code table text in a format compliant with the "Duoduo Input Method Generator", and integrate it with other auxiliary code table texts to form the "Total Code Table of Alphabet and Punctuation Encoding". Its specific content includes the following: (1) Help document code table: Facilitate users to input special symbols, query the pronunciation of radicals, query the encoding rules of word groups, etc., for quick help information; (2) More than 770,000 standard shape codes of "pure shape character codes" for GBK Chinese characters (excluding the southern accent compatibility encoding); (3) More than 480,000 "pure shape character codes" in the "southern accent compatibility encoding" that can be optionally configured for GBK Chinese characters (only for users with southern accents to select); (4) More than 25,000 full spelling "word and phrase encodings" for GBK Chinese characters (a multi-syllable Chinese character can have multiple full spelling "word and phrase encodings"); (5) More than 21,000 stroke "word and phrase encodings" for GBK Chinese characters; (6) More than 5.98 million "word and phrase encodings" of the first letter of the pinyin of word groups (a multi-syllable word group can have multiple "word and phrase encodings" of the first letter of the pinyin); (7) More than 21.48 million standard phonetic, shape and word codes of word groups (including multiple compatible encodings of word groups, but not including the "southern accent compatibility encoding"); (8) More than 13.5 million optionally selected southern accent compatibility phonetic, shape and word codes of word groups (only for a small number of users with southern accents to select); (9) More than 5.98 million full spelling "word and phrase encodings" of word groups; (10) Keyboard input root sound encoding: 1. Stroke root sound encoding: Enter {; + the strokes of the radical} to view the "preferred pronunciation" of 2029 radicals on the character selection screen; 2. Shape code root sound encoding: Enter {; + the shape code of the radical} to view the "preferred pronunciation" of the radical on the character selection screen.

[0306] 2. Create the word and phrase association code table and the code lookup dictionary text: (1) The word and phrase association code table has 117,000 entries (independent code table), providing an association library of 2-word and 3-word phrases with each common Chinese character as the first character; (2) The code lookup dictionary text (independent text) details more than 3.6 million lines of content such as the pinyin, strokes, location codes, recommended splitting methods and encodings of all GBK Chinese characters and 600,000 word groups.

[0307] Section 6. Create the text of the "Total Code Table of Digital Keyboard Codes" of the present invention

[0308] According to the "Radical -> Alphabet and Punctuation -> Digital Keyboard Key Mapping Method" (see Section 1, Point 3 (2) 4), convert all the letter and punctuation symbol encodings in the "Total Code Table of Alphabet and Punctuation Encoding" described in the previous section into digital "word and phrase encodings" according to the mapping method in [Table 5] of Section 3, and save it as the "Total Code Table of Digital Keyboard Codes".

[0309] Section 7. Create the root keyboard of the present invention

[0310] 1. Create the root alphabet and punctuation keyboard: Its characteristics are as follows:

[0311] (1). An entity keyboard with a "QWERTY" classic layout and no less than 48 keys, which at least includes: ① 26 English letter keys and 5 punctuation keys ",. / ’[" (note: the key names and symbols in this article are enclosed in double quotes, but the content stated does not include double quotes), used to map Chinese radicals and input Chinese words and phrases; ② 1 "Backspace" key used to delete the input content; ③ 2 page turning keys used to turn the character selection screen forward and backward; ④ 10 Arabic numeral keys "0123456789" used to input numbers and select desired words and phrases from the character selection screen; ⑤ 1 semicolon key used to input ";"; ⑥ 1 "Shift" key used to temporarily switch the key input mode, and the two keys "shift" and " / " can be pressed simultaneously to input the question mark "?"; ⑦ 1 "Space" key; ⑧ 1 "Enter" key used to confirm the input and line break, and other functions set by the user (for example: assist in inputting half-width key symbols, clear the content of the coding input box floating with the cursor).

[0312] (2). A screen virtual keyboard with a "QWERTY" classic layout and no less than 34 keys, which at least includes: ① 26 English letter keys used to map Chinese radicals and input Chinese words and phrases; ② 1 "Backspace" key; ③ 2 page turning keys; ④ 1 semicolon key; ⑤ 1 question mark key; ⑥ 1 "Shift" key; ⑦ 1 "Space" key; ⑧ 1 "Enter" key.

[0313] (3). The mapping relationships among the radicals, letters, and punctuation symbols and other keys on the radical letter punctuation keyboard are as shown in [Table 1], [Table 2], [Table 3], [Table 4] in Section 3 and their related descriptions.

[0314] 2. Create a radical number keyboard: Its characteristics are as follows:

[0315] (1). A number keyboard with no less than 17 keys, which at least includes 10 number keys "0123456789", 5 floating-point operation keys ".+-* / ", and 2 operation keys: the Enter key and the NumLock key used to switch between the number or Chinese input mode.

[0316] (2). When the number keyboard is set to input Chinese: ① Keys such as "0123456789" are used to input numbers, map Chinese radicals and input the codes of Chinese words and phrases, and input the serial numbers on the character selection screen to select the corresponding words and phrases after the input of the word codes is completed; ② The "*" key is used to delete the input content; ③ The "+" and "-" keys are used to turn the character selection screen forward and backward; ④ The "." key is used to input the dot "."; ⑤ The " / " key is used to input " / "; ⑥ The "Enter" key (which can be set by the user).

[0317] (3) The mapping relationship between the radicals and the numeric keys on the radical numeric keyboard can be found in the statement in Section 3 of Part 4.

[0318] Section 8. Creating the Input Method Installation Application Software of the Present Invention

[0319] The present invention utilizes the domestic input method generator software "Duoduo Input Method Generator". By setting relevant parameters and importing Chinese code table texts that conform to its specified format, the input method installation application software applicable to the Windows system of the present invention can be created:

[0320] 1. Set the input method - related parameters in "Duoduo Input Method Generator" according to the needs of the present invention: For example, set the function of dynamic ranking of character and word encodings, set the function of intelligently adding new phrases and automatically encoding...; 2. Create the alphabet version: Based on the "Total Code Table of Alphabet and Punctuation Encodings" (see Point 1 of Section 5 for details), generate the installation software of "Self - composed Character Input Method (Alphabet Version)"; 3. Create the numeric version: Based on the "Total Code Table of Numeric Keyboard Codes" (see Section 6 for details), generate the installation software of "Self - composed Character Input Method (Numeric Version)"; 4. Create the word and phrase association function: Based on the "Word and Phrase Association Code Table" (see Point 2 of Section 5 for details), generate the "word and phrase association" function; 5. Create the code - checking dictionary: Based on the "Code - checking Dictionary" (see Point 2 of Section 5 for details), generate the "code - checking dictionary" function; 6. Create multiple quick code - checking tutoring functions: By providing the following code - checking tutoring functions: (1) Pop - up code - checking of the "Code - checking Dictionary": ① In the input method running environment of the present invention, use the built - in full - spelling code and stroke code of the present invention to input the Chinese character or phrase for which the code is to be checked. When the Chinese character or phrase appears on the character selection screen, hover the mouse over it for 1 second, and a small floating window will pop up. Click "View Detailed Explanation" in the small floating window to activate the "Code - checking Dictionary", and the code - checking dictionary of the present invention will pop up in a pop - up window. The preferred character - splitting encoding method of the Chinese character or phrase is displayed in the dictionary pop - up window; ② The user can input any word or phrase in the input box of the "Code - checking Dictionary" to query its "Popular Big - Root Character - splitting Scheme"; ③ The user types {help} in the input box of the "Code - checking Dictionary" and selects to input "Help·" to view explanations such as "Help Method, Typing Rules, Radical Details, One - key Typing, Word Code Rules, Master Skills"; (2) Stroke - based radical pronunciation: Type {; + stroke encoding of the radical} = view the "preferred pronunciation" of the corresponding radical on the character selection screen; (3) Shape - code - based radical pronunciation: Type {; + shape code of the radical} = view the "preferred pronunciation" of the corresponding radical on the character selection screen; (4) View the quick tutoring text: Type {;} = view the quick tutoring text on the character selection screen (use the page - turning keys to view the full text by page); (5) View the word code rules: Type {czbmgz} = view a brief introduction to the "phonetic - shape combined encoding" rules of the phrase on the character selection screen (czbmgz = the first letters of the pinyin of "phrase encoding rules").

[0321] Section 9. Partial Technical Effect Data of the Present Invention

[0322] 1. The root set includes 3,196 roots, among which there are 5 "stroke" roots, 137 "radical" roots, and 3,054 "character within character" roots. More than 95% of the roots are "character within character" roots, and the intuitiveness of the roots is extremely strong, conforming to the cognitive law of the public.

[0323] 2. More than 770,000 shape code "word and phrase codes" are created for 21,000 GBK Chinese characters (on average, each character has more than 36 shape codes).

[0324] 3. More than 230,000 blind typing shape codes for 21,000 GBK Chinese characters are created, including 1,942 two - digit blind typing shape codes, 18,600 three - digit blind typing shape codes, and 215,900 four - digit blind typing shape codes. On average, each character has more than 11 blind typing shape codes; all 5,000 GBK Chinese characters with the highest word frequencies can be blindly typed using the "popular large root character code"; more than 90% of GBK Chinese characters can be blindly typed using the "popular large root character code"; more than 99% of GBK Chinese characters can be input without screen flipping using the "four - digit character code".

[0325] 4. More than 598,000 phrases and more than 2.25 million sound - shape word codes are provided (on average, each phrase has more than 3.7 sound - shape word codes). More than 420,000 phrases can be blindly typed, and the blind typing rate of phrases exceeds 70%, and the initial blind typing rate of daily phrases exceeds 96%.

[0326] 5. The average number of encoding characters per single - character encoding is <3.84, and the average number of encoding characters per phrase encoding is <2.25.

[0327] Note: The above data are all statistical data without duplicate encoding, and only the standard word and phrase encodings that do not include the "Southern accent error - tolerant encoding" are counted (since the "Southern accent error - tolerant encoding" belongs to niche requirements and has no general representative significance, it is not included in the statistical report).

Claims

1. A Chinese encoding input method based on an optimized radical keyboard, characterized in that: 1.1 Optimize the key composition and function configuration of the root keyboard 1.1.1 Root letter punctuation keyboard, its characteristics are ： 1.1.1.1 A physical keyboard with no less than 48 keys in the classic "QWERTY" layout, which includes at least the following keys and their functions: ① 26 English letter keys and 5 punctuation keys ",. / ′[", for mapping Chinese radicals and inputting Chinese "word codes" (Note: "word codes" refer to keyboard key symbol codes that can be recognized and processed by computers by converting Chinese characters or phrases according to specific rules, so that users can input Chinese information into computers through the keyboard. The specific forms of "word codes" include "stroke codes", "full spelling codes", "pure shape code" in "3.3" of this article, and "phonetic shape code" in "4.2" of this article; "mapping" refers to the process and corresponding relationship of assigning radicals to keyboard keys according to rules; the key symbols stated in this article are uniformly enclosed in double quotes, and all The content of the statement does not include double quotes, for example, ";" only represents a semicolon, not double quotes), ②1 "backspace key" is used to delete the input content, ③2 page keys "PageUp" and "PageDn" are used to flip forward and backward through the content on the word selection screen, ④10 Arabic numeral keys "0123456789" are used to input numbers, select and input the words the user wants from the word selection screen, ⑤1 semicolon key is used to input ";", ⑥1 shift key "Shift" is used to temporarily switch the key input mode, and the "shift" and " / " keys can be pressed at the same time to input a question mark "?", ⑦1 "space bar" is used to input a space, select and input the first word on the word selection screen, ⑧1 "Enter" key (the specific function can be personalized according to user needs); 1.1.1.2 A virtual keyboard with a classic "QWERTY" layout and no fewer than 34 keys, including at least: ① 26 English letter keys, used to map Chinese character roots and input Chinese word codes, ② 1 "Backspace key" to delete the input content, ③ 2 page turning keys "PageUp" and "PageDn" to flip forward and backward through the content in the character selection screen, ④ 1 question mark key to input "?", ⑤ 1 semicolon key to input ";", ⑥ 1 shift key "Shift" to temporarily switch the key input mode, ⑦ 1 "Space bar" to input space, ⑧ 1 "Enter" key to confirm input and line break, and other functions set by the user (for example: assist in inputting half-width key symbols, clear the content of the code input box floating with the cursor); 1.1.1.3 The mapping relationship between radicals and key positions in the radical letter punctuation keyboard is as shown in [Table 1], [Table 2], [Table 3], and [Table 4] in claim 2; 1.1.2 Root numeric keyboard, its characteristics are ： 1.1.2.1 A numeric keyboard with no less than 17 keys, including at least 10 numeric keys "0123456789", 5 floating point operation keys ".+-* / ", and 2 operation keys "Enter" and "NumLock". The functions of each key are as follows: 1.1.2.2 The NumLock key is used to switch between numeric and Chinese input modes; 1.1.2.3 When the numeric keypad is set to input Chinese, the functions of the relevant keys are as follows: ① The ten numeric keys "0123456789" are used to input numbers, map Chinese radicals, and input the codes of Chinese words and phrases. After inputting the codes of Chinese words and phrases, input the serial numbers on the character selection screen to select and input the words and phrases corresponding to the numbers. ② The "*" key is used to delete the input content. ③ The "+" key and the "-" key are used as screen flipping keys to flip the character selection screen forward and backward. ④ The "." key is used to input the dot ".", ⑤ The " / " key is used to input " / ", ⑥ The "Enter" key is used to confirm the input and line break, as well as other functions set by the user (for example: assist in inputting half-width key symbols, clear the content of the floating coding input box following the cursor); 1.1.2.4 The mapping relationship between the radicals and the numeric keys on the radical numeric keypad is as shown in [Table 5], [Table 1], [Table 2], [Table 3], and [Table 4] in Claim 2; 1.2 Select 3,196 radicals as the structural components for forming Chinese characters. The specific radicals and their mapping relationships with the corresponding keys on the optimized radical keypad are as described in Claim 2, and it is characterized in that: 1.2.1 A "radical" is a structural component that makes up a Chinese character, and is divided into three categories: "strokes", "radicals", and "characters within characters"; 1.2.2 "Stroke" is the most basic structural component that makes up Chinese characters, which can be divided into five types: "dot, horizontal, vertical, left-falling stroke, and fold". For the convenience of description, in this article, "丶, 一, 丨, 丿, ” are used to represent the strokes "dot, horizontal, vertical, left-falling stroke, and fold" respectively. The characteristics of stroke radicals are as follows: 1.2.2.1 "Dot": All short strokes without turning angles, and all strokes that extend diagonally downward without turning angles (right-falling strokes), are classified as "dots" (for example but not limited to "丶"); 1.2.2.2 "Horizontal": All strokes that extend horizontally to the right without turning angles (such as the character "一"), and all strokes that extend diagonally upward to the right without turning angles (such as the "raising stroke" in the character "七"), are classified as "horizontals" (for example but not limited to "一"); 1.2.2.3 "Vertical": All vertical strokes without turning angles (such as "丨"), and all strokes that extend vertically downward and then turn diagonally upward to the left (such as the "vertical hook" in the center of the character "水"), are classified as "verticals"; 1.2.2.4 "Left-falling stroke": All strokes that extend diagonally downward to the left without turning angles are classified as "left-falling strokes" (for example but not limited to "丿"); 1.2.2.5 "Turn": All strokes that have turning angles or are circular in the middle of the extension, except for "vertical hooks", are classified as "turns" (for example but not limited to: "乙", "", "", "乛", "乚", "○"); 1.2.2.6 During the character decomposition and coding process of the present invention, any "radical" composed of a single stroke is preferentially regarded as a "stroke" radical, rather than being recommended as a "character within character" radical or a "radical" radical (for example but not limited to "、一丨乙"); 1.2.3 "Radical": It is a Chinese writing structure composed of no less than two strokes, without a complete meaning, and cannot be used as an independent and complete Chinese character in modern Chinese sentences, but can be used as a component within Chinese characters (for example but not limited to: "艹, 氵, 扌, 钅, 疒, 勹, , 亻, 夂, , 癶, 灬"); 1.2.4 "Chinese characters within Chinese characters": It is a Chinese character writing structure composed of no less than two strokes, with a complete meaning. It can be used as a complete and independent Chinese character in modern Chinese sentences and can also serve as a structural component within Chinese characters (such as but not limited to: "wood, fire, earth, metal, water, sun, moon, sky, people, mouth, field, mountain, grain, white, black").

2. According to Claim 1, for the root characters and the mapping relationship between them and the corresponding key positions on the optimized root character keyboard, it is characterized in that: 2.1 The mapping relationship between the root characters and the initial letter keys of pinyin is as shown in [Table 1]: [Table 1]: 3,196 root characters, the classification identifiers of the root characters, and the methods of assigning the root characters to the letter keys are described after the table. Explanation of [Table 1]: 1) The root characters are arranged in the table frame of the corresponding letter key area according to the initial letter of their first pronunciation ("first pronunciation" refers to the only pronunciation of the root character or the most popular pronunciation of the root character in society). Different root characters are separated by semicolons: 2) The pinyin of each root character is marked with square brackets "[]" behind it. A root character with multiple pronunciations is called a "polysyllabic root character" (the same below). The different pinyins of the polysyllabic root character are separated by commas, and some pinyins are annotated with their pronunciations in parentheses at the back. For example, in the a key area of the table, "[ang (bottom of the character 'ang'), yang (raise the head)]" indicates that the root character "卬" can be pronounced as "ang (bottom of the character 'ang')" or "yang (raise the head)"; 3) Each root character can be assigned to the corresponding letter key position according to the initial letter of its first pronunciation. The polysyllabic root character can also be assigned to the corresponding letter key position using the initial letters of other pronunciations listed in the square brackets behind it: 4) For some high-frequency polysyllabic root characters, their first pronunciation is well-known and commonly used, while their other pronunciations are rarely used. For such root characters, only their first pronunciation is used as the pinyin mapping clue for assigning them to the letter key position. The pinyin that is abandoned as the mapping clue is marked with "(abandon...)". For example, in the z key area, "[zhui, (abandon chui)]" indicates that for the root character "椎", only its first pronunciation "zhui" is used as the mapping clue for assigning it to the initial letter key of pinyin, and the second pronunciation "chui" is abandoned as the clue for assigning it to the initial letter key of pinyin; 5) The root characters marked with an asterisk in front are all "basic root characters" that are common and frequently used by ordinary users. When the user extracts "sub-root characters" from Chinese characters (the definition of "sub-root characters" is detailed in "3.2.1" of this article), it is recommended that the user extract these "basic root characters" as a whole, rather than disassembling these "basic root characters" into more subdivided root characters. For example, in the a key area, "*凹[ao]" indicates that when the user extracts "sub-root characters" from Chinese characters, the root character "凹" should be extracted as a whole, rather than disassembling "凹" into more subdivided root characters when extracting "sub-root characters". On the contrary, the root characters without an asterisk in front are all "prone-to-disambiguation root characters" that may cause cognitive differences among ordinary users. When the user extracts "sub-root characters" from Chinese characters, the user is默许 to disassemble these "prone-to-disambiguation root characters": 6) The radicals marked with a % sign in front are "rare radicals" that are rarely used by ordinary users. The present invention does not recommend that ordinary users give priority to extracting "rare radicals" from Chinese characters. The present invention does not use these "rare radicals" in the "Popular Big Radical Character-Splitting Scheme": On the contrary, the radicals without a % sign in front are "popular radicals" that are relatively easy for ordinary users to recognize. The present invention recommends that ordinary users give priority to extracting "popular radicals" from Chinese characters. The "Popular Big Radical Character-Splitting Scheme" of the present invention gives priority to using these "popular radicals": 2.2 The "radical → similar-shaped letter key position" extended mapping method, as shown in [Table 2]: [Table 2]: 18 radicals that can be additionally assigned to the corresponding letter key positions according to the clue of morphological similarity and their key position extended mapping methods Explanation of [Table 2]: In addition to being assigned to the corresponding first letter key position of the pinyin according to the method described in [Table 1], the radicals in [Table 2] can be additionally assigned to the corresponding similar-shaped letter key positions according to the clues shown in [Table 2]. Among them, the radical "冂" can be assigned to the u key or the n key, and the user can choose one of the mapping methods according to their own cognitive preferences; 2.3 The "radical → punctuation key position" extended mapping method. Six radicals, namely "木, 月, 扌, 氵, 钅, 釒", can be additionally assigned to the corresponding punctuation key positions, as shown in [Table 3]: [Table 3]: Key position extended mapping scheme for 6 radicals that can be assigned to punctuation key positions according to key position location or morphological similarity clues Explanation of [Table 3]: In addition to being assigned to the corresponding letter key positions according to the mapping relationships described in [Table 1] and [Table 2], the radicals in [Table 3] can be additionally assigned to the corresponding punctuation key positions according to the clues shown in [Table 3]: 2.4 The "radical → standard pinyin first letter key position → southern accent first letter key position" fault-tolerant mapping rule, as shown in [Table 4]: [Table 4]: Radical → standard pinyin first letter key position → southern accent first letter key position fault-tolerant mapping rule Explanation of [Table 4]: When there is a radical in the radical string that should be assigned to the "standard pinyin first letter" key position in [Table 4] according to the mapping method described in [Table 1], users with a stronger southern accent can assign the radical to the "southern accent first letter" key position with the same serial number as the "standard pinyin first letter" key position in [Table 4] ("southern accent first letter" refers to the non-standard pinyin first letter affected by the southern accent); for radicals starting with hu in pinyin, its southern accent first letter is f or w (users choose according to their own accent preferences). For example, "胡" is pronounced fu in Hunan dialect and wu in Cantonese; for radicals starting with ji in pinyin, its southern accent first letter is g. For example, "家" is pronounced ga and "街" is pronounced gai; for characters starting with wen in pinyin, its southern accent first letter is m or y (users choose according to their own accent preferences). For example, "文" is pronounced men and "翁" is pronounced yong in Cantonese; for radicals with the pinyin cheng, its "southern accent first letter" in Cantonese is s. For example, "成" is pronounced seing in Cantonese: for radicals with the first letter c in pinyin but not the pinyin cheng, its "southern accent first letter" in Cantonese is q. For example, "车" is pronounced qe in Cantonese: 2.5 "Character root → letter punctuation → number key" extended mapping rule: After the character root is assigned to the letter, punctuation mark and space key on the letter punctuation keyboard, it can be further assigned to the number key, as shown in [Table 5]: [Table 5]: Radical → Letter Punctuation → Number Key Mapping Table [Table 5] Description: After mapping the radicals to letter punctuation key positions according to [Table 1], [Table 2], [Table 3], and [Table 4], the radicals can be further assigned to corresponding numeric key positions according to the mapping relationship described in [Table 5].

3. According to claim 1, a method for inputting a single Chinese character code, characterized in that: 3.1 Method of splitting a single Chinese character into radical strings 3.1.1 Based on the user's existing knowledge of the writing structure and stroke order of Chinese characters ("Chinese characters" refers to Chinese characters and the radicals within Chinese characters), the user searches for a "character within a character" or "radical" that the user recognizes from the starting point of the remaining part of the split Chinese character and splits it out. If the user cannot find a "character within a character" or "radical" that the user recognizes at that point, the user splits out the first "stroke" of the remaining part. For a Chinese character composed of a single stroke, the single stroke is directly taken out as the "radical" within the Chinese character; 3.1.2 Repeat the operation described in "3.1.1" of this article until the first four "radicals" in the writing order of the Chinese character to be split are split (when the number of "radicals" contained in the Chinese character is less than 4, then split until all the "radicals" are split), and then refer to the writing order of these "radicals" in the Chinese character, arrange them into a string, which is called a "radical string"; 3.1.3 According to the method described in "3.1.1" of this article, users with different cognitive levels and cognitive preferences may have different ways of splitting non-single-stroke Chinese characters. Users should use one of the splitting methods according to their own cognitive levels and cognitive preferences; 3.2 Method of adding radicals to radical strings When the number of "radicals" contained in the "radical string" described in "3.1.2" of this article is less than 4, the following method can be used to add "radicals" to the "radical string" so that the number of "radicals" contained in the "radical string" increases to 4: 3.2.1 Split the Chinese character that the user wants to input according to the method described in "3.1.1" of this article, and the resulting "radical string" is called the "sub-radical string" of the Chinese character, and the "radicals" in the "sub-radical string" are called the "sub-radicals" of the Chinese character; Split the "sub-radical" of the Chinese character according to the method described in "3.1.1" of this article, and the resulting more subdivided radicals are called the "grandson radicals" of the Chinese character (when the "sub-radical" of the Chinese character is a single stroke, the "grandson radical" of the "sub-radical" is also the single stroke); Split the first "sub-radical" in the "sub-radical string" of the Chinese character according to the method described in "3.1.1" of this article, and the resulting first "grandson radical" is called the "first grandson radical" of the Chinese character; Split the second "sub-radical" in the "sub-radical string" of the Chinese character according to the method described in "3.1.1" of this article, and the resulting first "grandson radical" is called the "second grandson radical" of the Chinese character; 3.2.2 Append the "first Sun radical" described in "3.2.1" of this article to the end of the "radical string" described in "3.1.2" of this article. If the "radical string" already has 4 radicals after the "first Sun radical" is appended, then the operation of appending radicals is terminated. Otherwise, append the "second Sun radical" described in "3.2.1" of this article to the end of the "radical string" to increase the number of radicals in the "radical string" to 4. If the Chinese character that the user wants to input is a single-stroke character, repeat appending the stroke until there are four; 3.2.3 According to the methods described in "3.2.1" and "3.2.2" of this article, when users with different cognitive levels and cognitive preferences have different ways of splitting the first "sub-radical" and the second "sub-radical" of the Chinese character they want to input, correspondingly, users with different cognitive levels and cognitive preferences may have different implementation methods for adding the "first grandson radical" and the "second grandson radical" to the Chinese character. Users should adopt one of the specific implementation methods according to their actual cognitive level and cognitive preference; 3.3 Mapping method of assigning radicals to keyboard keys Each "radical" in the "radical string" obtained by splitting the Chinese character according to the method described in "3.1" and "3.2" of this article is converted into a corresponding key position symbol according to the mapping relationship between the radical and the keyboard key position stated in [Table 1], [Table 2], [Table 3], [Table 4], and [Table 5] in claim 2 (hereinafter referred to as "mapping encoding", and the radical with multiple mapping encoding methods is called "multi-mapping radical"). The "word code" consisting of no more than 4 characters obtained thereby is the "pure form code" of this Chinese character. The specific mapping method is as follows: 3.3.1 Ordinary users should assign each root in the "root string" to the corresponding pinyin initial letter key according to [Table 1]. When the root is a polyphonic root, the user should select the pinyin initial letter of a specific pronunciation of the polyphonic root as its key mapping letter to map and encode it according to his or her own cognitive preference; 3.3.2 Users who prefer the mapping method of root-like letters can assign some or all of the relevant roots in the "root string" to the corresponding letter-like keys according to [Table 2]; 3.3.3 Users who pursue a quick and easy touch typing experience can assign some or all of the relevant radicals in the "radical string" to the corresponding punctuation key positions according to [Table 3]; 3.3.4 Users with a strong southern accent can assign some or all of the relevant roots in the "root string" to the "Southern accent first letter" key according to their own cognitive preferences in the manner described in [Table 4]; 3.3.5 Users who use the numeric keyboard should first map the "radical string" of Chinese characters to the first letter of the pinyin, similar letters, and punctuation key characters according to [Table 1], [Table 2], [Table 3], and [Table 4], and then further assign all the key characters obtained in the previous step to the corresponding numeric keys according to the mapping method described in [Table 5]. The numeric string obtained is the "numeric code" of the Chinese character; 3.3.6 According to the mapping method described in "3.3" of this article, when the user knows that a writing structure is a radical, but does not know which specific key position the radical should be assigned to, on the radical letter punctuation keyboard, "?" can be used as a "universal radical substitute" and the radical can be assigned to the "?" key, and on the radical numeric keyboard, " / " can be used as a "universal radical substitute" and the radical can be assigned to the " / " key; 3.4 Diversified and compatible methods of character encoding According to the methods described in "3.1", "3.2" and "3.3" of this article, users with different cognitive levels and cognitive preferences have different specific splitting methods for the same Chinese character and different specific mapping methods for the same radical. As a result, users with different cognitive levels and cognitive preferences may have different results when splitting the same Chinese character into radical strings and mapping the resulting "pure character codes"; 3.5 Priority sorting method for "pure character code" 3.5.1 Tier division of "pure character codes" and initial priority ranking method The "pure character code" that uses the "common character root" with the largest number of strokes in the Chinese character and uses the pinyin initial letter mapping encoding of the "first pronunciation" of the character root is the "priority character code" (for "common character root" and "first pronunciation", please refer to [Table 1] and its description in "2.1" of this article), and the "pure character code" that uses other character decomposition encoding methods is the "compatible character code"; Each Chinese character has only one "priority code" and can have multiple "compatible codes". The code table containing all the "pure form codes" of Chinese characters is called the "pure form code table". The "pure form code table" contains three "code echelons". All "priority codes" are placed at the front of the "pure form code table" as the first code echelon; all "compatible codes" encoded with the "Nanyin initials" are placed at the end of the "pure form code table" as the optional third code echelon (for "Nanyin initials", please refer to [Table 4] and its description in "2.4" of this article); the remaining "compatible codes" are placed at the back of the first code echelon and in front of the third code echelon as the second echelon code echelon; 3.5.2 Initial Priority Sorting Method of "Pure Character Code" in Each Coding Echelon 3.5.2.1 Calculation formula for "recognition priority" of radical: R = SIGN (INT (0.7 * F)) Formula description (applicable to the whole text): ① "R" is the "recognition priority of the root", ② "SIGN(X)" is the "sign function" (when X<0, X=0, X>0, it returns -1, 0, 1 respectively), ③ "INT()" is the "rounding function" (returns the integer part of the input value), ④ "F" here is the "character frequency" of the root (the "character frequency" of a Chinese character refers to the number of times the Chinese character appears in every 100,000 Chinese characters in ordinary text), ⑤ 0.7 is the first-level character frequency coefficient = 1 / 1.

43. Since the lowest character frequency of the top 3500 Chinese characters is 1.43, the roots with a frequency lower than 1.43 are regarded as low-recognition roots, ⑥ The value range of R: when F>=1.43, R=1, when F<1.43, R=0; 3.5.2.2 Initial priority sorting method for different decomposition encoding methods of the same Chinese character within the same character code echelon On the basis of the initial priority ranking described in "3.5.1" of this article, when a Chinese character has multiple splitting schemes and contains "multi-mapped radicals", the ranking of the "pure character codes" corresponding to each splitting encoding scheme in the same encoding echelon is arranged from large to small according to the "(U)" value calculated by the following formula: U=(R1*B1*(10^8))+(R2*B2*(10^6))+(R3*B3*(10^4))+M1*100+M2*10+M3 Formula description: ①U is the "character splitting encoding scheme priority", ②R1 is the recognition priority of the first root in the root string, B1 is the number of strokes of the first root in the root string, ③R2 is the recognition priority of the second root in the root string, B2 is the number of strokes of the second root in the root string, ④R3 is the recognition priority of the third root in the root string, B3 is the number of strokes of the third root in the root string, ⑤M1 is the "mapping implementation method priority (M)" of the first root in the root string, ⑥M2 is the "mapping implementation method priority (M)" of the second root in the root string, ⑦M3 is the "mapping implementation method priority (M)" of the third root in the root string, ⑧The "mapping implementation method priority (M)" of the root The value of "First degree (M)" is as follows: when the root is assigned to the first letter key of the first pronunciation, M=6; when the root is assigned to the first letter key of the second pronunciation, M=5; when the root is assigned to the first letter key of the third pronunciation, M=4; when the root is assigned to the first letter key of the fourth pronunciation, M=3; when the root is assigned to the letter-like key, M=2; when the root is assigned to the punctuation key, M=1; when the root is assigned to the first letter key of the Nanyin, M=0; ⑨ The following operators are applicable to the whole text: "+" is the addition sign, "*" is the multiplication sign, and "^" is the power operator (for example, 10^2=100, 10^3=1000); ⑩ When there is no nth root in the root string, Rn=0, Bn=0, Mn=0; 3.5.2.3 Initial Priority Sorting Method for Repeated Code Words in the Same Code Echelon On the basis of the priority hierarchical sorting described in "3.5.1" and "3.5.2.2" of this article, Chinese characters and their "pure form character codes" are sorted in the character code echelon according to the composite index of "C+STR((10^6)-INT(0.5+100*F))" (where "C" is the "pure form character code" string and "F" is the frequency of the Chinese character). When a "pure form character code" corresponding to multiple Chinese characters is input, the Chinese character with the same code and higher frequency in the same character code echelon is presented at a front position in the character selection screen; Index formula description (applicable to the entire text): "-" is a "minus sign", "STR()" is a "value to character function" (input a value and return the Arabic numeral character corresponding to the value); 3.5.3 Dynamic ranking method of "pure character code" On the basis of the initial priority hierarchical sorting described in "3.5.1" and "3.5.2" of this article, as the user types in the pure form code and selects to input the corresponding Chinese character, the position of the inputted Chinese character in the candidate word list on the character selection screen will gradually move closer to the first position in the candidate word list on the character selection screen. If the user uses a pure form code to input a Chinese character this time, and the position of the Chinese character in the candidate word list on the character selection screen is the Nth position, then the next time the user inputs the pure form code, the position of the Chinese character in the candidate word list will become the INT(1+K*N)th position. K in the formula is the moving coefficient. The initial value of K is 0.

5. The user can set the value of K between 0 and 1 according to his or her preference. The smaller the K value, the faster the forward movement. 3.6 How to type "pure character code" and select and enter Chinese characters The user should enter the "pure character code" obtained according to the methods described in "3.1" and "3.3" of this article on the character root keyboard (when the user enters the "numeric character code" described in 3.3.5 of this article, the period ""." must be used as the start and end symbol of the "numeric character code": first enter 1 "".", then start entering the "numeric character code", and after entering the "numeric character code", enter another "." to end the input of the "numeric character code"). After the user has entered the "pure character code", the corresponding operation is performed according to the following 6 situations: 3.6.1 When the user has completed inputting a "pure character code" of less than 4 coded characters, and the Chinese character that the user wants to input does not appear on the character selection screen, the user who wishes to input the Chinese character by blind typing without turning the screen can enter the code of the supplementary radical in the "pure character code" obtained by the methods described in "3.2" and "3.3" of this article on the radical keyboard (the additional code of the part of the "pure character code" of the Chinese character obtained by the methods described in "3.1" and "3.3" of this article); 3.6.2 When the user has finished inputting the "pure character code" and the Chinese character the user wants has not yet appeared on the character selection screen, the user should use the screen-turning key to flip through the contents on the character selection screen until the Chinese character appears on the character selection screen or the user has flipped through all the contents of the candidate word list on the character selection screen; 3.6.3 When the Chinese character that the user wants appears on the character selection screen, the user should enter the serial number corresponding to the left side of the Chinese character on the character selection screen to select and enter the desired Chinese character; 3.6.4 When the Chinese character the user wants appears in the first position on the character selection screen, the user can press the space bar to enter it; 3.6.5 When the Chinese character desired by the user appears on the character selection screen in the touch screen, the user should click on the Chinese character on the character selection screen in the touch screen to select and input it; 3.6.6 When the user flips through all the candidate words and characters on the character selection screen using the screen flipping key but still cannot find the Chinese character they want to input, it indicates that the encoding input by the user contains a special character splitting encoding method that is incompatible with the present invention. The user should use the code lookup dictionary built into the application software of the present invention to query the preferred character splitting encoding method for this Chinese character (for the usage of the code lookup dictionary, see the content of "3.8" in this article). After the user has learned the preferred character splitting encoding method for this Chinese character in the code lookup dictionary, they should restart the operation described in "3.6" of this article; 3.7 Character splitting encoding method for improving the blind typing rate of Chinese characters According to the preferred grading and sorting method of Chinese characters and their "pure shape character codes" stated in "3.5" of this article, the "pure shape character codes" obtained using the following character splitting encoding methods can increase the blind typing rate: 3.7.1 All recognizable roots can be used, and split the character according to common sense: The user does not need to distinguish which inner characters and radicals are available roots and which are not. All inner characters, radicals, and strokes that the user recognizes are available roots, and all character splitting methods that conform to the common sense of ordinary users are available character splitting methods; 3.7.2 Do not split recognizable roots into fine parts, and preferentially select the larger root: When the user extracts the "sub-root" from the Chinese character they want to input (for the definition of "sub-root", see the content of " 3.2.1" in this article), for all "roots" that the user recognizes, they should be extracted as a whole without splitting. The user does not need to distinguish whether the root is an "inner character" or a "radical", and only needs to extract the "root" with a larger number of strokes that they recognize. If a larger inner character or radical with a larger number of strokes can be extracted, then do not extract a smaller inner character or radical with a smaller number of strokes. If an inner character or radical can be extracted, then do not extract a single stroke. Only when the user cannot find a recognizable inner character or radical do they extract a single stroke; 3.7.3 For unrecognizable roots, incorporate them into a larger root or split them into smaller parts: When the user encounters an unrecognizable root during the character splitting process, if it can be incorporated into a larger "root" that the user recognizes, then incorporate it into the larger "root" and extract it together. Otherwise, split it into smaller "roots" that the user recognizes, and then extract these smaller "roots" one by one according to the stroke order. When the user cannot find a recognizable inner character or radical, then extract the first single stroke of the remaining part of the Chinese character being split; 3.7.4 For single-structure characters, simulate writing according to the stroke order and extract the larger root: When there are no obvious discrete components inside the Chinese character being split, according to the writing stroke order, simulate writing each stroke of the remaining part of the Chinese character from the beginning. The root with the largest number of strokes that the user recognizes and is written first is the root recommended to be extracted by the present invention. For example: the character "果" is recommended to be split into "日木", the character "大" is recommended to be split into "丶", the character "男" is recommended to be split into "田力", the character "王" is recommended to be split into "干一", the character "田" is recommended to be split into "冂土", and so on; 3.7.

5. When two radicals inside a Chinese character share the same stroke, the shared stroke should be preferably broken in the middle to split out two relevant radicals respectively. When the user does not break the shared stroke but splits one of the relevant radicals instead, the desired Chinese character can still be typed, but the resulting code does not enjoy the privilege of blind typing with same-code priority. For example: "戠” is preferably split into "音戈” (but it can also be split into "音 丿、” or "立日戈” or "立日 丿”);"重” is preferably split into "千里” (but it can also be split into "千日二” or "丿一日土” or "丿一申二” or "丿一里”);"成” is preferably split into "万戈” (but it can also be split into "厂 丿” or "万 丿丶”); 3.7.6 For "polyphone radicals", users should use the pronunciation they are most familiar with to implement mapping and encoding for the radicals: For the various encoding methods of Chinese characters containing polyphone radicals, the present invention will put the encoding of the pinyin initials of the most popular pronunciation among the general public at the highest priority position in the code table, so users can use the pronunciation they are most familiar with to map and encode the polyphone radicals, which will make it easier to obtain the convenience of priority blind typing when typing; 3.7.

7. Use the quick mapping method: Use the similar letter iouv in the "radical→similar letter mapping method" described in [Table 2] of claim 2 and the "radical→punctuation mark mapping method" described in [Table 3] to implement mapping encoding on the radical, which can obtain more blind typing opportunities than using other mapping encoding methods; 3.7.

9. Moderate splitting of Chinese characters can improve touch typing. For example, if a Chinese character is split into 4 to 5 radicals, the resulting code can have a higher touch typing rate. For example, after completely splitting the character "捡" into "扌佥", the mapped code is {sq}. When typing {sq} for the first time, the character "捡" appears at the 20th position on the character selection screen. And when completely splitting the character "捡" into "扌人一 ” and mapping the resulting code {srhx}, when typing {srhx} for the first time, the character "捡" appears at the 1st position on the character selection screen, enabling touch typing for the first time; 3.7.9 The more times a user uses a certain "pure character code" to input the Chinese character, the greater the chance that the user will use the "pure character code" to blindly input the Chinese character (see the dynamic ranking method of "pure character code" described in "3.5.3" of this article for details); 3.8 How to use the code query dictionary: ① In the input method running environment of the present invention, use the full spelling or stroke word encoding provided by the present invention to input the Chinese characters that the user wants to check the code, ②When the Chinese character appears on the character selection screen, move the mouse cursor over the Chinese character on the character selection screen and keep it there for 1 second, and a small pop-up window will pop up next to the character selection screen. ③ Click the blue font "View detailed explanation" at the bottom edge of the small pop-up window, and the code-checking dictionary of the present invention will pop up in a pop-up window, and the detailed character-decomposition encoding method of the Chinese character will be displayed in the code-checking dictionary pop-up window; ④ The user can use the scroll key to browse the relevant content displayed in the code-checking dictionary, and can also directly enter other words that they want to search in the "code-checking dictionary" input box to search for the priority decomposition encoding method of other words in the code-checking dictionary.

4. According to claim 1, the coding input method of Chinese phrases is characterized in that: 4.1 Method for splitting Chinese characters in a phrase and encoding root mapping In the process of encoding the phrase, the relevant Chinese characters in the phrase are split according to the method described in "3.1" of this article, and the relevant radicals are assigned to key symbols according to the radical mapping method described in "3.3"; 4.2 Methods for encoding phrases The phrase is encoded according to the following "phonetic-graphic combined encoding" encoding formula, and the resulting "word code" consisting of letters and punctuation marks is the "phonetic-graphic code" of the phrase: 4.2.1 Sound and shape joint coding formula: The phonetic and morphological code of a two-character phrase = W1p1+W2p1+;+W1s1+W2s1+W2s2 The phonetic and morphological code of three-character phrase = W1p1+W2p1+W3p1+;+W1s1+W2s1 The phonetic and morphological code of the four-character phrase = W1p1+W2p1+W3p1+W4p1+;+W1s1 The phonetic and morphological code of the five-character phrase = W1p1+W2p1+W3p1+W4p1+W5p1+; The phonetic and morphological code of N-word phrase (N>=6) = W1p1+W2p1+W3p1+W4p1+W5p1+W6p1 4.2.2 Explanation of the sound-shape joint coding formula: 1) W1p1 = the first pinyin letter of the first Chinese character in the phrase, W2p1 = the first pinyin letter of the second Chinese character in the phrase, W3p1 = the first pinyin letter of the third Chinese character in the phrase, W4p1 = the first pinyin letter of the fourth Chinese character in the phrase, W5p1 = the first pinyin letter of the fifth Chinese character in the phrase, W6p1 = the first pinyin letter of the sixth Chinese character in the phrase; 2) W1s1 = mapping key character of the first radical of the first Chinese character in the phrase, W2s1 = mapping key character of the first radical of the second Chinese character in the phrase, W2s2 = mapping key character of the second radical of the second Chinese character in the phrase; 3) "Sound-shape word code" consists of two parts: "Pinyin initial letter string" and "radical mapping key string" (hereinafter referred to as "shape code part"). The semicolon ";" is used as a separator between the two parts and appears in the instance of "Sound-shape word code". The plus sign "+" is only used as a connector in the formula and does not appear in the instance of "Sound-shape word code"; 4.3 Diversified and compatible methods of "phonetic-morphological code" 4.3.1 According to the methods described in "4.1" and "4.2" of this article, users with different cognitive levels and cognitive preferences have different specific decomposition methods for each relevant Chinese character in a phrase and different specific mapping encoding methods for each relevant root. As a result, users with different cognitive levels and cognitive preferences may have different results for the "shape code part" of the "sound-shape word code" obtained by decomposition encoding operations on the same phrase; 4.3.2 According to the methods described in "4.1" and "4.2" of this document, when there are multiple split character encoding methods for Chinese characters related to the "shape code part" of the phonetic-shape word code in a phrase, and when the root related to the "shape code part" of the phonetic-shape word code in a phrase is a "multi-map root" (for "multi-map root", refer to the content of "3.3" of this document), the present invention creates multiple "phonetic-shape word codes" corresponding to the multiple split character encoding methods for the phrase to adapt to the different split character encoding preferences of users with different cognitive levels and cognitive preferences; 4.4 Encoding rules for "digital code" of phrases After obtaining the "phonetic-morphological code" of the phrase according to the methods described in "4.1", "4.2" and "4.3" of this document, the letters and punctuation marks in the "phonetic-morphological code" are further mapped and converted into Arabic numerals according to the mapping relationship described in [Table 5] in claim 2, and the implementation form of the "phonetic-morphological code" of the phrase on the numeric keyboard is obtained: "numerical code"; 4.5 Priority sorting method for "phonetic-morphological code" 4.5.1 Tier division of "phonetic-morphological code" and its initial priority ranking method In the process of character splitting, the character splitting scheme using the "common root" with the largest number of strokes in the Chinese character, and the phonetic-morphological code encoded by mapping the pinyin initials of the "first pronunciation" of the root is the "priority code" (for "common root" and "first pronunciation", please refer to [Table 1] and its description in "2.1" of this article), and the phonetic-morphological code using other character splitting encoding methods is the "compatible code"; Each phrase has only one "priority code" and can have multiple "compatible codes". All the phonetic and morphological codes are divided into three "code echelons". All "priority codes" are placed at the front of the phrase code table as the first code echelon. In the mapping and encoding process of the shape code part of the phonetic and morphological code, all the phonetic and morphological codes that partially or completely use the "Nanyin initial letter" mapping method are placed at the end of the phrase code table as the optional third code echelon (for "Nanyin initial letter", please refer to [Table 4] and its description in "2.4" of this article). The remaining phonetic and morphological codes are placed at the end of the first code echelon and in front of the third code echelon as the second code echelon. 4.5.2 Initial Priority Sorting Method for Repeated Code Phrases in the Same Code Tier On the basis of the initial priority ranking of the "phonetic-morphological code" described in "4.5.1" of this article, the phrase and its "phonetic-morphological code" are sorted according to the composite index of "C+STR((10^6)-INT(0.5+10*F))" within the word code echelon ("C" here is the "phonetic-morphological code" string; "F" here is the word frequency of the phrase, and the "word frequency" of the phrase refers to the number of times the phrase appears in every 10 million phrases in ordinary text on average). When a "phonetic-morphological code" corresponding to multiple phrases is entered, the same-code phrase with a higher word frequency in the same word code echelon is presented at a higher position in the word selection screen; 4.5.3 Dynamic ranking method of "phonetic-morphological code" On the basis of the initial priority ranking described in "4.5.1" and "4.5.2" of this article, as the user types in the phonetic-morphological code and selects to input the corresponding phrase, the position of the input phrase in the candidate word list on the word selection screen will gradually move closer to the first position in the candidate word list on the word selection screen. If the user uses a phonetic-morphological code to input a phrase this time, and the position of the phrase in the candidate word list on the word selection screen is the Nth position, then the next time the user types in the phonetic-morphological code, the position of the phrase in the candidate word list will become the INT(1+K*N)th position. K in the formula is the moving coefficient. The initial value of K is 0.

5. The user can set the value of K between 0 and 1 according to his or her preference. The smaller the K value, the faster the forward movement. 4.6 Method of typing "phonetic-morphological code" and selecting and inputting phrases The user types the "phonetic-graphic code" described in "4.2", "4.3" and "4.4" on the root keyboard (after typing the "numeric code" described in "4.4", a period ". " must be typed as the end symbol of the "numeric code"), and after typing the "phonetic-graphic code", the user performs corresponding operations according to the following five situations: 4.6.1 For some high-frequency phrases, you only need to enter the 3 to 5 digits before the "phonetic-morphological word code" and the corresponding high-frequency phrase will appear on the word selection screen. At this time, you can follow the methods described in "4.6.3", "4.6.4" and "4.6.5" of this article to implement the operation; 4.6.2 When the phrase that the user wants to input does not appear on the word selection screen, the user must use the scroll key to scroll through the contents on the word selection screen until the phrase appears on the word selection screen or the user scrolls through all the contents of the candidate word list on the word selection screen; 4.6.3 When the phrase the user wants to enter appears on the word selection screen, the user should type the corresponding serial number on the left side of the phrase on the word selection screen to select and enter the phrase the user wants; 4.6.4 When the phrase that the user wants to enter appears at the top of the candidate word list on the word selection screen, the user can press the space bar to enter it; 4.6.5 When the phrase that the user wants to input appears on the word selection screen in the touch screen, the user should click on the phrase on the word selection screen in the touch screen to input it; 4.6.6 When the user uses the screen-turning key to flip through the entire list of candidate words on the word selection screen and still cannot find the phrase he wants to enter, it means that the phonetic and morphological code entered by the user contains a special character splitting encoding method that is incompatible with the present invention. The user should use the code-checking dictionary built into the application software of the present invention to query the preferred character splitting encoding method of the phrase (for the usage of the code-checking dictionary, see the content of "3.8" of this article for details); after the user has learned the preferred character splitting encoding method of the phrase in the code-checking dictionary, he should restart the operation described in "4.6" of this article; 4.7 Character-splitting encoding method to improve the blind typing rate of phrases According to the method for sorting phrases and their "sound-shape word codes" stated in "4.5" of this article, in the process of "sound-shape word codes" for the phrases, the "character splitting encoding method for improving the blind typing rate of Chinese characters" stated in "3.7" of this article can improve the blind typing rate of the phrase and its "sound-shape word codes".

Images