JP4956181B2 - Character input device and method using keypad - Google Patents

Description

  The present invention relates to a character input device and method for a keypad, and more particularly to a character input device and method for a keypad having a small number of keys such as a telephone dial.

  With the development of mobile communications, a function for transmitting and receiving digital information such as characters with a portable terminal centered on a voice call has been added. Therefore, initially, the keypad provided in the portable terminal for the purpose of inputting the telephone number includes means for inputting characters.

  However, since the size of the keypad used as the input means of the portable terminal is gradually reduced, the number of buttons included in the keypad has its limit. On the other hand, the characters of each language greatly exceed the number of 12 keys included in the keypad. Therefore, in order to input one character using the telephone keypad, the character must be expressed by combining one or more buttons on the keypad.

  The present invention provides improvements on the extension of the applicant's previous applications (application numbers PCT / KR00 / 00601, PCT / KR01 / 00076, PCT / KR01 / 02267, PCT / KR02 / 01355, PCT / KR03 / 01433, etc.). Present. Although some of the contents of the prior application are presented together for the purpose of explaining the present invention, it should be clearly stated that necessary parts not presented in the present application should be referred to the contents of the prior application.

  Since it is indispensable to give a specific example to explain the characteristics of the present invention, the detailed contents will be described through examples.

(Example)
First, supplementing the contents of the prior application for each language is as follows. In the following, it is obvious that the content that can be applied to other languages among the content that is referred to in one language can be applied to other languages without special mention in other languages.

1. Common Supplementary Items 1.1 Scope of Keypad Application in Prior Application and Present Invention The keypad presented in the prior application and present invention is a mobile terminal, a numeric keypad of a standard keyboard, and a key configured as software on a screen. It is obvious that it can be applied to all fields of telephone dials such as pads or door locks. In addition, the numeric keypad provided in the standard keyboard is different in the arrangement of the numeric buttons from the keypad presented in the previous application and the present invention, but the arrangement on the keypad buttons in the previous application and the present invention is provided in the keyboard. Obviously, it can also be applied to keypads. For example, the character arranged in the [1] button in the prior application and the present invention is arranged in the [1] button of the numeric keypad provided in the keyboard, and the character code input, simple code It can be used for use and memorization of various codes.

1.2 Setting of continuous hit delay time and release delay time In languages where consonants and vowels appear alternately, such as Korean and Hindu, a pair of representative consonants and representative vowels is assigned to each button. Applying the method of selecting vowels with two strokes, the algorithm is implemented so that the same two keystrokes input within a certain delay time (for example, 0.1 seconds) interval are recognized as vowels primarily. Then, it can be efficiently implemented while reducing the refusal feeling of the user. This delay time should be determined in consideration of the time interval that is universally pressed for a certain time when the same button is continuously pressed. Such a delay time is referred to as “continuous hit delay time” for convenience. Further, it is possible to implement such that two hits of the same button input at intervals of a certain delay time (for example, 1 second) are recognized as two consonants. For the sake of convenience, such a delay time is referred to as a “release delay time”. This can also be applied to pressing the same button three or more times.

として１次的に認識し、仮に１．１秒の遅延時間を伴って入力された場合には１次的に２個の子音

と認識するものである。もし、０．５秒の時間遅延を伴って入力された場合には、該当言語の子音と母音の出現構造を通じて１つの母音が入力されたのか、又は二つの子音が入力されたのかを判断できる。また、０．０８秒又は１．１秒の遅延時間を伴って入力された場合にも、該当言語の子音と母音の出現構造を通じて１つの母音が入力されたのか、又は二つの子音が入力されたのかを最終的に判断することができる。 For example, if [1] + [1] is input with a delay time of 0.08 seconds in FIG. 4A or FIG.

If the input is performed with a delay time of 1.1 seconds, two consonants will be primary.

It is something to recognize. If it is input with a time delay of 0.5 seconds, it can be determined whether one vowel is input or two consonants are input through the appearance structure of the consonant and vowel in the corresponding language. . In addition, even when a delay time of 0.08 seconds or 1.1 seconds is input, whether one vowel is input through the appearance structure of a consonant and a vowel in the corresponding language, or two consonants are input. It can be finally determined whether it was.

  Conventionally, only one type of a certain time interval (for example, 1 second) is set, and if the same button is continuously input within the corresponding time, it is recognized as two consecutive strokes, and the same time is delayed after the corresponding time. When two keystrokes were entered, each one was recognized as one stroke. The difference is that a reference time delay value (for example, 0.1 second) determined to be two consecutive strokes and a time delay value (for example, one second) determined to be one stroke each are different.

  In general, it can be said that the iterative selection method is superior in terms of convenience of input among the partial whole selection method and the iterative selection method of the prior application. Therefore, this means that the advantage of the iterative selection method (simpleness and convenience of input rules) can be utilized, and the ambiguity can be largely avoided by using the structure of a specific language in which consonants and vowels appear alternately, There is a merit that the implementation of the algorithm can be simplified by setting the delay time for repeated hits and the delay time for takeoff differently so that the user can specify them.

2. Multi-dimensional Cross Control Processing Method FIG. 2-2 is an example in which the alphabet of the 50th diagram “A” stage is assigned to each button in the order of the dictionary as the representative alphabet. In the case of the previous application, the 4th and 5th controls were additionally assigned to the control buttons to which the 2nd and 3rd controls were assigned. Here, the number of strokes is minimized and the control buttons are maximized. An example of assigning only the second and third controls for limited use is shown.

  The method for inputting the subsequent letter in FIG. 2-2 is as presented in the previous application. For example, if a post-control input is applied, “I = A + [*]”. Next, the input method of the long sound, the muddy sound, and the reflex sound which are the deformation character of each character is shown. In FIG. 2-2, all the buttons that can be used as control buttons ([*] button, [#] button) are used as subsequent control buttons, and therefore there is no control button for a modified letter. However, if the basic character of the target character is the subsequent character, the subsequent control button that is not used as the subsequent control button after the input of the subsequent character is used as a modified character control button (this is referred to as the “opposite control button” for convenience). Can be entered.

  For example, “I = I +“ opposite control button ”= A + [*] + [#]”, “go = ko +“ opposite control button ”= ka + [#] + [#] + [*] "become. This applies to the control processing method by regarding “I” as a modified letter of the basic letter “I”. In order to input “Yes”, the opposite control button that is not used as a subsequent control button is used as a modified letter control button. It is an example utilized as. In the above example, the skip control processing method of the prior application can also be applied here. From the point of view of the chain-type control processing method, by pressing the opposite control button once, the transformation mother (for example, “ko”) of the entered character (for example, “ko”) is entered. It will be understood that the “next control” is selected.

  There are two types of moji characters in the line: muddy sound and anti-turbid sound. For muddy sounds, the opposite control button is pressed once, and for muddy sounds, the opposite control is pressed twice. That is, “bu = fu +“ opposite side control button ”= has + [*] + [*] + [#]”, “pu = fu +“ press opposite side control button twice ”= has + [*] + [*] + [#] + [#] ".

  Next, when the basic character of the target character is the representative character, the above method cannot be applied as it is. In this case, the modified character of the representative character is additionally set as the subsequent character to any button of the subsequent control buttons, and the target character can be input.

  For example, when a deformed character (long sound, muddy sound, semi-voiced sound) of the representative character is input using the [*] button, “a = a + [*] + [*] + [*]”. That is, the modified alphabet (long sound, muddy sound, semi-voiced sound) of the representative character is regarded as the fourth subsequent character that can be input using the [*] button. The above contents are summarized as shown in the following table.

  In this way, using the opposite side control button to expand the application of the control button and using it for more characters or other input, for convenience, "cross control method" or "zigzag control processing method" or "multidimensional This is referred to as “cross control method”. This can also be applied when using three or more buttons as control buttons.

  The remaining column fields other than the first column field in the above table can also be used to input other characters or various symbols. In the above example, the opposite control button is applied once. In other words, when “pu” is input, “pu = fu +“ press the opposite control button twice ”= has [[*] + [*] + [#] + [#]]. The opposite control button ([#] button) is applied only once to the [*] button used as the subsequent control button. This is referred to as “two-dimensional cross control processing” or “two-stage cross control processing” for convenience.

  That is, when only one control button is used to input one character, this is referred to as “one-dimensional control processing”, which is considered to use the control button “one-dimensionally”. On the other hand, in the two-dimensional (cross) control processing method, two other control buttons are used for inputting one character. In other words, when “G” is entered, “G” is regarded as a modified letter of “K”, and “G == +“ Control button on the opposite side ”= or + [#] + [#] + [*]” However, although the opposite control button is already defined as a control button for inputting the first and second succeeding characters “ki” and “ku” of the representative character “ka”, The opposite side control button ([*] button) can be used like a modified letter control button for inputting "G", which is the modified letter of this letter. This is because one control button is used one-dimensionally for one character input, whereas two control buttons are used “two-dimensionally”.

  “Pu = fu +“ press the opposite control button once ”+“ press the opposite control button once against the previous control button ”= has + [*] + [*] + [#] + [* ] ".

  That is, when inputting “bu”, if it is not possible to input the modified letter by repeatedly using the finally used control button (eg, [#] button), the same method is used for the opposite control button. The opposite control button can be used again to enter another deformed or subsequent character. This is referred to as “three-dimensional cross control processing” or “three-stage cross control processing” for convenience. This means that by using the cross control processing method, the number of input characters that can be input in the keypad or related to other character inputs can be expanded infinitely. Furthermore, it has a great meaning that a succeeding character or a modified character having a relationship with a character input in advance can be naturally input by a control processing method.

  FIG. 2-3 is a graph showing an example of “ha line” in which two types of deformed characters (turbid sound and anti-turbid sound) exist. In Fig. 2-3, "..." is displayed in a light color in circles. This is because you can enter other characters or symbols in the blanks in the table above, and enter them as needed. This means that things (characters or symbols) can be input by a multidimensional control processing method. In FIG. 2-3, the [*] button and the [#] button are used in the right angle direction.

  In the previous application, the second, third, fourth, and fifth controls, which are not representative characters, are assigned to the [*] button, and the [#] button is a modified character such as long sound, muddy sound, and muddy sound. Compared with the method of using as a control button for inputting, it can be seen that the two-dimensional control processing is applied also in the method of the prior application. That is, two control buttons are used for inputting “I = I + {Deformation} = I + [*] + [#]”. In the earlier application, the use of the control button is only used as a subsequent control (second, third, fourth, fifth) button, and only as a modified letter (long sound, muddy sound, semi-muddy sound) control button. On the other hand, in the two-dimensional control process of the present invention, after the subsequent character is input, the subsequent control button is further used as a modified character control button. In order to distinguish the use of the control button for the main purpose (for example, the subsequent control button) as the modified letter control button from the two-dimensional control processing method of the previous application, the “two-dimensional reuse control processing method” is used for convenience. Or “two-dimensional multipurpose control processing method”.

  The method of the prior application has the advantage that the input method is simple and consistent, but has the disadvantage that the number of input strokes is relatively large. Fig. 2-4 shows the case where the control button is used only as a subsequent control (second, third, fourth, fifth) button and only as a modified letter (long sound, muddy sound, reflex sound) control button in the previous application. Is shown in a graph. Differences from the present invention can be understood from FIGS. Also in FIGS. 2-3 and 2-4, the portion displayed in a light color can be expanded and applied.

  In the above, an example of post-control input in the application of the cross-control processing method is shown. However, as in the previous application, the control-destination input or the post-input is applied in the application of the control processing method, the input by the control-destination input is also possible here. is there. For example, “I =“ opposite control button ”+ I = [#] + [*] + A”. However, post-control input is more convenient in applying the cross control processing method.

  This point can be applied not only to Japanese but also to all other languages. For example, if the representative character exclusion control processing method is applied to Arabic and [*] and [#] are used to input subsequent characters, the multi-dimensional cross control processing method is used to input the vowel in the subscript form. be able to. Note that even when the [*] button and [#] button are used as control buttons for inputting subsequent consonants and subsequent vowels in Thai, respectively, it is possible to use a multidimensional cross control processing method for inputting tone codes. it can.

3. Method of utilizing vowel elements in Korean In Fig.4-1 to Fig.4-3, basic consonants and basic vowels in Korean are paired and assigned to each button, and the basic consonants and basic vowels displayed on the keypad are displayed. A method for inputting the value by the iterative selection method will be described. In FIG. 4-1, a treble sound, a hard sound, and an extended vowel are input by a control processing method. In FIG. 4B, only a fierce sound and a hard sound are input by a control processing method. In FIG. In this example, hard sounds, basic vowels, and extended vowels are input by the control processing method.

  FIG. 4-4 is merely an example of a flowchart for implementation, and more efficient programming is possible. For example, in the case of the consonant part at the end in FIG. 4-4, a more efficient implementation is possible by first checking whether the consonant can form a double final consonant.

  The Korean example presented in the earlier application can be applied to other languages having similar characteristics (a structure in which consonants and vowels appear alternately). In the case of other languages, the consonant appearance characteristics of other languages may be reflected.

  For example, when trying to construct a Hindu input system using a pair of basic consonants and basic vowels, it can be embodied in the same way as in Korean, reflecting the characteristics of Hindu. In the case of words, the appearance rules of consonants and vowels are simpler than in the case of Korean, so it can be implemented more easily.

  FIGS. 4-5 to 4-8 are diagrams showing examples of using vowel elements in Korean and using vowel buttons as control buttons. Please refer to the contents of the prior application in relation to the present invention.

4). Simple code utilization method and abbreviated input method / parallel input method 4.1 Character linkage simple code generation method When using an information communication terminal to access an information system, it is necessary to input a character. Alternatively, it is possible to input a character code to be input as a numeric code. In many cases, a miniaturized information communication terminal employs an ordinary keypad interface. The codes here refer to all types of codes. Typical examples include telephone numbers, securities line (listed company) codes, city codes, department codes, electric railway station codes, bank codes, etc. Very many. The advantage of coding and inputting various names is that the input can be simplified.

  Information communication terminals include all forms of information communication terminals such as PCs, mobile communication devices, smartphones, PDAs, two-way character transmission / reception devices, ATM machines (automatic deposit / withdrawal machines), and communication like electronic notebooks. Covers terminals with no function. The information system includes not only a system that can be visually accessed through a GUI, but also all forms of systems such as a system that can be touched only by voice, such as ARS. The system means a server-side system in a narrow sense, and includes client software of a terminal corresponding to the server system in a broad sense.

  It can be usefully used for memorizing various codes by using the characters arranged on the keypad. When used for memorizing codes, there can be simple naming, initial naming, and full naming methods. The contents are briefly summarized as follows.

のコードを

と連関した“１７９９”をシンプルコードとすることである。この場合、

でシンプルコード“１７９９”と連関した

を強調して表示することによって特定単語のシンプルコードをさらに容易に分かるようにでき、さらに、そのような語句からシンプルコードを抽出することもできる。シンプルネーミングの場合、コード化しようとする単語に属するいずれかの字母と連関した数字をコードに決めることであるので、例としてあげた

のコードを必ずしも“１７９９”とする必要はない。例えば、

のシンプルコードは

と連関した値をシンプルコードとできるように、

の場合の他の例をあげると、

と連関した“１７２９”とすることもでき、

と連関した“１７４９”とすることもでき、

を構成する全体字母と連関した“１３２９４２９３”とすることもできる。語句を構成する全体字母と連関したコードをシンプルコードとすることを便宜上“全体連関シンプルネーミング”といい、特定語句の一部字母と連関してシンプルコードを指定することを便宜上“一部連関シンプルネーミング”と呼ぶ。いずれの場合でもシンプルネーミング（すなわち、シンプルコード）は語句単位でその語句を構成する字母に連関してコードを指定することと言える。これは韓国語以外の言語にも同様に適用できる。例えば、“captain”で子音“CPTN”と連関する“２７８６”をシンプルコードにできるが、これは一部連関シンプルコードであり、これを特に便宜上“子音連関シンプルコード”と呼ぶことにする。 Simple naming refers to designating a number associated with a word or phrase to be coded (referred to as a word or phrase, hereinafter simply referred to as a “phrase”) in the code. For example, in case of Korean, company name code is based on Fig. 4-2.

Code

“1799” related to the code is to be a simple code. in this case,

And associated with the simple code "1799"

By emphasizing and displaying, it is possible to more easily understand the simple code of a specific word, and it is also possible to extract the simple code from such a phrase. In the case of simple naming, the code is a number associated with one of the letters belonging to the word to be coded.

Is not necessarily set to “1799”. For example,

The simple code of

So that the value associated with can be a simple code,

Another example of

Can be set to “1729”

It can also be “1749” in conjunction with

"13294293" associated with the overall character that constitutes. For convenience, it is called “whole association simple naming” to make the code associated with the entire character constituting the phrase simple code, and for convenience, it is “partially associated simple” This is called “naming”. In any case, simple naming (that is, simple code) can be said to specify a code in association with the characters that make up the phrase in phrase units. This applies to languages other than Korean as well. For example, “2786” associated with the consonant “CPTN” in “captain” can be made a simple code, but this is a partially associated simple code, which will be referred to as a “consonant-related simple code” for convenience.

  When a consonant-related simple code is generated for a word starting from a vowel such as “escape”, it becomes “727” corresponding to “SCP”, which is the same as the consonant-related simple code of “scape”. Therefore, although a reduced simple code is used, in order to minimize the duplication of the defined simple code and simple code-corresponding words and to improve the convenience in use, a simple code in which the first vowel and the consonant are linked can be considered. This is referred to as “first vowel + consonant reference simple code” for convenience. Similar to other simple codes, such a first vowel + consonant-related simple code has an advantage that a simple code can be mechanically generated for a specific word.

という語句で意味を最もよく含蓄している字（音節）を活用して

にマッピングされる“８３１４”をシンプルコードにできる。英語の場合、先出願で例としてあげた“date tonight”で音価を有しているｄ、ｔ、ｔ、ｎに関連する“３８８６”をシンプルコードにできる。 Not only words but also phrases can be coded by simple naming. Taken as an example in an earlier application

Using the most commonly implied characters (syllables)

“8314” mapped to can be made a simple code. In the case of English, “3886” related to d, t, t, and n having a sound value in “date tonight” given as an example in the previous application can be made a simple code.

の場合、音節基準イニシャルネーミングによるイニシャルコードは各音節（字）の子音初声に関連する“１７９９”となる。音節基準イニシャルネーミングの場合も韓国語以外の言語に対しても適用できる。例えば、英語の場合、“entertainment”という単語の音節基準イニシャルネーミングによるイニシャルコードをｅ、ｔ、ｔ、ｍに関連する“３８８６”にできる。音節基準イニシャルネーミングは１つの音節が１つの字を構成する韓国語の場合により有用に使用され得る。韓国語だけではなく、韓国語と同様に１つの音節が１つの字になる中国語、日本語などの他の言語に対しても有用に使用することができる。中国語の場合、北京（Beijing：最初の‘ｅ’には４声調符号が、最後の‘ｉ’には２声調符号が付記される）の音節基準イニシャルコードは、図１−１を基準としてはｂ、ｊに関連する“２５”となり、図１０−１〜１０−４を基準としては“１４”となる。 Initial naming is a special case of partially linked simple naming. In the case of Korean, there can be a method in which initial naming is based on syllables (characters) and a number that is mapped to the initial voice (first consonant) is designated as a code. This is called “syllable (character) reference initial naming” for convenience.

In this case, the initial code by the syllable reference initial naming is “1799” related to the consonant initial voice of each syllable (character). The syllable reference initial naming can be applied to languages other than Korean. For example, in the case of English, the initial code by the syllable reference initial naming of the word “entertainment” can be “3886” related to e, t, t, m. Syllable reference initial naming can be more usefully used in the case of Korean where one syllable constitutes one letter. Not only Korean but also other languages such as Chinese, Japanese, etc. in which one syllable becomes one character as in Korean can be usefully used. In the case of Chinese, the syllable reference initial code in Beijing (Beijing: the first 'e' is a four-tone code and the last 'i' is a two-tone code) is based on Fig. 1-1. Becomes “25” related to b and j, and becomes “14” with reference to FIGS.

”という句節で各単語の初の字母である

と関連した“８１”をイニシャルコードにできる。英語の場合、先出願で例として“dance with the wolf”という語句に対する単語基準イニシャルネーミングのイニシャルコードはｄ、ｗ、ｔ、ｗと関連した“３９７９”がイニシャルコードとなる。単語基準イニシャルネーミングは句節全体に対してコードを付与する場合にすべての言語で有用に使用され得る。 Similarly, initial naming is possible for a phrase, but the example given in the earlier application is “

”Is the first letter of each word

"81" related to the initial code can be used. In the case of English, the initial code of the word reference initial naming for the phrase “dance with the wolf” as an example in the earlier application is “3979” related to d, w, t, and w. Word-based initial naming can be usefully used in all languages when assigning codes to the entire phrase.

  Simple code (whole association simple code, partial association simple code) and initial code (syllable reference initial code, word reference initial code) are collectively referred to as “simple code (ie, simple code in a broad sense)” or “abbreviated code” for convenience. I will call it. The whole-associative simple code, consonant-related simple code, syllable-based initial code, and word-based initial code have regularity especially in their code generation, so they can be used in general, and others can use simple codes with such rules. Even when it is generated, it can be easily used.

の場合、図４−２を基準に部分全体選択方法（先出願で言及）によるフルコードは“７７４５８８８９４４”となる。基準反復選択方法（先出願で言及）によるフルコードは“７４４８８８４”となる。もし、図４−２と異なるキーパッドを適用したり、他の字母入力方法を適用したりする場合には、それに従うフルコード値が存在する。 Full naming is an input value of a word to be encoded by a specific character input method. Therefore, this can be changed depending on the applied character input method, and it can be said that a numerical value corresponding to a word or phrase which is a unit of character is used as a code. Taken as an example in an earlier application

In this case, the full code based on the partial whole selection method (referred to in the previous application) based on FIG. 4-2 is “7744588884”. The full code according to the standard iterative selection method (referred to in the previous application) is “7448884”. If a keypad different from that shown in FIG. 4-2 is applied or another character input method is applied, there are full code values corresponding to the keypad input method.

4.2 Existing Chinese Kanji Input Method The Chinese input method is the same as the Kanji input method in Hangul, where you can input English characters corresponding to Chinese pronunciation and convert them using the “Kanji conversion key”. When is displayed, it is a common form to select and input the corresponding kanji. That is, when a Chinese pin sound is input in Roman characters, the system searches for the corresponding Chinese character and provides it to the user. Refer to Fig. 5-1. In Chinese, the full code can be determined based on the English pronunciation of Chinese.

4.3 Uniqueized Simple Code If simple code can be decrypted on the client side (that is, if the client side has a specific phrase and a simple code value corresponding to that specific phrase), it is entered The words corresponding to simple codes can be transmitted to the server side. Even if simple code can be decoded on the client side, if simple code is required on the server side depending on the nature of the application, the simple code itself (ie, a list of numbers) is transferred to the server side, and this simple code is interpreted on the server side. Good. This means that simple code decryption can be done on the client side or on the server side.

  When a simple code is defined and used for a large number of words / phrases, there can be many words / phrases corresponding to the same simple code. Such ambiguity that can occur between simple codes is called “secondary ambiguity” for convenience. In this case, the system can store a simple code having the same value by adding a serial number to form a unique code value. However, since the user mainly uses a simple code associated with a specific phrase, secondary ambiguous Occurs. In such a case, it is natural that the system should recommend such words to the user according to the priority of use. When the same simple code is used while being different words, the serial number can be used as a priority for recommending to the user by adding the serial number to the simple code according to the priority according to the frequency of use of the word in the system. Of course, the serial number does not necessarily have to be assigned, but the system may have priority information separately, and only an example in which it can be done is shown.

のシンプルコードが音節基準イニシャルコードである“９１９６”であり、

のシンプルコードが音節基準イニシャルコードである“９１９６”であって、同一である場合、各単語の常用頻度によって一連番号を付け、その一連番号をユーザに推薦する優先順位として用いることができる。

の使用頻度が多いとそれに優先順位を付与して

＝“９１９６１”とし、

＝“９１９６２”にできる。同様に先生（xiansheng：母音ａが第１声調）と学生（xuesheng：最初のｅが第２声調）の音節基準イニシャルコードが同じ“９７”であるなら、各単語の使用頻度により一連番号を付け、その一連番号を優先順位として活用することができる。先生（xiansheng）の使用頻度が学生（xuesheng）の使用頻度よりも高くなると、先生（xiansheng）＝９７１とし、学生（xuesheng）＝９７２とすることができる。このように一連番号を付加して生成されたシンプルコードを便宜上“ユニーク化シンプルコード”ということとし、一連番号を付けず重複性があるシンプルコードを単純に“シンプルコード”とし、両者をすべて包括して“シンプルコード”という。 For example,

The simple chord is “9196” which is the syllable reference initial chord,

Is a syllable reference initial code “9196” and is the same, a serial number is assigned according to the frequency of use of each word, and the serial number can be used as a priority order recommended to the user.

If the frequency of use is high, give priority to it

= "91961"

= "91962". Similarly, if the teacher's (xiansheng: vowel a is in the first tone) and the student (xuesheng: first e is in the second tone) have the same syllable reference initial code “97”, a serial number is assigned according to the frequency of use of each word. The serial number can be used as a priority order. When the usage frequency of the teacher (xiansheng) becomes higher than the usage frequency of the student (xuesheng), the teacher (xiansheng) = 971 and the student (xuesheng) = 972 can be obtained. Simple codes generated by adding serial numbers in this way are referred to as “unique simple codes” for convenience, simple codes that do not have serial numbers and are simply referred to as “simple codes”, and both are included. This is called “simple code”.

  In the teacher (xiansheng) and student (xuesheng) examples, if the user enters only “97”, the system will provide the user with the teacher (xiansheng) and student (xuesheng) and the user will choose between Can do. If the user recognizes it from the beginning and inputs “971”, the system can recognize it as “xiansheng”.

  Here, by highlighting “x, s” used as the basis of the syllable reference initial code, a better visual effect can be given to the user. One way to highlight it is to display it in capital letters, like teacher (XianSheng). The system can also analyze the simple code “97” from the words displayed in capital letters.

4.4 Use cases of simple codes The following are examples of simple codes (syllable reference initial codes) assigned to city names. This can be usefully used in railway information systems and the like.

の音節基準イニシャルコードが同一であるので、システム内ではこれを回避するために連番を付けて

のようにシンプルコードをユニーク化できる。ユーザがシステムに“７８”のみを送出すると、システムでは適切にフィードバックを提供して（例えば、

のリストを提供又は音声で知らせる）、ユーザが

のうちの１つを選択するようにする。ユーザが最初からこれを認知して“７８１”を入力した場合には、システムではこれを

と認識できる。 here

Since the syllable reference initial codes are the same, serial numbers are assigned to avoid this in the system.

Simple code can be made unique like this. If the user sends only “78” to the system, the system will provide appropriate feedback (eg,

(Provide a list of or inform by voice)

One of them is selected. If the user recognizes this from the beginning and enters “781”, the system

Can be recognized.

という単語自体を必要とする場合には、クライアント側でシンプルコード“７８”を解釈して

をサーバ側に転送すればよい。又は、クライアント側でシンプルコードを解釈できる場合でも、アプリケーションの性格によってサーバ側でシンプルコード自体を必要とする場合には、入力されたシンプルコード自体をサーバ側に送出すればよい。 If the server side system does not require the simple code “78”,

If you need the word itself, interpret the simple code “78” on the client side.

Can be transferred to the server side. Alternatively, even if the simple code can be interpreted on the client side, if the simple code itself is required on the server side depending on the nature of the application, the input simple code itself may be sent to the server side.

  To give another example of assigning a simple code (whole association simple code, syllable reference initial code, consonant association simple code, first vowel + consonant association simple code) to the city name, Beijing based on Figure 1-1 The (Beijing) syllable reference initial code is “25” related to b, j, the entire associative simple code is “2345464”, and the consonant-related simple code is “2564” related to b, j, n, g. .

The following are examples of securities items (listed companies). This can be usefully used in various securities information systems.

の場合、音節基準イニシャルコードの基礎として用いられた

を強調して表示することによって、ユーザに視覚的に一層良い効果を与えることができる。 Where, for example,

Was used as the basis of the syllable reference initial code

By emphasizing and displaying, a better visual effect can be given to the user.

Examples of bank codes are as follows. This can be usefully used in ATM machines and various financial information systems.

  Here, when a user inputs a simple code already defined in this way, it is obvious that it can be used for inputting a phrase by interpreting it on the client (terminal) side and providing it to the user. This is called a “shortening input method” and will be described in detail together with a “parallel input method”.

の優先順位が各々１、２位に指定されていたが、特定ユーザが

を選択する回数が顕著に多い場合には、

の優先順位を

の優先順位より高いものに調整できる。前述のように多様な方法があるが、一連番号を変えてこれを実現できる。優先順位情報を別途にシステムで有している場合はその優先順位情報を変更することによって可能である。 4.5 Automatic re-priority specification based on selection frequency

Was assigned to the first and second priority, but the specific user

If the number of times

Priorities for

Can be adjusted to a higher priority. There are various methods as described above, but this can be realized by changing the serial number. If the priority information is separately provided in the system, this can be done by changing the priority information.

を選択する回数が顕著に多いか否かの判断基準はシステムが決定することもでき、ユーザが（再）指定することもできる。このような判断基準の例をあげると、１０回のうちの８回以上既に指定された優先順位と相違に選択すると、既存の優先順位を自動的に訂正するようにできる。オプションに応じてユーザに訂正の如何を聞いて確認を受けることもできる。

The system can determine the criterion for determining whether or not the number of times to select is remarkably large, and can be (re) designated by the user. As an example of such a criterion, if a priority is selected that is different from a priority that has already been designated 8 times or more out of 10 times, the existing priority can be automatically corrected. Depending on the option, the user can be asked for corrections and confirmed.

をそれぞれ“先生（xiansheng）”と“学生（xuesheng）”に変更した場合にも同一に適用することができる。 This can be applied to other languages in the same way,

The same can be applied to the case where each is changed to “xiansheng” and “student (xuesheng)”.

4.6 Simple code automatic specification and simple code-related character highlighting In particular, consonant-related simple codes, syllable-based initial codes, and word-based initial codes have regularity in simple code generation as described above. When a user designates a simple code for a specific word / phrase, if the user inputs a specific word / phrase with a simple code generation rule designated, the corresponding simple code can be automatically extracted and stored in the system. Moreover, the convenience of use can be improved by highlighting and displaying the characters associated with the simple code. In the case of English, a simple code and a related letter are highlighted and displayed using capital letters.

  In the previous application, “shortening input method” and “shortening / full parallel input method” have been presented. A simple code for abbreviated input can be defined by the system, the user can change it, and the user can also specify a simple code for a new phrase.

に対するシンプルコードを生成することにおいて、音節基準イニシャルコードをシンプルコードとして使用しようととする場合、シンプルコード生成モードで

を入力し、“９１９６”を入力しなければならない。 In defining a simple code for a new phrase, it is possible to use simple code generation rules such as a whole association simple code, a partial association simple code, a consonant association simple code, a syllable reference initial code, and a word reference initial code introduced in the previous application. For example, if a user generates a simple code for “dance with the wolf” and wants to use the word reference initial code as a simple code, enter “dance with the wolf” in simple code generation mode and The word reference initial code “3983” must be entered. Similarly,

If you want to use a syllable reference initial code as a simple code in the simple code generation mode,

And “9196” must be entered.

のみを入力することによって

に対するシンプルコードが自動的に“９１９６”に指定されるようにすることができる。同様に、音節基準イニシャルコードを使用することをシステムに設定し、先生（xiansheng）を入力することにより、先生（xiansheng）に対するシンプルコードが自動的に“９７”に指定されるようになる。 However, if the user designates a simple code for a specific phrase, the user can enter the phrase for which the user wants to generate a simple code by storing the type of the desired simple code in the system. A simple code can be automatically specified without entering. Set the system to use the syllable reference initial code in the above example,

Only by entering

The simple code for can be automatically designated as “9196”. Similarly, by setting the system to use the syllable reference initial code and inputting the teacher (xiansheng), the simple code for the teacher (xiansheng) is automatically designated as “97”.

  In the case of English, capital letters can be used by highlighting characters that are to be used as simple codes. Therefore, the simple code specification method can be specified in advance to use uppercase letters as simple codes, and "3886" corresponding to uppercase "DTTN" can be automatically specified as simple codes when "DaTe ToNight" is entered. Or, when “ToNight ShoW” is input, “8679” corresponding to the capital letter “TNSW” can be automatically designated as a simple code.

4.7 Short phrase input method and parallel input method using simple code Enter simple code (initial code is a special case of simple code, so it will be included in simple code unless otherwise specified) Sometimes, the system (client-side system or server-side system) can recognize and process it as a word corresponding to the simple code. Therefore, when the phrase corresponding to the specific simple code is recognized and provided to the user again, it is natural that this can also be used for inputting a word.

  Actually, in foreign character input method, an index with “Whole Linked Simple Code” is stored in the terminal (client side system) for each word, and when the user inputs a code, the corresponding word is assigned a priority by word. The character input system is realized by allowing the target word to be determined by displaying the above. This can be found at http: // www. tegic. com and http: // www. zicorp. com's two internet sites. For convenience, such a method is referred to as an “overall-relation input method” or “foreign method” in the present invention. Alternatively, a typical input system that uses such an approach is “T9” from Tegic and is referred to as “T9 Like Input Method (T9LIM)”. FIG. 5-2 shows an input example in the Tic system of Tegic. As shown in FIG. 5B, when the user inputs “622...” To input meet, the system initially provides “off” to the user, but when “6228” is input, “meet”. "Is displayed.

  Comparing the Tegic and Zi methods with the keypad input method presented by the applicant, the Applicant's character input method assigns a unique code to each character unit, and the target character or target phrase is given by the full code. The foreign method mentioned above assigns a whole associative simple code in units of words, and allows the target word to be input by the simple code.

  The disadvantage of foreign methods is that you can enter only pre-defined words by assigning a code on a word-by-word basis, or use a toggle button to enter a word that is different but not frequently used when it has the same code It is not easy to input by selecting and confirming the target word using the move button, the word that is not the target word can appear temporarily at the time of input, it takes up a lot of memory capacity of the system and requires a lot of cost to implement There is a point.

  Here, it is pointed out that a simple code (partially linked or totally linked) is assigned to a common term phrase (a concept including all common words or common phrase clauses), and a target word can be input using the simple code. The simple code for the common phrase can be provided to the user already defined in the system, and can be arbitrarily specified by the user. Alternatively, the user can arbitrarily modify the simple code already defined in the system. What the user can specify arbitrarily has the merit that the user can easily perceive the simple code value for the specific common phrase.

  In the present invention, a method for inputting a target word / phrase using a simple code (including all partially associated simple codes, all associated simple codes, and initial codes) is called a “shortened input method” for convenience, and a method for inputting a target character using a full code. Is referred to as a “full input method” for convenience. Further, as described later, the abbreviated input method and the full input method can be applied in parallel, and the method of applying the abbreviated input method and the full input method in parallel is referred to as a “shortened / full parallel input method” for convenience. Or, simply called “parallel input method”.

  Among the full input methods that give a unique code for each character unit and allow the target character to be input by inputting the code, the ambiguity generated by using the iterative selection method is particularly convenient for the sake of convenience. It will be called “sa” or “character ambiguity”. On the other hand, in the method of assigning a code to all words and inputting the target word using this code as in the foreign method, there is an ambiguity that there are many different words for the same code. This is called “secondary ambiguity” or “phrase ambiguity” for convenience. In the present invention, the simple ambiguity means primary ambiguity.

  A specific input value is primarily interpreted as a simple code (that is, the abbreviated input method is applied primarily or the abbreviated input mode is applied as the basic input mode), and there is no simple code corresponding to the input value. In some cases, it is possible to construct a scenario such as secondarily recognizing with a full code (that is, secondarily applying a full input method). (Ie, the full input method is applied primarily, or the full input mode is applied as the basic input mode), and if a full code cannot be formed, it is recognized by a simple code (ie, shortened secondary) A scenario such as applying an input method is possible. Interpreting input values primarily with simple codes can be seen as applying the "shortened input mode" as the basic input mode, and conversely, interpreting primarily with full codes is fundamental. It can be seen that the “full input mode” is applied as the input mode. In character input, it is preferable that the user who mainly uses the common phrase is applied the abbreviated input method primarily (that is, the abbreviated input mode is used as the basic input mode), and the user who does not use the abbreviated input method is the primary. Preferably, the full input method is applied (that is, the full input mode is used as the basic input mode).

という単語のシンプルコードを“８７７”とし（音節基準イニシャルコードを使用）、これを入力する時に１次的にフルコードと見なすのでこれを

として認識することができる。これは

などのように２、３番目の字の初声が同一ボタンに対応する場合に生じる。反対に、基本入力モードとして短縮入力モードに設定した場合、入力値に対して１次的にシンプルコードとみなして解釈するので、フルコードが入力された場合にもターゲット単語ではない他の単語として認識されることがある。つまり、これはシンプルコードとフルコードの間の曖昧さであり、これを便宜上“３次曖昧さ”と呼ぶ。 Here, when the basic input mode is set to the full input mode, the input value is interpreted as being primarily regarded as a full code, so even if a simple code is input, the input value forms a full code. Other words that are not the user's desired target word may be entered. For example, when applying the reference iteration selection method as the full input method in FIG.

The simple code of the word “877” is used (uses the syllable reference initial code), and when this is input, it is considered as a full code primarily, so this

Can be recognized as. this is

This occurs when the first and second voices correspond to the same button. On the other hand, when the short input mode is set as the basic input mode, the input value is interpreted as a simple code for the first time. May be recognized. In other words, this is an ambiguity between simple code and full code, which is called “tertiary ambiguity” for convenience.

  Such a tertiary ambiguity can be overcome by a toggle method or a list selection method as in the existing method. Or, there may be other methods, but before inputting an input value that may cause such ambiguity, it is possible to switch from the full input mode to the abbreviated input mode or from the abbreviated input mode to the full input mode on a word basis. Is to be able to do it. This is because, as presented in the applicant's earlier application, an “a / a” control is provided and a word unit hiragana / katakana switching control is provided and selected to input one word of katakana in hiragana mode. It is similar to enabling the user to input one word of hiragana in katakana mode. For example, if the basic input mode is the full input mode, the system recognizes this as a simple code from the beginning and refers to the index for the input value that is input after the “short / full” control is provided and selected. Thus, the target word corresponding to the input value can be provided to the user. Similarly, when the abbreviated input mode is used as the basic input mode, the input value input after selecting the “abbreviated / full” control can be recognized and processed as a full code from the beginning. The “abbreviated / full” control can be pre-inputted or post-inputted to the target word as well, but in this case, it is convenient to pre-input.

  The determination of whether the input value when applying the parallel input method is a full code or a simple code can be performed in units of words in the same manner as referring to the index in order to remove the primary ambiguity. It can be determined whether the input value is a full code or a simple code in the middle of the input as follows.

  When the parallel input method is applied using the full input method as the basic mode, the input value is checked to see if it forms a full code for each code input, and input is performed when it is determined that no full code is formed. The efficiency of the parallel input method can be doubled by determining the value as a simple code, referring to the simple code index, and providing the corresponding phrase to the user. Similarly, when the parallel input method is applied with the abbreviated input method as the basic mode, the input value is compared with the index every time each code is input, and it is confirmed that there is no word that matches the input value in the index. This can be processed by determining that this is a full code of the full input method already set. This means that the tertiary ambiguity that occurs in the middle of input can be removed at the beginning of input by applying the rules for the respective full input methods. This can be similarly applied to the case where the character input method not presented in the previous application is used as the full input method. An example based on the full input method (reference repetitive selection method, partial whole selection method) presented in the previous application is as follows.

  In the case of Korean, for example, when the standard repetition selection method is applied based on Fig. 4-2 with the full input method, the second input value and the third input value of all syllables by the full chord always have the same value. Therefore, in the case of contradicting such conditions, the input value can be determined as a simple code and processed. When it is allowed to process a twin consonant as a combination of basic consonants, it is possible to apply the accompanying criteria. This is applicable in all cases where the reference iteration method is applied.

  Also, in all languages, when the partial whole selection method is applied as a full input method, two input values correspond to one letter, and 2 for the first input value of the two input values. The second input value is restricted. For example, in the case of English, when only the left / right straight line combination is used as shown in FIG. 1-1, if the number is not selected by the whole part selection method, the buttons ([1], [2], [[ 3]) is used as the first input value corresponding to one letter, the values that can be used as the second input value are also the buttons ([1], [2], [3]) in the first row. is there. Similarly, when the second row button ([4], [5], [6]) is input again after [2] + [1] is input, it will come next to form a full code. The possible values are also one of the buttons in the second row ([4], [5], [6]). At the moment when the input value is contrary to this, the system regards the input value as a simple code and can recommend a word corresponding to the simple code to the user according to priority. For example, as shown in FIG. 5C, in the parallel input method in which the partial whole selection method is applied as the full input method and the full input method is applied as the basic input mode, the user inputs a simple code 4357 of “help”. At the moment when the second input value 3 is input, the system recognizes that the input value does not become a full code, and can process the input value as a simple code.

  If the partial whole selection method is applied to FIG. 4-2, the next button that can come in response to the input of the first [1] button to form a full code is [1] or [2] There are only buttons. In contrast to this, it is possible to determine that the input value is a simple code that is not a full code, and refer to the index to recommend the target word corresponding to the input value to the user based on the priority order. When four letters of P, Q, R, and S are assigned to the [7] button as shown in FIG. 1-3, as pointed out in the previous application, one letter among the four letters is combined with the upper and lower adjacent combinations. In this case, when the [7] button is input to form a full code for one character, the button that can be placed next is the button ([7 ], [8], [9]) or only the upper and lower adjacent combination buttons ([4] button), the input value can be processed into a simple code when it is contrary to this. This is applicable to all languages when the partial whole selection method presented in the previous application is applied.

  For example, using the partial code selection method as the full input method with Beijing simple code “25” stored as the syllable reference initial code in the simple code index based on FIG. Then, when the full input mode is applied as the basic input mode and the user inputs “25”, the system first inputs (2) after [2] and then inputs [5]. Recognizes that the input value “25” does not become a full code (because the button for the left / right straight line combination that comes after [2] is one of [1], [2], or [3]) and is simple With reference to the index of the code, “Beijing” corresponding to the input value “25” can be provided to the user. In the case of Chinese, a Chinese character (ie, Beijing) can be provided to the user as a word corresponding to the simple code “25” specifically exemplified (because the Chinese character “Beijing” is the target word). For languages using phonetic characters other than Chinese, the system provides the user with “Beijing” corresponding to the simple code “25”. When there are several words / phrases corresponding to the simple code “25”, they are provided in the form of a list, or the user selects a desired word / phrase by repeatedly pressing a specific button (Toggle method). can do. Reference is made to FIG.

  In the end, the parallel input method as shown in Fig. 5-4 applies the simple code of frequently used words such as "Beijing" to the index, and shortens it to the general full input method. This is because the input method is used without changing the mode so that a word registered as a simple code can be input with a small number of strokes.

  In the case of Fig.4-5, one basic consonant per button is assigned, so when using a syllable standard initial code that can be used in Korean as a simple code, abbreviated input method and full input method are used. Even when applied in parallel, there is a good characteristic that it is possible to input without tertiary ambiguity. In other words, if the user inputs a syllable reference initial code when the parallel input method is applied, the input value cannot form a full code from the second input (when hard sound or intense sound is input by the control processing method). You can search the index and recommend words to your users based on priority. The same applies to full code input.

  As described above, when applying the parallel input method, it is one of the “cores” of the present invention that it is possible to determine whether the input value is a simple code or a full code in the middle of input. This applies not only when using the full input method presented by the applicant, but also when using other full input methods. As an example, there is a method of inputting a character by combining a first button to which each character is assigned and a second button corresponding to the character order of the first button. That is, as shown in FIG. 1-1, the input is “P = [7] + [1]”. In this case, since the second button is one of [1], [2], and [3] buttons in Fig. 1-1, the system simplifies the input value at the moment when the input value violates this rule. It can be processed as a code. In particular, the full input method presented by the applicant has a good characteristic that it can be easily known even in the middle of input whether or not the input value constitutes a full code when the parallel input method is applied as shown in the above example.

  Also, as the applicant mentioned in the earlier application, interpretation of simple code or full code may occur on the client side or on the server side. Then, in looking up an index for overcoming primary ambiguity (character ambiguity) in the previous application, a scenario in which the client side index is referred to firstly and the server side index is referred to secondarily. Applicable in interpreting simple code or full code. Conversely, it is also possible to refer to the server-side index primarily and to refer to the client-side index secondary. In addition, if the input value is first interpreted into simple code, the client side index is referenced first, and the server side index is referenced secondarily, and if the simple code for the input value is not searched, the code is changed to full code again. Interpretation, primary interpretation on the client side, and if there is no such interpretation function on the client side, can be secondary interpretation on the server side. As another example, it is also possible to interpret the input value into simple code temporarily, refer to all the client-side index and server-side index and provide it to the user, and allow the user to select the target word. It is. By applying the interpretation method (simple code, full code) and interpretation location (client side, server side) in this way, similar modifications can be made. In other words, when organized, various combinations are possible in the application of the interpretation method (simple code, full code) and interpretation location (client side, server side). In other words, as shown in FIG. 5-5, for the input value, in the order of “(A) − (B) − (C) − (D)” or “(A) − (C) − (B )-(D) ”, not only in the order of (A), (B), (C), (D) but also in the order of various combinations.

  Advantages of applying the shortened input method and the full input method in parallel are as follows. First, since the full input method can be used, the user can enter all words even if the word does not exist in the dictionary other than the already defined words, and a simple code for the use of the abbreviated input method is used for the regular phrase. It can be specified according to convenience (partially linked / whole linked), and the number of input strokes can be greatly reduced by specifying the partially linked simple code. There is an advantage that the word reference initial code can be given not only to the word but also to the phrase. However, since the foreign method uses a method that uses a word-by-word index to input all words, use the entire associated simple code to minimize the same code being given to different words. I had to.

  The system must also have an “index” that has a specific phrase and a code value for that phrase, but even less capacity than a foreign method that has an “index” for every word. Need a storage location. And this “index” must have in the system an “index” for correct words or an “index” for wrong words for words that may cause ambiguity in order to remove ambiguity. Can be used in conjunction with an index prepared to remove the problem.

  In all languages, the consonant has a note value of a specific word, so the technique of extracting a consonant to form an abbreviation has already been widely used. In the case of English, for example, the consonant “CPT” having a note value in the military term “captain” is extracted and used as an abbreviation, and “PVT” from “private” is used as an abbreviation. “SGT” from “sergent” is used as an abbreviation, “SSG” from “staff sergent” is used as an abbreviation, “SFC” from “sergent first class” is used as an abbreviation, and so on. Of course, “captain” and “private” given as examples include two syllables, but consonants extracted as abbreviations here can be regarded as consonants representing each syllable. Therefore, in the case of “captain”, “278” related to “CPT” can be designated as a simple code.

  In this way, the ability to apply a shortened input method by specifying a partially associated simple chord arbitrarily for an ordinary phrase on the basis of a syllable is significant in addition to reducing input effort. Phonologically, syllables are defined as “psychological entities”. Also, a consonant has a note value in the syllable. When only the vowel “AAI” is extracted from “captain” given as an example, it is almost impossible to analogize “captain”. However, if the consonant “CPTN” or “CPT” is extracted, “captain” can be easily inferred. Even if you remove only the consonants and remove only the consonants from each word in a sentence in English, the original sentence can be inferred. In other words, using a partially associated simple code in association with each consonant that is a syllable subject means that the user can naturally adapt to the abbreviated input method, and can provide convenience to the user.

  In particular, it is common to use abbreviations in English-speaking countries, and in the case of a listed company name, a certain number of abbreviations are specified, so simple codes can be used based on these abbreviations. .

  Making it possible to specify a simple code (partial association, whole association) for a specific phrase in a convenient manner for the user has an advantage of easily storing the code value for the regular phrase. In addition, if a user needs only a few common phrases (for example, using only ten or fewer common phrases), without giving the code related to the common phrase phrases, A simple code such as 1, 2, 3,.

4.8 Grouping of simple codes / corresponding phrases and search range specification When simple codes are specified for a large number of phrases, a lot of duplication may occur. However, grouping phrases corresponding to simple codes, The ambiguity (secondary ambiguity) between the simple codes can be reduced by searching the simple code only for the specific phrase group. A word does not necessarily have to belong to only one group but can belong to various groups.

  For example, simple-named phrases can be grouped into groups such as listed company names, city names, common-word phrases,..., And regular-phrase groups can be grouped again as detailed groups such as social fields, political fields,. In this example, a two-stage tree-type grouping is presented, but a three-stage, four-stage or more-stage tree-type grouping is also possible. Please refer to FIG. When the user (or the system) limits the search range of simple codes to listed company name groups, the system searches for named phrases corresponding to simple codes entered only within the category of listed company name groups when entering specific simple codes. As a result, the secondary ambiguity can be reduced. Similarly, if the search range is an ordinary term phrase, all the detail groups below the regular term phrase group can be included in the search range. If the social field group of the common phrase group is designated as the search range, only the group including the social field and thereafter (in the group structure of the tree form) becomes the search range.

を選択すると、次の選択範囲は

にある“区”となり、さらにユーザが

を選択すると、次の選択範囲は

内にある“洞”となる。 The tree-like group in FIG. 5-6 is like a folder of a searcher of Windows (registered trademark). The search range can be automatically specified by the system. For example, when a hierarchical structure of administrative districts is stored as a subgroup of city names, when the user selects a specific city, the search range for the next input value is the name of the sub-administrative unit of the specific city. is there. For example, if a user is in a city

If you select, the next selection is

And the user will be

If you select, the next selection is

It becomes the “dong” inside.

4.9 Utilization of Relay Server Simple code can be interpreted on the client side or on the server side, so simple code is dedicated to decoding such simple code (including full code in some cases). It is possible to provide a relay server that provides a word or phrase corresponding to the above to the client side or another server side. Reference can be made to FIG. Also, in FIG. 6A, if the client cannot interpret the phrase corresponding to the simple code input by first decoding the simple code, it corresponds to the simple code input secondarily at the relay server. It is possible to interpret the word and interpret the word corresponding to the simple code input at each server again in a tertiary manner when the interpretation fails. A tertiary simple code decoding server (referred to as “tertiary server” for convenience) is a server equipped with an application that uses an input simple code or a phrase corresponding to the simple code.

  When a relay server is provided, even if a phrase that is not a simple code is required on the tertiary server side, the user inputs a simple code and the phrase corresponding to the simple code and the simple code is stored on the tertiary server side. Even if the relay server is not required, the relay server can interpret the simple code input by the user, and transmit the phrase corresponding to the simple code to the tertiary server side.

  Each time a code value of a simple code is input, it is possible to search the index storing the words corresponding to the simple code and give feedback to the client side or each server side. Can also be given.

4.10 Word Unit Classification In the present invention, the word unit means from the word start to the word end. This can be identified through a combination of all elements (word start, white space, mode switching, enter,...) That can be assigned a word and a word division. For example, blank-to-blank, blank-to-mode switching, etc. can identify that a word has been input through word start to word end. Giving feedback on a word-by-word basis can now be fully achieved through languages that support network environments.

4.11 Download Simple Code / Corresponding Phrase Furthermore, the simple code existing on the server side and the phrase corresponding to the simple code can be downloaded to the client side without the user directly inputting and storing them.

  The unit to be downloaded can be downloaded in units of words, or can be downloaded in units of groups of words (groups of a tree structure) as exemplified above. When selecting the upper group, you can download to the lower group. It is possible to download while maintaining the tree structure of the word group on the server side as it is, and it is also possible to download the words belonging to the corresponding group and the lower group as one group designated by the user on the client side. If there is a relay server whose main function is to decode simple codes, the relay server can take charge of such functions.

を入力しようとしたが、[１]＋[＊]を入力して

を入力した場合、削除ボタンを押して最終入力（［＊］）を取消すことによって

に還元させることができる。これはコントロールボタンを反復的に押して後続字母を入力する場合に有用に用いることができる。取消ボタンを連続して押す場合は通常の方法通りに既に入力された字母が削除されるようにすればよい。例えば、

を入力した状態で取消ボタンを１回押すと

になり、１回さらに押すと

になり、もう一度さらに押すと、

になる。ローマ字の場合、“ａｂａ..”を入力した状態（最後の“ａ．．”は“．．＋ａ”からなる変形字母）で削除ボタンを一度押すと“ａｂａ”になり、一度さらに押すと“ａｂ”になり、もう一度さらに押すと“ａ”になる。つまり、既に形成された字母については字母単位で削除される。 5. How to use the delete button By using the delete button, this can be used for the “final input cancellation” introduced in the previous application. For example, in Fig. 4-2

I tried entering [1] + [*]

If you delete the last entry ([*]) by pressing the delete button

Can be reduced. This can be useful when the control button is repeatedly pressed to enter the subsequent letter. When the cancel button is pressed continuously, the entered characters may be deleted as usual. For example,

If you press the cancel button once with

And press once more

And press again to

become. In the case of Romaji, when “aba ..” is entered (the last “a ..” is a modified letter consisting of “.. + a”), pressing the delete button once will change it to “aba”, and pressing it once will “ ab ”and then press again to“ a ”. That is, the already formed characters are deleted in units of characters.

6). Language Restriction Input Method The language restriction input method is to reduce the ambiguity in character input by using consonant and vowel combination rules in word generation of a specific language, and will be described in detail below.

  In the following, it is apparent that the contents applicable to other languages among the contents mentioned in one language can be applied to other languages without any particular mention. A similar concept can be applied to languages that do not use Roman characters.

（＾ｅ）は殆ど使用されない）がある。声母は子音に該当し、韻母は母音に該当する。中国大陸では中国語音を表記する方法として漢語ピン音方案を用い、台湾では注音符号を用いて中国語音を入力する。漢語ピン音は、ラテン子母、すなわちローマ字を用いて中国語音を表記することである。 注音符号とそれに対応する漢語ピン音（ローマ字）を括弧の中に表示すると次のようである。 6.1 Chinese restricted iterative selection method with consonant separation keypad in languages that use Romaji 6.1.1 Composition of Chinese initials and finals There are 21 initials and 16 finals in Chinese ,

(^ E) is rarely used). The vowel is a consonant and the final is a vowel. In Chinese mainland, the Chinese pin sound method is used as a method of notation of Chinese sounds, and in Taiwan, Chinese sounds are input using a phonetic code. The Chinese pin sound is the notation of a Chinese sound using a Latin child, that is, a Roman character. A note code and its corresponding Chinese pin sound (Roman characters) are displayed in parentheses as follows.

  Next, the Chinese combined final will be described. The combined final is the one in which three vowels i, u,... U are located in front, and other finals are connected behind it. The following table is a list of combined finals and shows the cases where they can be combined.

  As shown in the above table, it can be seen that ia can be bonded, but iu cannot be bonded.

6.1.2 Inputting Chinese Chinese Pin Tones Using Romaji After all, as you can see from the table above, using Romaji will input all 21 Chinese initials in a combination of 18 Romaji. All 16 finals can be entered as a combination of 7 simple finals or as a combination of Roman vowels and Roman consonants.

  In addition, ^ e,. Next, adding four tone codes to five basic vowels can also be input by the control processing method. After all, in the case of Chinese using Roman characters, all Chinese finals can be input using five basic Roman vowels. This is to input a deformed character that does not exist in English characters (that is, Roman characters) by the control processing method, and has already been described in the input of French, German, etc. The following is an example of setting the attachment relationship between a basic character (basic vowel) and a subsequent character (following vowel) when inputting a Chinese final character by the control processing method.

  In the above example, the letter with the tone code of 1 to 4 tones becomes the following letter, and ^ e is hardly used, and the attachment rank is the lowest. The order of attachment can be changed according to the frequency of use. For example,... Can be changed to the second succeeding letter instead of the sixth, and the attachment order of the remaining succeeding letters can be lowered one by one.

As an input example, when using the [*] button as a control button to apply the control after ^{input, - e = e + [*} ], / e = e + [*] + [*], v e = e + [*] + [*] + [*], `E = e + [*] + [*] + [*] + [*], ^ e = e + [*] + [*] + [*] + [*] + [ *]become that way. Applying partial entire selection method to "e" is displayed on the keypad of FIG. 1-1, it becomes e = [3] + [2 ], thereby ^{- e = [3] + [} 2] + [ *]. When the character displayed on the keypad in FIG. 1-1 is selected by the partial whole selection method, the full code of the person (r / en: second voice) becomes “7832 ** 65”. Here, in the input of “e” displayed on the keypad, other input methods (for example, simple repeated selection method) can be used in addition to the method of selecting the entire part.

  After all, only one control button is used, so the input method can be simplified. This can be applied in the same way even when using a phonetic code without using Roman characters (for example, the same input method is applied to a keypad displaying a phonetic code corresponding to the English letter in FIG. 1-1).

6.1.3 Chinese restriction repeated selection method with consonant separation keypad All Chinese consonants can be input in 18 roman letters, but only in the case of zh, ch, sh, roman consonants can be entered. Appears continuously. If a syllable consisting only of a final name in Chinese without a phonetic is expressed as a Chinese pin sound, it is displayed with y, w, etc. formally. For example, “cloth = yi (1 voice)”, “5 = wu (3 voices)”.

Also, as mentioned in the previous application, when a Chinese character represents one syllable, it is more convenient to use a syllable reference initial code as a simple code. Therefore, it is preferable that as many as 18 Roman consonants be assigned to each numeric button. In the present invention, a case is shown in which 18 Roman consonants are grouped by 2 and assigned to [1] to [9] number buttons.
bp / dt / gk / zj / cq / sx / mn / lr / hf

  The above example is an example of grouping based on pronunciation similarity, and when using a simple chord, there are many words (words or phrases) corresponding to the same simple chord (syllable reference initial code) This is an example in which words with similar pronunciations have the same syllable reference initial code. The above example is only one example, and various modifications are possible. Grouping can be performed based on pronunciation similarity, or grouping can be performed based on various criteria such as the dictionary order of English alphabets and the dictionary order of phonograms corresponding to English alphabets. Another advantage of grouping similar pronunciations in the same group and assigning them to the same button is that consonants with similar pronunciations rarely appear consecutively in all languages that use Roman letters. This means that there is less chance of ambiguity. Even in the use of simple codes (especially syllable reference initial codes), even if there are many words that correspond to the same simplicity, these words are words with similar pronunciation (syllabic value), so user confusion This makes it possible to use simple code smoothly.

  Moreover, although zh, ch, sh exist as Chinese phonetics, they prevent z and h, c and h, or s and h from grouping into the same group (Chinese syllables are generally “vowel + In the present embodiment, there is no problem even if grouping into the same group. However, when w, y, v, etc. not shown in FIG. 10-1 are input, s, h, w Etc., and input with the assigned button 3 strokes).

  The nine groups exemplified above can be arbitrarily assigned to nine buttons [1] to [9] as shown in FIG. 10A, and the iterative selection method can be applied to the input of each character. . Due to the characteristics of Chinese, when inputting the initials, consonants in Roman letters do not continue except for zh, ch, sh. Therefore, it is possible to input characters without ambiguity even when inputting a character by an iterative selection method. For example, it is possible to input so that b = [1] and p = [1] + [1] in FIG. Which one of the letters assigned to each button is to be used is determined according to the usage frequency of the letters. Of course, it is possible to select a frequently used character with one stroke.

  Romaji “v”, which is hardly used in Chinese, can be placed in any group among the nine groups. For example, “v” can be further arranged (or not explicitly assigned) in a group to which “f” with similar pronunciation belongs, and “v” can be input with three hits of the corresponding button. Also, w and y used for “clothes = yi (1 voice)”, “5 = wu (3 voices)”, etc. are placed in appropriate consonant groups (or not explicitly assigned), It is possible to input by an iterative selection method (for example, w and y are input with three arranged buttons). For example, if y is placed in the group to which / l and r / belong, the corresponding button ([8] button in FIG. 10-1) inputs “1” with one stroke, inputs “r” with two strokes, When “y” is entered with 3 strokes and w is placed in the group to which / m and n / belong, “m” is entered with 1 stroke of the corresponding button ([7] button in FIG. 10-1). “N” is input with two strokes, and “w” is input with three strokes.

が連続しない性質を利用し、激音コントロールを反復選択方法により選択できるようにしたものがあり、これは“韓国語制限反復選択方法”と言える。 When the iterative selection method is applied and a Chinese pin sound is input, the fact that the consonant and the vowel of the Romaji character appear alternately and the ambiguity is greatly reduced is referred to as a “Chinese restriction repetitive selection method” for convenience. When this property is generally applied not only to Chinese but also to all languages, it is called “Language Restricted Repeat Selection Method (LRRSM)” for convenience. It is called “Chiese Restricted RSM”. The language-restricted iterative selection method is less ambiguous by using the property that consonants and vowels appear alternately when assigning a pair of basic consonants and basic vowels to a button and applying the iterative selection method in Korean or Hindi. Just understand the same context that you can input. Similarly, as a method using Korean vowel elements,

There is one that uses the non-continuous nature and makes it possible to select intense sound control by an iterative selection method.

  In particular, as shown in FIGS. 10-1 to 10-4, a button to which a consonant is assigned (referred to as “consonant button” for convenience) and a button to which a vowel is assigned (referred to as “vowel button” for convenience) are separated. However (when consonants and vowels are not assigned to the same button), when applying the repetitive selection method, ambiguity can be greatly reduced by utilizing the characteristics of each language in which consonants and vowels are combined. In this manner, a keypad configured by separating a consonant button and a vowel button and assigning a predetermined number (for example, 1 to 3) of consonants or vowels to each button as in FIGS. This is referred to as “Consonant Vowal Separated Keypad (CVSK)”.

  From FIG. 10-1, it can be easily understood that the Roman consonants and vowels displayed on the keypad can be input without ambiguity even when inputting the Chinese-Chinese pin sound by the iterative selection method. When the same button among the consonant buttons ([1] to [9] buttons in FIG. 10-1) is continuously pressed for the input of the initial, the second consonant displayed on the keypad is input. The system can easily understand that it has been done. This is because the same Romaji characters do not appear in succession when inputting a Chinese phonetic. Here, the button pressed twice is a button to which w, y, v, etc. are assigned, and the same button is pressed once again (that is, when it is pressed three times) as already explained. The system can easily recognize that the operation is for inputting w, y, v, and the like. This is because if [w] 777 is input with 3 hits of the [7] button, this may be interpreted as “mmm”, “mn” or “nm”. This is because it is not configured.

  As shown in FIG. 10-1, the six vowels a, e, i, o, u,... U are divided into two groups of three, and three buttons (for example, [[ *], [0], [#] buttons), and each vowel can be input without ambiguity by the iterative selection method. This is because the same roman vowel does not continue twice in Chinese. For example, when a certain Chinese character is expressed with a Chinese pin sound, it does not become "... aa ...".

  However, it can be seen from this that it is not necessary to group a and i and e and i into the same group in compound finals ai and ei and combined finals ia and ie. If a and i are grouped into the same group and assigned to the same button (for example, [*] button), the input of the corresponding button 3 hits (that is, “***”) is “ai”. This is because an ambiguity that cannot be distinguished between an input and an input of “ia” occurs. FIG. 10A shows an example of grouping considering such contents. The vowel group and arrangement in FIG. 10-1 are not absolute, and it is obvious that the vowel group and arrangement can be modified within the limits satisfying the above restrictions. In FIG. 10A, each vowel can be input by an iterative selection method. For example, i = [0], o = [0] + [0], ioo = [0] + [*] + [0] + [0] can be input.

  As long as a and i, e and i, and o and u are not grouped in the same group, “a, e, o” may be combined into one group (see FIGS. 10-2 and 10-4), and the rest “I, u,... U” may be grouped into one group. Again, various modifications are possible. These two vowel groups are assigned to arbitrary buttons (for example, [*] and [#] buttons, respectively) in the 3 * 4 keypad, and the consonants “w, y, v” are assigned as separate groups to the remaining buttons ( For example, [0] button) can be assigned. When three letters are grouped in one group (for example, “a, e, o”), this corresponds to three hits of the corresponding button indicating a combination of vowels (for example, “ae”, “ea”, “aaa”) Since there is no Chinese final, the “o” can be recognized without ambiguity by hitting the corresponding button 3 times. For example, when “ao” is input, the corresponding button is hit 4 times, but “oa” does not exist as a Chinese pin sound, so it can be recognized without ambiguity. Reference can be made to 10-6. In FIG. 10-6, “v” is not used for inputting Chinese characters, and thus may not be displayed. Grouping “w, y” and the like into one group and assigning them to the same button has the effect of assigning half consonants (ie, half vowels) to the same button.

  In FIG. 10-2, FIG. 10-4, and FIG. 10-6, “a, e, o” are grouped in the same group, but only “ao” combinations are possible. If only “a, o” is assigned to one button, the corresponding three strokes of the button can be recognized as “ao” without ambiguity. However, if three letters are assigned and selected in the order of “a-o-e” depending on the number of pushes of the button, it is unclear whether the three-time push of the button is “ao” or “e”. May occur. Therefore, in such a case (when three or more vowels are assigned to the same button and one of two vowels can be combined among the three vowels), two possible vowels are combined with the corresponding button for one stroke and two strokes respectively. Instead of being selected with, each of them is selected with 1 stroke, 3 strokes (or 2 strokes and 3 strokes are possible, and the corresponding button 5 strokes can be recognized as “ao”). Thus, ambiguity can be removed using language restrictions. That is, when determining the selection order based on the number of button pushes in the three character groups, the selection order is determined so that characters that can appear in succession are not selected with one or two hits of the corresponding button.

  10-1 to 10-4, the same button can be pressed a maximum of 2 to 3 times when inputting the initial in romaji. The system recognizes the number of times the button is repeatedly pressed and can identify the target character by the number of times the button is repeatedly pressed. When inputting a final character in Romaji, if the same button can be pressed three times in succession in FIGS. 10-1 and 10-3, [0] is input to input “io” of the combined final character “iou”. Only when the button is pressed three times in a row. In response to a user input, the system temporarily recognizes “0 = i” and confirms when a button other than the [0] button is pressed. When the [0] button is pressed once again, the system temporarily recognizes “00 = o”, and similarly, any button other than the [0] button is pressed again. Here, if the [0] button is pressed once more, the system can determine that “000 = io”. This is to use “Chinese restriction” that “oi” does not appear when inputting Chinese pin sound. In the case of the [*] button, it can be pressed twice at maximum, so the system can temporarily recognize “* = a” and determine “** = e”. This is because the same Romaji vowels do not come in succession and the vowels assigned to the same button do not come in succession when inputting Chinese pin sounds.

  The reason why the input system can be configured without ambiguity with the consonant separation keypad (or with very little ambiguity in cases other than Chinese) is that the same consonant button (for example, [1] button in FIG. 10-1) While being pressed once or continuously, the system recognizes that one of the consonants in the same button is input, and then another consonant button (eg, [9] button in FIG. 10-1) is pressed. This is because the system recognizes that continuous consonants are input. Another reason is that about 3 letters are assigned instead of about 3 letters for each button. As mentioned in the previous application when grouping consonants and vowels as a pair, the possibility of ambiguity is significantly reduced when two letters are assigned compared to when three letters are assigned per button. Is easily understood. In addition, when two letters are assigned per button, it is possible to more effectively apply the repeated hit delay time and the release delay time presented by the applicant.

  FIG. 10-7 shows a general iterative selection method procedure in the consonant separation keypad except for the (R) portion. In FIG. 10-7, for convenience, it is assumed that the input value is a consonant button or a vowel button. That is, it is assumed that there is no special use or function (eg, a subsequent control button for entering a subsequent letter). FIG. 10-7 is not absolute and can be used as a reference only. In FIG. 10-7, a detailed procedure such as temporarily recognizing a letter corresponding to a single push of the specific button as a target letter for a specific button being pressed once is not displayed. However, such a process is performed in the (R0) process in FIG. Language restrictions can be determined in the (R) process, and the (R) process can intervene in any part of the flowchart. That is, the first input value is any one of the consonant buttons, only the two consonants are assigned to the continuously pressed buttons, and the letters assigned to the same button are consecutive. If there is no case (for example, [1] button in FIG. 10-1), when the same button is pressed twice in (R1), the input value is set to one consonant (for example, “P” in FIG. 10-1). Can be determined. This is similar to using linguistic restrictions in FIG. 4-4 to control ambiguity (determining the target alphabet with no or less ambiguity when applying the iterative selection method). It can be seen that the procedure of FIG. 10-7 is simpler than the procedure of FIG. 4-4.

  When the vowel button is pressed after the same or non-identical consonant buttons are continuously pressed, the system determines the target alphabet from the input values of the already input consonant buttons and performs the vowel processing process. The first input value in the vowel processing process is the last input vowel button in the consonant processing process. The same applies when the vowel processing process changes to the consonant processing process.

  In FIG. 10-7, the method of determining by language restriction (R) is a list of consonants and vowels that can be combined (eg, “ch”, “sh”, “zh”, “ioo”, “iou”, “ia”). ”,“ Ie ”,“ uai ”,“ uei ”,“ ua ”,“ uo ”,“ ue ”,“ ai ”,“ ao ”,“ ua ”,“ uo ”,“ ..ue ”,“ ou ” ", ...) or a list of consonants and vowels that cannot be combined (if ambiguity may occur with a particular keypad) (eg" bb "," aa "," ee "," oa ", ... ), But the input value may cause ambiguity, it is common to make the system recognize the target character when it can be combined except when it cannot be combined. Is possible. For example, when “ao = 0000” is input in FIG. 10-2 (Chinese restriction), the system has “ao” in the vowel combination list while the system has “ao” for the input value “0000”. Can be recognized as the target character. The other method is that only the [0] button can be pressed up to four times when there is no user input error with the consonant separation keypad of FIGS. 10-2 and 10-4. In this case, it can be recognized as “0000 = ao”. The remaining steps may be performed according to the processing steps shown in FIG.

  It can be seen that the procedure omitted in FIG. 10-7 is similar to the procedure already displayed. When FIG. 10-7 is further generalized and expressed without any omitted parts, it is as shown in FIG. 10-8. FIG. 10-8 has the same meaning as FIG. 10-7. In FIG. 10-8, (1) portion virtually means that button input is performed, and “n <− (n + 1)” displayed on the arrow entering (1) portion indicates that the number of button pushes is already It means to increase once.

  When Chinese syllables are expressed in pin tones, when finals such as “en”, “eng”, “an”, “ang”, “er” are used, the end of the syllable is an English consonant (the consonant of the English alphabet) ) (For example, ren: person). That is, consonants of English alphabets that can come to the end of a syllable in the middle of a word or section are “n”, “g”, “r”, and the like. A consonant that can come to the end of a syllable with a pin sound in this way is called “Consonants which can be at the end of a Syllable of Pinyin” for convenience.

  In the case of a syllable in which such a pin syllable end possible consonant is used (that is, a syllable in which a final word such as “en”, “eng”, “an”, “ang”, “er” is used) is representative) In addition, ambiguity may occur when the English consonant of the next syllable is the English consonant of the previous syllable. For example, in the case of Zhongguo (China), “˜gg˜” can be recognized as “˜k˜”. Ambiguity can occur in this very special case, which is a pin syllable end possible consonant (eg, “n”, “g”, “r”...) Ambiguity can be removed by selecting the corresponding group with two hits (as shown in FIG. 10-6, the group belonging to “n”, “g”, “r”, etc.) When only two letters are assigned).

  In FIG. 10-6, “n”, which is one of the consonants capable of ending the pin syllable, is selected by two hits of the corresponding button ([7] button). Therefore, in the case of “rennai”, “88 * * 7777 * # "is the same as that which can be input without ambiguity. In order to input “nn”, “7777” is input, but this can be recognized as “nn” without ambiguity (“mmmm”, “nmm”, “mnm” in the middle of a word or section in a pin sound) , "Mmn" and other combinations are not possible). That is, if the selection order of “g” and “k” is “kg” in FIG. 10-6, the pin sound can be input without ambiguity.

  Thus, in order to remove the ambiguity by selecting the pin syllable end consonant with two strokes from the corresponding group, two letters must be grouped in the corresponding group. If three English letters “m, n, w” are grouped as shown in FIG. 10-4 and are selected in the order of “m−n−w” according to the number of button pushes, “n” Is selected with two hits of the corresponding button, it causes ambiguity with “rewin” when inputting “renmin” (where “rewin” is not actually present in Chinese pin tones, The system can recognize the input value as “renmin”, but it is a pin sound combination that allows “rewin”).

  Therefore, as shown in FIG. 10-4, if three English consonants can be grouped by three, consonants capable of ending a pin syllable (“n”, “g”, “r”, etc.) are two English. It can be seen that it is necessary to group into groups where consonants are grouped. And when consonants are grouped two by two, if “n” and “r” and “g” and “r” are grouped in the same group among the consonants at the end of pin syllables, the ambiguity is still present. It can be seen that this may occur. For example, if two English consonants “n” and “r” are grouped into the same group, it will cause ambiguity as “˜rn˜” when inputting “˜nr˜”. The same applies when “g” and “r” are grouped into the same group.

  In FIG. 10, since “n” and “g” are assigned to different buttons, they can be input without ambiguity. Of course, even if “n” and “g” are grouped in the same group, if only “n” and “g” are grouped, they can be input without ambiguity. For example, if “n” and “g” are grouped into the same group and assigned to the [5] button, when the [5] button is pressed three times to input “vowel + ng” (ie, When “555” is input), it can be recognized as “vowel + ng” instead of “vowel + gn”. This is because it is possible to use a language restriction in which “vowel + gn” cannot come in the middle of a word in a pin sound. In this way, when “n” and “g” are grouped in the same group, if the selection order by the number of button pushes is “gn”, “g” is displayed when the corresponding button is pressed twice. If you press it once more, “ng” can be displayed, so you can naturally realize WISWYG (What You See is What You Get, more precisely, What You Press is What You See). be able to. 10-2, 10-4, and 10-6, the vowels “a”, “e”, and “o” are grouped into one group and assigned to the [*] button (example in FIG. 10-6). Therefore, if the selection order according to the number of pushes of the corresponding button is “oea”, the system recognizes and displays “a” when the [*] button is pressed three times. In addition, when the [*] button is pressed once more, the user can be given more affinity by displaying “ao” naturally.

  When only “n” and “g” are grouped into the same group and assigned to any button (eg, [5] button), if “5555” is input after the vowel button is pressed, If it is recognized as vowel + ngg and "55555" is input, it is recognized as "vowel + ngn". If “vowel + ng” is used relatively frequently in Chinese pin tones, the convenience of input can be increased by grouping “n” and “g” into the same group in this way. However, it can be said that it is not a good approach because it is natural to group consonants having similar pronunciations into the same group in order to utilize simple chords.

  When English consonants other than “n” and “g” are grouped together (for example, “m”), in order to input a pin sound without ambiguity, “mn-g” or “mg-n” If they are selected in the order of “” (ie, if consonants other than “n” and “g” are selected with one hit of the corresponding button), they can be recognized without ambiguity. For example, when “m−g−n” is selected in accordance with the number of times the button is pushed, pressing the corresponding button five times after the vowel button is pressed is recognized as “(vowel) + ng”, and six times. Pressing is recognized as “(vowel) + ngm”, pressing 7 times is recognized as “(vowel) + ngg”, and pressing 8 times can be recognized unambiguously as “(vowel) + ngn”. However, it is not preferable because the number of repeated pressings when “ng” is input becomes excessive.

  In FIGS. 10-7, 10-8, and 10-9, “nX”, “ngX”, “rX” (capital letters “X” are used as an English consonant in the consonant processing in the middle of the word, not at the beginning of the word. Even if consonants come in succession as in (), input values may be processed with consonant combinations that are possible with pin sounds. This is because English consonants such as “nX”, “ngX”, and “rX” do not always appear at the beginning of a word. After all, in the input of pin tones, there are theoretically up to four English consonants that come in the middle of a word or section like "~ ngch ~", "~ ngsh ~", "~ ngzh ~" In FIG. 10-7, FIG. 10-8, and FIG. 10-9, such a language restriction may be used in the process of consonant processing in the middle of a word.

  We presented a method that allows unambiguous input even when ambiguity may occur due to the English consonant coming at the end of a syllable. In the character group assigned to each button, the order of selection based on the number of pushes of the button is arbitrarily determined, and this is considered to be a detailed explanation of the contents presented in the previous application. The pin syllable end consonant is a consonant that can be used at the beginning or end of a syllable, so it can be said to be a relatively common consonant. The possibility that ambiguity may occur in this way is not considered to occur frequently, and even if ambiguity may occur, the pin / kanji index is referenced and does not exist. Pin sounds (for example, “rewin”) can be excluded, considering the convenience of input, etc., and determining the order in which pin syllable end possible consonants are selected in each group to which pin syllable end consonants belong Can do.

  10-1 including the above contents can be input with an average of 1.5 strokes per Roman character, but in consideration of the frequency of use in actual Chinese, it is possible to select a high-frequency character with one stroke. In this case, it is possible to input with a small number of input strokes.

6.1.4 Original function input of function buttons used as control buttons In Fig. 10-1, when inputting a modified letter that adds a suffix to a vowel in Roman letters, in addition to the 3 * 4 keypad, an optional "separate" Can be used as a control button. For example, to use the left move button in Chinese input mode (<) as control ^{buttons, - e = e + [<} ], / e = e + [<] + [<], v e = e + [< ] + [<] + [<], `E = e + [<] + [<] + [<] + [<], ^ e = e + [<] + [<] + [<] + [<] + [<] Can be entered, and if the left move button is pressed five times in succession, no more controls can be selected that are associated with “e”, so the move function is activated (“e” Is entered). On the other hand, a delete button if utilized as a control button for inputting affixed characters (for convenience, "[X]" displayed) <sup>is, - e = e + [X</sup> ], / e = e + [X] + [X], <sup>v</sup> e = e + [X] + [X] + [X], `e = e + [X] + [X] + [X] + [X], ^ e = e + [X] + [X] + [X ] + [X] + [X] can be entered, and when the delete button is continuously pressed five times, the already inputted “e” is deleted. In this way, buttons with other functions are used as control buttons, but when the control is selected by the number of times the control button is pushed and there is no more control to be selected, the original function (for example, movement function) is selected. It can be so.

  In the above example, when the left movement function is activated, the word start state is entered. At this time, if the left movement button [<] is pressed only once, the left movement function is input. That is, after inputting “e”, if the [<] button, which is a control button, is pressed five times, the cursor moves one side to the left with only “e” being input, and if it is pressed again, the Move one. This is applicable to all languages.

6.1.5 Chinese restriction repeated selection method with consonant separation keypad (input of modified characters on 3 * 4 keypad)
Here, a 3 * 4 keypad can also be used to input a modified alphabet (including a case where a tone code is added) with a subscript on a Roman vowel.

As shown in the list of combined finals, “i” and “u” do not come after “i” (in other words, “a” or “e” or “o” comes). Also, “u” and “i” do not come after “u” (in other words, “a”, “e”, or “o” comes). Therefore, the vowel arrangement shown in FIG. In FIG. 10-2, when adding a tone symbol to the left vowel “i, a, e”, the [#] button to which the right “u” is assigned is used as a control button. Similarly, when adding a tone code to the right vowel “o, u”, the left [*] button is used like a tone symbol control button. For example, ^/ o (second voice) = o + [*] + [*] = [0] + [0] + [0] + [*] + [*], and ^/ a (second voice) = a + [ #] + [#] = [0] + [#] + [#], ^ e = e + [#] + [#] + [#] + [#] + [#] = [0] + [0 ] + [#] + [#] + [#] + [#] + [#], Respectively. It is possible to input compound vowel ao = a + o = [0] + [0] + [0] + [0], and vowels a and e do not come after vowel a, and there is no case like oa. The system can be identified without ambiguity. Since the vowels u are similar in pattern to the vowels u, they can be input with two hits of buttons to which u is assigned (a concealed repetitive selection method).

  This utilizes the property that the vowel u does not come after a and e among the vowels “a, e, o” in Chinese, and the property that the vowel i does not come after the vowel o. The constraints described above, ie, a and i are not placed in the same group, e and i are not placed in the same group, and o and u are not placed in the same group, two vowels on three buttons It can be seen that it is impossible to use a button to which a vowel is assigned as shown in FIG. In FIG. 10-2, i and u are arranged in the [*] button and the [#] button, respectively, so that it can be easily recognized that this button is used like a control button. This is because the balance of the arrangement is taken into consideration. The average number of vowel input strokes in FIG. 10-2 is about 1.5 as in FIG. 10-1.

  As a modified example of FIG. 10-2, a part of the three letters assigned to the [0] button can be arranged on “another button” in addition to the 3 * 4 keypad. For example, a part of three letters can be assigned to any one of the up / down / left buttons.

  10-1 to 10-4, the Chinese pin sound can be input without ambiguity while applying the iterative selection method. The consonant buttons ([1] to [9] buttons) and the vowel buttons ([*]) , [0] and [#] buttons), and the appearance rules of Roman consonants and vowels when Chinese is expressed in Chinese pin tones are used.

6.2 English restriction selection method with consonant separation keypad In the case of English etc., as with Chinese, iterative selection method with consonant separation keypad (CVSK) similar to FIGS. 10-1 to 10-4 Can be entered with almost no ambiguity. This is because all languages using Roman letters have a structure in which word structures are alternately combined with consonants and vowels.

  In the case of English, when the consonant group is the maximum repetition from “word start”, “CCCVCCCC” is obtained as in “strengths” (C is a consonant and V is a vowel). At the beginning of a word, up to three consonants can be consecutive, only starting with “st˜” or “sp˜” (eg, spree, splene, strength, etc.).

  As in FIG. 10-1, a keypad to which about 2-3 consonants are assigned (for convenience, it is assumed that only “... U” is not assigned in FIG. 10-1), and the user uses the English word “student”. If you enter "622" to enter "", the system can treat "622" entered after "Start Word" as "st" input instead of "sdd" input (in English If there are three consecutive consonants in a particular word, it only begins with "st ~" or "sp ~", and the system remembers such English word generation rules or letter combination rules). When “u” of “stu˜” is input, the button to which the vowel is assigned is pressed, so the consonant group (eg, “st”) ends and the vowel (eg, “u”) comes. I understand that. When “2” is input to input “d” after “stu” is input, the system knows that the consonant starts again. Similarly, when the user inputs “˜ ** 7722” in order to input “to ent” of “student”, when “**” is input, in the case of English as described in the prior application, Since “a” is not continuous, the system can easily recognize that “e” is input instead of “aa”, and “7722” is changed to “mmt”, “nddd”, “nt”, “mmdd”. However, as mentioned in the previous application, the ambiguity generated from the iterative selection method can be greatly reduced by setting the “repetition hit delay time” and the “release delay time” differently. .

  Even in the current standard English keypad (Figure 1-1) assigned to each button with consonants and vowels mixed, application of the iterative selection method reflects the word generation rules (character combination rules) of the specific language in this way. The ambiguity that sometimes occurs can be reduced. However, when consonants and vowels are mixedly assigned to one button as described above, it is very difficult to apply the language restriction input method (particularly, the repetitive selection method is applied as the full input method). For example, when “student” is input by the iterative selection method in FIG. 1-1, “stu˜” is “777888-”, but even if “777” is recognized as “s”, the following “ 888 ”is“ ut (ie, sut) ”,“ tu (ie, stu) ”, or“ v (ie, sv) ”(of course, after“ s ”when the English restricted input method is applied) Since “ttt” does not come, the system can consider the input value “888” not to be “ttt”). Similarly, when “333” is input to input “˜de˜”, it is not known whether “de” or “ed”, but “888877” inputted before “˜de˜” is “ Since “sut”, “stu”, and “sv” are analyzed, both “˜de˜” and “˜ed˜” can be performed for each of the three cases.

  Even if the language restriction input method is applied in this way, a lot of ambiguity occurs when the iterative selection method is applied as the full input method. This is because the language restriction input method uses the appearance rules of consonants and vowels (that is, word generation rules and vowel combination rules). At this time, the input value is for inputting consonants or for inputting vowels. It is because it cannot be clearly distinguished whether it is a thing. As shown in FIGS. 10-1 and 10-4, when consonants and vowels are assigned separately, when the iterative selection method is applied, a vowel input is input when the vowel assignment button is pressed. When the consonant assignment button is pressed, the system recognizes that the input is a consonant, so the ambiguity can be greatly reduced.

  In the case of English using Romaji (English), there are many cases where two or more vowels continue (for example, 'ai' of captain). When the same vowels (basic vowels a, e, i, o, u) are continuous in actual words, there are cases of oo, ee like food and teen, and uu like vacum. There is a case. At this time, the former is relatively frequent, and the latter is not so many. The applicant has not yet seen the words ... aa ..., ... ii ... in the English dictionary. Therefore, when assigning five basic vowels in English to the [*], [0], [#] buttons, 'a' is selected with one stroke, and any one of e, o, and u is selected with two strokes. Doing so actually removes the ambiguity in many cases. For example, when the vowel 'a, o' is assigned to the [*] button and the iterative selection method is applied, when the [*] button is input twice, it is not the input of the two vowels aa but the input of the vowel o. The system will recognize. Similarly, when “i, u” is assigned to the [0] button and the iterative selection method is applied, when the [0] button is input twice, it is not the input of the two vowels ii but the input of the vowel u. The system will recognize. In a word, it cannot be said that it does not continue at all like aa, ii, but the ambiguity can be reduced by setting differently the “repetition delay time” and the “off delay time” mentioned in the previous application. Can be eliminated. The remaining vowel 'e' can be assigned to the remaining [#] buttons and input with a single stroke.

  If one of the buttons [*], [0] and [#] is to be used as a special purpose button, the vowel ‘e’ is assigned to an arbitrary button to which the vowel is assigned. For example, 'i, e, u' can be arranged on the [0] button. The vowel u is input with three hits of the [0] button. This is because the vowel u is the least frequently used vowel among English vowels. After all, if 'i, e, u' is assigned to the [0] button and i in English does not appear continuously (ie, ... ii ...), i and e are input without ambiguity. When entering u with 3 strokes of the [0] button, there will be ambiguity that u, ie, or ei are not known. The low use frequency of u means that there are not many cases like this.

  Eventually, the vowel separation keypad can be used to greatly reduce ambiguity, and when assigning more vowels to one button, the same vowels of English words in the dictionary do not appear consecutively. When a vowel is selected with one stroke and a vowel is input by a repetitive selection method using a button to which a plurality of vowels are assigned, the vowel can be input with almost no ambiguity.

  The selection of vowels that do not appear in succession in this way and the selection of the least frequently used vowels in three strokes is also applicable when approximately two to three consonants are assigned per button. can do.

  As described above, the case of English using Roman characters has been described as an example, but the same applies to the case of other languages using Roman characters.

6.3 Indonesian Restricted Iterative Selection Method on the Consonant Separation Keypad In Indonesian, the word is written in Roman letters (English letters). Indonesian syllables are organized as follows (C is a consonant and V is a vowel).

V: be-a tariff VC: am-bil grab CV: go-sock rubbing CVC: pon-dol hut CCV: tra-di-si tradition CCVC: con-trak contract CVCC: teks-tur textile CCCV: kon-struk -Si construction CCCVC: strip-tis nude dance

  If there are more than two consonants, it is a foreign language from English, etc. From the above explanation, even in the case of Indonesian language, it is expected that three or more consonants will not continue at the beginning of a word unless it begins with “st˜” or “sp˜”. Therefore, it is possible to apply the Indonesian language limited iterative selection method using such a word generation rule (character combination rule).

  It is said that q and x are used in natural science symbols such as physics and mathematics. In other words, it is considered that the characters are hardly used for character input. Therefore, q and x can be not explicitly arranged in a specific group, and can be input with three strokes. Similarly to the case of Chinese, two or more Roman alphabets may be combined to express one Indonesian sound, and there are four notations such as ny, sy, kh, and ng.

  Other consonant groups that are not split are bl, br, dr, dw, dy, fl, fr, gl, kr, ks, kw, pl, ps, rps, rs, sk, skr, sl, sp, spr. , Sr, str, sw and the like. In particular, since “skr˜”, “spr˜”, “str˜”, etc. come at the beginning of a word, means for discriminating whether an input value is a full code or a simple code by a language restricted parallel input method described later. (As an applicant, I don't know if "rps" may come at the beginning of a word, but if it exists, it can be used as well.)

  Of the 21 English consonant consonants, the 19 consonants excluding q and x can be grouped into 9 groups, but any method can be used. For example, grouping can be performed as BP / DT / GK / CJ / MN / LR / SZ / FV / HWY.

  q and x can be grouped into appropriate groups. For example, q can be grouped into a “GK” group and x can be grouped into an “SZ” group.

  To represent Indonesian vowels, five vowels a, i, u, e, and o are used. Also, there are three double vowels ai, au, and oi, and oi is rarely used. Therefore, when grouping the five vowels into two or three groups, it is preferable that a and i and a and u are not grouped into the same group. For example, grouping into ae / uo / i. Further, as in English, it is preferable to select a vowel that does not appear the same vowel continuously (the appearance frequency is low even if it continuously appears) with a single button hit to which the vowel belongs in each group. .

6.4 Japanese Restricted Repeated Selection Method on the Consonant Separation Keypad When inputting Japanese, there is a wide range of methods for inputting Japanese pronunciation in romaji and then converting it into Japanese. It is possible to input Japanese sounds in Romaji using a consonant separation keypad similar to FIGS. 10-1 to 10-4 and convert them into Japanese. “A”, “I”, “U”, “E”, “O” are respectively expressed as “a”, “i”, “u”, “e”, “o”, and “NA”, “NI” ”,“ Nu ”,“ Ne ”,“ No ”are written as“ na ”,“ ni ”,“ nu ”,“ ne ”,“ no ”, respectively, and the remaining letters are also consonants and vowels in Roman letters. It can be expressed by a combination of.

  When a consonant in Roman characters appears twice in succession in Japanese, it is a case where a prompt sound or a stuttering sound (expressed in small letters) is used. In addition, when the vowels in Romaji appear consecutively, on the assumption that the combination of “A”, “I”, “U”, “E”, “O” in the words does not appear consecutively, It is about 2 times at most, and it is extremely rare to appear 3 or more times in succession. Therefore, the five vowels (a, i, u, e) of the Roman alphabet are divided into three groups as in FIGS. You can see that there is no ambiguity even if you enter. In particular, in the case of Japanese, it can be easily understood that the Japanese corresponding to the input of “a” becomes “a” and the Japanese corresponding to the input of “na” becomes “na” (the remaining Japanese The same is true for characters). Therefore, at the moment when the user inputs “na” and the system confirms that the input value is “na”, the system can provide “na” to the user.

  There are 14 consonants in Romaji used for Japanese notation of 50-symbol charts: k, s, t, n, h, m, y, r, w, g, z, d, b, p. When stuttering is expressed, it is expressed by combining two Roman consonants (for example, cha, sha), by combining “y” (for example, kya), or by using “j”. When Romaji consonants appear twice or more consecutively, there are ch, sh, ky, ny, hy, my, ry, gy, py, etc., and there may be a prompt sound. When a prompt sound is input, the same letter may appear continuously among k, s, t, and p (for example, ippai). Accordingly, it can be seen that 16 (14 + c, j) Roman consonants are essential for Japanese input. In this way, a consonant separation keypad can be configured to easily input consonants essential for Japanese input. The remaining five Romaji consonants (f, l, q, v, x) are also necessary for input such as English, but once grouped around the 16-character base. For example, it is as follows.

BP / DT / GK / CJ / H / MN / R / SZ / YW / => Group into 9 groups.
BP / DT / GK / CJ / H / MN / Y / SZ / RW / => 9 groups.
BP / DT / GK / CJ / HR / MN / SZ / YW / => Group into 8 groups.

  The five Roman consonants required for English input can be added appropriately to each group as in the case of Chinese. In the case of grouping into 8 groups, the remaining 4 buttons can be used as vowel buttons with a 3 * 4 keypad. If only three vowel buttons are used, the remaining one button can be used for consonants necessary for English input.

6.5 Intentional language restriction cancellation The language restriction iteration selection method sacrifices the merit of the “full input method” in which all words can be entered regardless of whether or not they are in a dictionary. It is preferable that the user sets in advance whether or not to apply such language restrictions. However, even in the “Language Restricted Input Mode” that enables the language restricted input method, if the user wants to input a word that does not match the word generation rule (character combination rule) that is not in the dictionary, after entering the character All character combinations can be entered by entering specific functions (for example, blank and left-hand progression or word end), deliberately confirming the target character and then entering the next character. . For example, when the user inputs “622” in the “English restriction input mode” of FIG. 10A, the system recognizes that “st˜” is not “sdd˜”. If you want to enter, enter "62", then enter the space and left move function, enter "2", or enter "62" and then enter the "word end function (or word end function) Input another means for activating "the control that gives the effect of (2)", and then input "2". When the word end function is activated after entering “62”, the system may recognize “d” as “d” because “2” to be input next appears for the first time after “word start”. it can. In this manner, intentionally inputting the word end function and overcoming the language restriction in the specific language restriction input mode is referred to as “intentional language restriction release” for convenience. An example of “sdd˜” is “intentional English restriction release”.

  Similarly, in “Chinese restriction input mode (application of Chinese restriction repetitive selection method)” in FIG. 10B, the user is a combination of vowels that are not present in Chinese but are present in English. When "u" is to be input, after inputting "u", further means for ending the word (referred to in the above example) is input, and then "i" is input. On the other hand, in the “Chinese restriction input mode (application of Chinese restriction repetition selection method)” with reference to FIG. 10-2, after inputting “u” and pressing the button assigned with “i”, As mentioned, "u" variants (eg, "u" with a tone code or "u" with "..." on it are considered "u" variants. (When the Chinese restriction repeat selection method is applied, when a button assigned with “i” is pressed after the vowel “u”, “o”, it is used like a control button. Because) Eventually, the user can input combinations of words that do not exist in the combined Chinese language (for example, combinations of all letters such as English words) even in the Chinese restriction mode. In other words, a user who mainly uses Chinese is set to the Chinese restriction input mode (applied mode of the Chinese restriction repeated selection method), and all the settings that do not exist in Chinese even if the setting is not changed. You can enter a combination of letters. This is “intentional Chinese restriction cancellation”.

と母音

を入力しようとする場合は、“単語開始”状態で［１］を入力し、次いで単語終了機能を与える手段を入力し、再び単語開始状態で［＊］を入力する。［１］ボタンと［＊］ボタンを連続的に入力すると、

となるからである。標準キーボード（英語及び韓国語の標準キーボード）では、右側矢印ボタンを押すと、“単語終了”空白が入力せずに単語終了機能が活性化されるようになっており、本発明においても右側矢印ボタンがさらに備えられる場合、それは同一に適用することができる。 The method using the three vowel elements in Korean shown in FIG. 4-5 is also applicable to inputting consonants or vowels as one character. For example, in Figure 4-5, a consonant as a single character

And vowels

Is input in a “word start” state, then a means for providing a word end function is input, and [*] is input again in the word start state. When [1] button and [*] button are input continuously,

Because it becomes. With standard keyboards (English and Korean standard keyboards), pressing the right arrow button activates the word end function without entering the “word end” space. If a button is further provided, it can be applied identically.

6.6 Delay Time for Canceling Language Restriction It has been described above that “continuous hit delay time” and “offset delay time” can be applied even when three or more characters are assigned to one button. For example, if the delay time is set to 0.1 seconds with the standard English keypad in FIG. 1-1 and the [2] button is pressed twice within 0.1 seconds, the system recognizes that B has been entered. be able to.

  Similarly, when the [2] button is pressed three times in succession (ie, [2] + [2] + [2]), the first input value and the second input value (ie, the first input value) The delay time interval between the [2] button and the second [2] button is within a set time (for example, 0.1 second) and the second input value and 3 If the delay time interval between the input value and the second input value is within the set delay time (for example, 0.1 second) (ie, [2] + within 0.1 second + [2] +0. Within 1 second + [2]), the system can recognize as C input. Or, when the [2] button is pressed three times in succession (ie, [2] + [2] + [2]), the total input time is twice the repeated hit delay time (for example, 0.2 seconds) If it is within the range, the system can also recognize it as an input of C.

  Further, in the case of English, for example, assuming that there is no “... U” in FIG. 10A, the property that the English vowel “a” does not appear continuously and the property that “i” does not appear continuously are strict. When an abbreviation such as “NII” is input, “NII” can always be input in the English restriction repeated selection mode through “intentional English restriction release” by inputting the word end function. However, after a certain period of time has elapsed after inputting "NI", NI can be determined without intentionally inputting the word end function. This fixed time can be set to be the same as the “release delay time” mentioned in the previous application, but is preferably set longer than the “release delay time”. For example, when 2 seconds elapse after inputting "NI", NI is determined without inputting the word end function, and the system returns to the "start word" state again. This is referred to as “temporary language restriction release delay time” for convenience, and it is preferable that this can also be set by the user. It is self-evident that this applies equally to all languages.

The various delay times are summarized as follows.
Repeat delay time ≦ Release delay time ≦ Temporary language restriction release delay time All three delay times can be set to the same, but the release delay time is set longer than the repeat delay time and the release delay time is set. It is preferable to set a longer temporary language restriction release delay time.

6.7 Portuguese Restricted Repeat Selection Method with Consonant Separation Keypad In Portuguese, k, w, y are used only for abbreviations or foreign words, but are still necessary for character input. In Portuguese, the consonants are:
bl, cl, dl, fl, gl, pl, tl
br, cr, dr, fr, gr, pr, tr, vr

  In addition, there are double consonants such as gn, mn, pn, ps, pt, tm, ch, lh, nh, rr, and ss. Therefore, in constructing a consonant separation keypad for the Portuguese restricted input method, it is necessary to prevent the consecutive letters from being grouped into the same group. For example, FIG. 10 can be referred to. However, although “mn” is grouped in the same group, it can be appropriately modified and arranged in another group. The reason why the keypad of FIG. 10 can be similarly applied to other languages is that groupings based on pronunciation similarity (for example, voiced and unvoiced sounds having similar note values are grouped in the same group, / bp /, / dt /, / Gk / ...), because there are few cases where letters with similar pronunciations come in succession.

  Portuguese basic vowels include five vowels a, e, i, o, and u. Among them, a, e, and o are strong vowels, and i and u are weak vowels. When two different vowels come in succession, six of “strong vowels + weak vowels” (ai, au, ei, eu, oi, ou) and one of “weak vowels + weak vowels” (ui) ). That is, the strong vowels and the weak vowels can be grouped into separate groups (for example, in a state where “... + U” is not present in FIG. 10-6). In addition to such grouping, it is possible to group weak vowels and strong vowels into an arbitrary number of groups so as not to be in the same group. For example, grouping can be performed like / a / eo / ui /.

6.8 Spanish Restricted Repeat Selection Method with Consonant Separation Keypad Spanish has “˜ + n” (characters with “on” on n) as a character that is not in English. k and w are used only for notation of foreign words, but are still necessary for character input. In Spanish, the consonants are:
bl, cl, dl, fl, gl, pl
br, cr, dr, fr, gr, pr, tr

  That is, in configuring a consonant separation keypad for the Spanish restricted input method, it is necessary that both l and r are not assigned, and the characters that can form the double consonant are in the same group. Must be grouped and assigned to the same button.

  Reference can be made to the consonant groupings of FIGS. 10-1 to 10-4 and FIG. 10-6.

  Spanish basic vowels include five vowels a, e, i, o, and u. Among them, a, e, and o are strong vowels, and i and u are weak vowels. When two different vowels come in succession, six “strong vowels + weak vowels”, six “weak vowels + strong vowels”, and two “weak vowels + weak vowels” (iu, ui) ). Triple vowels are also combined as “weak vowel + strong vowel + weak vowel”. That is, when strong vowels + weak vowels are grouped into separate groups (for example, in the state where “. In this case, since iu and ui are all possible, it is possible to overcome the ambiguity by applying the repeated hitting delay time / offset delay time or using the applicant's presentation method such as intentional language restriction cancellation. The same applies when the same vowel appears continuously, when the same consonant appears continuously, and when other ambiguities may occur.

6.9 Italian Restricted Repeat Selection Method with Consonant Separation Keypad In Italian, j, k, w, x, y are used only for the notation of old or foreign words. Necessary for input such as English. In the consonant grouping, the remaining five consonants except the five consonants can be grouped and the remaining five consonants can be appropriately grouped.

  In Italian, double vowels include ia, io, ie, iu, ai, ei, oi, ui, uu, ou, eu, and the like. Excluding iu and ui, this is a combination of strong vowels (a, e, o) and weak vowels (u, i). The triple vowel also has a structure of “weak vowel + strong vowel + weak vowel”. Therefore, grouping of strong vowels and weak vowels such as / aeo /, / ui / can be used, and in the case of ui, iu, there may be ambiguity.

6.10 German restricted repeat selection method with consonant separation keypad German consonants are ch, chs, ck, ds, dt, ng, nk, pf, ph, sch, sp, st, th, ts, tz , Tsch, etc. Similarly, when grouping contiguous consonants so that they are not grouped into the same group, the consonant grouping of FIG. 10- * can be referred to. Further, in the case of “sch˜” in German, since three or more consonants can come at the beginning of the word, this can be utilized in the language-limited parallel input method described later. In the case of Tsch, it can come to the beginning, but it is rarely used.

  German has five single vowels, and there are odd vowels with "..." on a, o and u. Examples of the double vowel include au, ei, ai, eu, “. When inputting a vowel using only a single vowel (basic vowel), there are cases of ie and ei, so it is necessary to group i and e into different groups. For example, grouping such as / a e o / and / i u / can be performed, and many variations are possible.

  If the same consonant or vowel comes in succession, ambiguity can be avoided by various methods presented by the applicant.

6.11 Vietnamese restriction input method 6.11.1 Vietnamese restriction input method with consonant separation keypad Vietnamese syllables are mainly "vowel", "consonant + vowel", "vowel + consonant", and " Consonant + vowel + consonant. In particular, Vietnamese words are based on single syllable words. Although the number of compound syllable words is increasing, the fact that there are many single syllable words means that it is easy to configure an input system in a consonant separation keypad.

  Vietnamese has five basic vowels a, e, i (y), o, u and “v + a (vowel with a on top of a, hereinafter“ x + vowel ”is“ x ”above vowel. There are six variable vowels such as “, + ^”, “^ + o”, “, + o”, “, + u”. “Y” pronounces “i” long. A consonant separation keypad can be composed of only five basic vowels, and the remaining variable vowels can be input by the control processing method. A consonant separation keypad can be composed of only a part of basic vowels and variable vowels, and a consonant separation keypad can be composed of 11 vowels as a whole.

  Vietnamese double vowels and triple vowels have various combinations so that five basic vowels are grouped into two or three groups similar to FIG. 10 so that there is no ambiguity when applying the iterative selection method. It is not easy to group. Vietnamese has 6 types of tone, and 5 types of tone codes are displayed above or below the vowel. In Vietnamese, the same vowels do not come continuously, but this is a useful property for inputting Vietnamese vowels. Therefore, it is conceivable that five basic vowels are arranged in five groups, a variable vowel is input by a concealment type repetitive selection method, and a tone code is attached to the basic vowel and the variable vowel by a control processing method. Here, y can be regarded as an odd vowel of i. That is, vowels are / 'a', 'v + a', '^ + a' / 'o', '^ + o', ', + o' / 'u', ', + u' / 'i', 'y' / '. It is divided into five groups, e 'and' ^ + e '/. It is possible to modify this and combine the groups with a small number of odd vowels into four groups. For example, / 'a', 'v + a', '^ + a' / 'o', '^ + o', ', + o' / 'u', ', + u' / 'i', 'y', 'e' , '^ + E' /. However, when arranged in five groups, a vowel can be input without ambiguity by applying the iterative selection method. In each group, an arbitrary character is displayed on the keypad as a representative character, and the remaining characters may not be displayed. It is natural that a single vowel mainly becomes a representative character. In the case of dividing into four groups, it is possible to display all i and e. The selection order based on the number of button pushes is determined in consideration of the usage frequency.

  Vietnamese consonants include “− + d” (the center of “d” ‘−’) ”that is not in English, and English f and z are not used. ‘− + D’ is regarded as a modified letter of ‘d’, and a (hidden) iterative selection method or a control processing method can be applied. Some textbooks do not use f, w, z in Vietnamese, and others use w, j as semiconsonants. It is recognized that at least f and z are not used and w is rarely used. In the present invention, w is assumed not to be used, but may be included in an appropriate group as necessary.

  In addition, there are ch, gh, gi, kh, ng, ngh, nh, ph, qu, th, tr, etc. as complex consonants. Among them, gi and qu are not combined here because they are a combination of consonant and vowel in terms of form. The fact that Vietnamese is mainly composed of single syllables means that unlike other languages, there is almost no continuation of syllables consisting of “child + mother + child”. Except in some cases, this means that consonants and consonants are rarely continuous. As a result, when applying the repetitive selection method by dividing the consonants into several groups, if the consonants constituting the double consonants are not grouped into the same group, the consonants are ambiguous as in the case of Chinese. It will be possible to input without.

  For example, it can be grouped into eight groups such as bp / dt / gk / cq / sx / mn / lr / hvj. Further, it can be divided into six groups such as bpv / dt / gkq / scx / mnj / hlr /. It is also possible to group into arbitrary groups. Unused consonants such as f and z can be further included in appropriate groups.

  If consonants are divided into 8 groups and vowels are divided into 4 groups, all consonants and vowels can be accommodated in the 3 * 4 keypad. If consonants are divided into 6 groups and vowels are divided into 5 groups, 11 buttons can contain consonants and vowels, and the remaining 1 button can be used as a control button to add tone codes to vowels. can do. If consonants are divided into six groups and vowels are divided into four groups, all the consonant buttons and vowel buttons can be accommodated with ten numeric buttons. This has the effect of making the simple code consist only of numbers in the use of the simple code. The remaining two buttons can be used as a control button for inputting a tone code and a subsequent control button for removing ambiguity.

6.11.2 Vietnamese Restricted Input Method Using Consonant and Vowel Pairs It is possible to apply the iterative selection method by grouping Vietnamese vowels and consonants as a group. You can refer to the case of Korean. If the number of consonant and vowel pairs is 10, the remaining consonant and vowels may be input by the control processing method.

6.12 Russian Restriction Input Method with Consonant Separation Keypad To configure a consonant separation keypad for Russian, applicant's earlier application PCT / KR02 / 00247 can be referred to.

  There are 33 letters in Russian. That is, it is composed of 10 vowels, 20 consonants, one half vowel (or half consonant), and two symbol vowels (hard consonant code, soft consonant code).

Next, 33 letters were arranged in upper case and lower case according to the order of the letters.
АБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ
абвгдеёжзийклмнопрстуфхцчшщъыьэюя

  The 20 consonants are divided into voiced and unvoiced consonants as follows. The brackets mean pronunciation.

  Vowels are divided into hard vowels and soft vowels.

  In appropriately grouping Russian consonants, voiced and unvoiced sounds with similar note values can be grouped into the same group. For example, it is classified as / б (b) п (p) / д (d) т (t) /. Consonants that are not paired for pronunciation can be appropriately grouped as presented in the earlier application. Some examples are as follows, and various modifications can be made while grouping unvoiced sounds and voiced sounds that make a pair.

10 groupings Example 1: БП / ДТ / ГК / ВФ / ЗС / ЖШ / ЛР / МН / ХЦ / ЧЩ
Example 2: БВ / ГК / ДТ / ЖЗ / ЛР / МН / ПФ / СХ / ЦЧ / ШЩ
Nine grouping cases: (b) п (p) / д (d) т (t) / г (g) к (k) / в (v) ф (f) х (x) / з (z) с ( s) / ж (zh) ш (sh) / ц (ts) ч (tsh) щ (shsh) / л (l) р (r) / м (m) н (n)
Eight groupings: (b) п (p) / д (d) т (t) / г (g) к (k) / в (v) ф (f) х (x) / з (z) с ( s) ж (zh) / ш (sh) ц (ts) ч (tsh) щ (shsh) / л (l) р (r) / м (m) н (n)
Seven grouping cases:
...

  In the above example, a semi-consonant sound, a hard consonant code Ъ, and a soft consonant code Ь may be included in an appropriate group. Further, when assigning such a group to each button, some characters (for example, characters other than Й, Ъ, Ь, etc.) may be omitted and not displayed in order to keep the keypad simple.

  As the vowels, ten vowels can be grouped into five groups using a pair of hard vowels and soft vowels. There are 10 Russian vowel vowels, and there are 11 vowel sounds as pronunciation, but the basic phonemes of vowels are 5 (a), (e), (i), (o), (u). There is one. Among these, (e) and (o) appear only under stress. That is, the vowel phonemes (a) and (o) can be regarded as strong vowels. There are only three remaining vowel phonemes at positions without stress. Therefore, 5 basic vowels of 10 Russian vowels are placed, and among them, vowels corresponding to strong vowels are divided into one group, and the remaining three basic vowels are divided into one or more groups. be able to. This is because it can be predicted that it is difficult for vowels of the same type to come continuously (for example, strong vowels next to strong vowels). For example, two groups such as а (a) о (o) / у (u) э (e) и (i) /, or а (a) о (o) / у (u) э (e) It can be divided into 3 groups like / и (i) /. The vowels that form a pair corresponding to the basic vowel can be placed together in a group to which the basic vowel belongs and the repetitive selection method is applied, or the control processing method can be applied by regarding the vowel as a modified vowel.

  Consonant and vowel groups can be used appropriately to construct a consonant separation keypad within a 3 * 4 keypad (eg, vowel 5 group and consonant 7 group, vowel 3 group and consonant 9 group). In Russian, a large number of consonants may continue (for example, CCCVC...) As in English, but a typical syllable structure is a consonant and vowel repetitive structure (CV, CVCV, CVCVCV,...). This means that when applying the iterative selection method with a consonant separation keypad, in most cases Russian words can be entered without ambiguity (ie with very little ambiguity).

6.13 Hindu Restriction Input Method with Consonant Separation Keypad Applicant's earlier application PCT / KR00 / 00601 shows an example of grouping Hindu consonants into nine groups based on pronunciation similarity. Then, the example of grouping into 10 groups was shown. In Hindu, it was shown that group consonants with strong pronunciation similarity could be grouped into the same group. For example, in the picture presented above, group consonants having pronunciations of (k), (kh), (g), and (gh) are grouped into one group. The same applies to the remaining similar pronunciation groups. Of the 35 Hindi consonants presented above, 33 consonants excluding _ (ud) and _ (udh) with a lower point can be arbitrarily grouped into 9 groups or 8 groups. _ (Ud) and _ (udh) can be regarded as modified characters of __ (d) and __ (dh), respectively, and can be arranged in the group to which the basic character belongs.

  The following are the Hindu vowels described in the previous application.

  Hereinafter, for the sake of convenience, Hindu vowels are expressed in English pronunciation in parentheses next to the underline (assuming that the underline has Hindu characters of the corresponding pronunciation) or simply in English parentheses in parentheses. For convenience, _ (aa) is (a-), _ (ee) is (i-), _ (oo) is (o-), _ (ae) is (e), and _ (aaa) is (ai). write.

  The vowel _ (ri) may be classified as a consonant. Ten vowels excluding the vowel_ (ri) can be grouped into five groups in pairs as described above.

で入力したのと類似である。 Hindu vowels are broadly divided into single vowels (a), (i), (u) and long vowels _ (a-), _ (i-), _ (u-), and the remaining four vowels are It is a complex vowel and can be displayed as a combination of single vowels. That is, _ (e) = (a) + (i) or (a) + (i−), _ (ai) = (a) + (e), _ (o) = (a) + (u) or (A) + (u−), _ (au) = (a) + (o) Therefore, the basic vowels (a), (i), and (u) are assigned to three buttons in three groups, the basic vowels are input with one keystroke, and the long vowel corresponding to the basic vowel is the basic vowel. It is possible to input with the assigned button 2 hits, and the remaining 4 complex vowels can be input as a combination of basic vowels. Inputting complex vowels in this way is

It is similar to that entered in.

がそれぞれ［＊］，［０］，［＃］ボタンに割り当てられている場合、__（ａ−）＝＊＊で、ｅ＝＊０又は＊００で入力する。もし＊＊０が押されると、“＊＊”まで押されたとき、システムは＿（ａ−）と認識するが、次に“０”が押される瞬間、＊＊０を（ａ）＋（ｅ）＝＿（ａｉ）と認識することができる。同様に＊＊＃が押されると、“＊＊”まで押されたとき、システムは、一時的に＿（ａ−）と認識するが、次に“＃”が押される瞬間、＊＊＃を（ａ）＋（ｏ）＝＿（ａｕ）と認識することができる。 For example,

Are assigned to [*], [0], and [#] buttons, respectively, ___ (a −) = ** and e = * 0 or * 00. If ** 0 is pressed, the system recognizes it as _ (a-) when it is pressed to "**", but the next time that "0" is pressed, e) can be recognized as = _ (ai). Similarly, if ** # is pressed, the system will temporarily recognize it as _ (a-) when it is pressed to “**”, but the next time “#” is pressed, It can be recognized as (a) + (o) = _ (au).

  This is possible if the ten vowels of Hindu do not come consecutively in one word. Ambiguity can occur if Hindu vowels come in series. That is, when the user inputs ** 0, the system does not know whether it is _ (a-) and _ (i) or _ (ai). However, even if 10 basic vowels may continue, if the frequency is very low, the system is basically used when it is possible to mean the complex vowel_ (ai) as ** 0. A Hindu restriction that recognizes input values as complex vowels can be applied. To implement such Hindu restriction, the system has a case where a composite vowel is composed of a combination of three basic vowels, and recognizes the vowel by detecting that a continuous vowel button is pressed. However, when a combination (for example, _ (a −) + _ (i)) regarded as impossible is recognized, this may be recognized as a promised composite vowel.

  In this case, in order for the user to input a combination of a long vowel, _ (a−), and a single vowel _ (i), “intentional language restriction release” or “language restriction release delay time” must be used. . That is, ** is input in the above example, and 0 is input after the word end function is activated for a certain period of time or intentionally. Similarly, since _ (a-) exists, it can be predicted that _ (a) 0 does not come continuously in the word. Even if _ (a) may continue infrequently, the Indian language restriction input method in which the system is recognized as “** = _ (a−)” is basically applied, and _ (a) In order to input words one after another, intentional language restriction release or language restriction release delay time may be utilized, that is, after * is input, word termination is activated by delay time or the like, and then * is further input.

  Grouping into arbitrary groups of vowels on the assumption that 10 Indian vowels (or 11 vowels including vowel_ (ri)) may or may not be consecutive And the iterative selection method can be applied with a consonant separation keypad.

  For example, since there are three long vowels corresponding to three single vowels, it can be expected that at least three single vowels may or may not be continuous in a word. Therefore, vowels are divided into three groups and assigned to each button. Three single vowels are assigned to each group, three long vowels are also assigned to each group, and the remaining four vowels (vowels) are assigned. When _ (ri) is included, 5 vowels) are appropriately divided into three groups and assigned to each group. Here, single vowels, long vowels, and compound vowels grouped in each group can be selected in a predetermined order by the iterative selection method. If the selection is made in the order of “single vowel-long vowel-complex vowel” according to the number of button pushes, the input of the single vowel and the long vowel is the same as described above, and only the input of the composite vowel is slightly different. Since single vowels and long vowels have strong pronunciation similarity, it is natural to input them with one stroke and two strokes. Input multiple vowels by repeatedly pressing the same button.

  The number of vowel groups and the grouping of vowels are arbitrarily determined, and the order in which vowels belonging to the same group are selected is also arbitrarily determined. The vowels of each group are assigned to arbitrary buttons, but only some vowels can be displayed on the keypad. In this case as well, if there are rare cases where Hindu vowels come continuously, it can be input by intentional language restriction release or language restriction release delay time.

  In Hindu, the main syllable structure is a form in which consonants and vowels alternate, but consonants may come in succession. If another consonant comes after a consonant ending with a vertical line, it is combined with the next consonant to form a combined consonant without the vertical line. A consonant without a vertical line is combined with the next consonant with a symbol similar to “\” under the letter. There are some other rules, and they may be irregular. In this way, when it is recognized that the consonant is continuously input (temporarily recognized or deterministically recognized), it is easy to implement an automata to display with the combined consonant corresponding to the combination, It is also implemented in existing Indian word processors.

  Even in countries that do not use English as their native language, the English character is also used on the keypad to input English. In India, Hindu and English are used as common languages, so assigning Hindu consonants and English consonants with similar pronunciation to the same button can add simplicity to the use of simple codes and character input. it can. For example, as follows, various modifications are possible.

のみを表示することができ、又は１つの字母のみを表示することができる。各グループにおいて、ボタンプッシュ回数に応じて選択される順序は、任意に定めればよい。母音＿（ｒｉ）を、ＲＬが割り当てられるボタンに割り当てることもできる。 In the above example, in the case of g, k, etc., an English vowel separation keypad can be configured by assigning it to the same button as the Hindu letter corresponding to the pronunciation relatively easily. In the above, English consonants such as z and w can be additionally or separately arranged in an appropriate group as in the case of Chinese. In addition, in order to simplify the arrangement on the keypad, only some characters in each group can be displayed on the keypad (application of the concealed iterative selection method). For example, in the grouping exemplified above, when assigning the first group to an arbitrary button,

Can be displayed, or only one letter can be displayed. In each group, the order of selection according to the number of button pushes may be arbitrarily determined. The vowel _ (ri) can also be assigned to a button to which RL is assigned.

  To simplify the arrangement on the keypad, only English vowels (English vowel vowels) can be written. As can be seen from the above example, Hindu vowels have strong similarity in pronunciation with English vowels, so it is very natural to write only English vowels. As a result, in the case of consonants, as shown in the comparison table of Hindu consonants and English vowels, / GK /, / MN /, / DT /, / BP /, The keypad can be further simplified by displaying only English consonants without omitting the Hindu consonants corresponding to the / RL / ... group. Of course, even if Hindu characters are omitted in this way, the Hindu language restricted input mode is similar to the Chinese restricted input mode in which the pin sound could be entered without ambiguity in the Chinese example. Now, Hindi is input without ambiguity due to Hindu restrictions.

6.14 Language restriction input method with semi-consonant separation consonant separation keypad The semi-consonant (ie, semi-vowel) is slightly different for each language. In English, w, y, j, etc. are generally handled as semiconsonants, and in some languages, j is excluded, and in some languages, v, etc. are included. Generally, it is determined that w and y are handled as semiconsonants in common in almost all languages.

  Here, since a semiconsonant has a property and a sound value similar to a vowel, arranging the consonant in a separate group is an extension of grouping consonants based on similarity of pronunciation.

  In addition, there are English alphabets that are not used in a specific language for each language (for example, k and w that are used only for foreign words in Spanish), and other alphabets (for the sake of convenience, the applicant can modify Used as a letter or additional letter). Unused English letters are also required for universally used English input. Thus, if there are not many Roman characters that are not used in a particular language, they can be grouped with semiconsonants and assigned to the same button. Reference can be made to FIG. 10-6 in Chinese.

  In addition, letters that are not used in a specific language can be grouped together and assigned to the same button.

6.15 Language Restriction Input Method with Incomplete Consonant Separation Keypad In the present invention, a vowel separation keypad optimized for each language is constructed and an iterative selection method is applied to eliminate ambiguity or Can be minimized. Further, by configuring the consonant separation keypad in this way, the algorithm of the input system can be simplified. It is obvious that various modifications can be made based on this. Therefore, even when this is slightly modified and consonants are assigned to some buttons together, this is a modified category of the language restriction input method in the consonant separation keypad presented in the present invention. For convenience, this is called an “incomplete consonant separation keypad”, and a button to which consonants are assigned together is called a “consonant mixing button” for convenience.

  For example, the vowel “i” and the (half) consonant “y” are grouped together in the consonant separation keypad, assigned to specific buttons, and consonants (here semiconsonants) and vowels in a predetermined order according to the number of button pushes. Can be selected (the remaining buttons are separated into consonant buttons and vowel buttons). If there is no continuous case like “iy” or “yi”, it is possible to input a Chinese pin sound without ambiguity by applying the Chinese restriction input method. In this case, it is assumed that the consonant buttons are selected in the order of “y” (1 push) and “i” (2 pushes) according to the number of pushes of the specific button assigned by mixing the consonants. After that, at the moment when this consonant mixing button is pressed once, the input of the previous consonant button is determined as the input of the consonant, so the system responds to the consonant mixing button being pressed once. It can be determined that the vowel “i” has been input without considering that “y” has been input temporarily. This is because, in Chinese, it is possible to use a language restriction in which a consonant “y” cannot come after a specific consonant.

  As another example, the vowel “i” and the consonant “j” or the vowel “i” and the consonant “k” are grouped into the same group, and the (language restriction) iterative selection method can be applied. If the Chinese pin sound is input continuously such as "... ik ..." or "... ki ...", ambiguity may occur. Assigning consonants and vowels together to some buttons in this way without any special reason complicates the input algorithm, reduces the efficiency of the input system configuration, and language-limited input on a complete consonant separation keypad Better than applying the method. However, due to the characteristics of each language, consonant mixing buttons can be placed on some buttons for special purposes such as assigning consonants to all 10 numeric buttons (see Korean example).

  This can be applied by modifying the flowchart of FIG. 10-7. Refer to FIG. The same is true when there is at least one consonant mixing button, and the same is true when at least two consonants are assigned to one consonant mixing button. Applying language restrictions with an incomplete consonant separation keypad is a bit more complicated (see the case of “i” and “y” above), applying the consonant-vowel combination rules for each language. As a result, the flow diagram with the detailed application of the language restrictions for each language can be more complex than FIG. 10-9. FIG. 10-9 may be considered to be a partial modification of FIG. 10-7 and a very simple representation of general contents only.

  For example, when inputting Chinese pin tones with the standard English keypad of FIG. 5-1, since five vowels are assigned to different buttons, vowels can be recognized without ambiguity. In the case of a consonant, if a consonant comes after “˜vowel + n”, it can be recognized without ambiguity. When “˜vowel + n + m” is input in “˜n + X” (“X” is a voice consonant displayed as an English consonant), a vowel (o) that can be combined with the vowel “o” assigned to the [6] button is combined ( For example, when a vowel 'a') comes, it can be temporarily recognized as “˜ao”. However, since a normal Chinese pin syllable does not end with “m”, the previous input value can be recognized as “˜anm + vowel” by looking at the vowel recognized by the next input value. If “n + m = 666” is input next to a vowel that cannot be combined with “a” (for example, “e”), “−eo” is used instead of “˜eo” using a language restriction that cannot be combined with “eo”. ~ Emm "(because '~ emn ~' is impossible in Chinese pin tones).

  When a consonant comes after a syllable ending with "~ vowel + ng", ambiguity may occur. For example, “˜ngg˜” can be recognized as “˜nh˜”. In the case of Chinese, the system can search the index for all Chinese characters matching “˜ngg˜” and “˜nh˜” and provide them to the user. In the case of “˜ngh + vowel˜”, it can be recognized as “˜ni + vowel”. If the vowel recognized one after the other is a vowel that cannot be combined next to “i” (for example, “u”), the input value can be recognized as “˜nghu”. It is possible to search and process kanji characters corresponding to all of “˜ngh + vowel˜” and “˜ni + vowel˜”. In addition to this, when some ambiguity occurs, all words that match the interpretation result as much as possible can be searched and provided to the user.

  When inputting the Chinese pin sound by applying the Chinese restriction repeated selection method in the standard English keypad of FIG. 1-1, the case of “˜ngh + vowel˜” may be recognized as “˜ni + vowel”. It was. In this case, it is possible to search for a phrase that matches both of them from the pin sound index and process it. When all the full codes (full codes in the iterative selection method based on FIG. 1-1) are stored in the pin tone index, the input value may be generated in the same manner as the Chinese restricted iterative selection method is applied. The system can recognize matching phrases by searching the index for all possible cases (eg, “˜ngh + vowel” and “˜ni + vowel”).

  For example, in the case of applying the iterative selection method in FIGS. 1-1 and 5-1, “222” can be recognized as “aaa”, “ab”, “ba”, “c”. Accordingly, in FIG. 1-1 and FIG. 5-1, “beijing = 22334445444464”, and here, a continuous number group can be recognized as a combination of a large number of characters. A continuous number group “22 to” can be interpreted as “aa” or “b”. “˜33˜” can be interpreted as “dd” or “e”. Therefore, “2233˜” can be interpreted as “aadd˜”, “aae˜”, “bdd˜”, or “be˜”. As the input value increases, the number of possible combinations increases geometrically. A combination of characters that can be interpreted with respect to the input value is referred to as a “possible character combination” or “candidate character combination” for convenience.

  However, when "22 ~" is pressed, it can be interpreted as "aa ~" or "b ~", but the system confirms that there are no more words in the index starting with "aa ~" by referring to the index. Then, the system can process the input values “22˜” into “b˜” except for the combinations starting with “aa˜”. This is similar to the fact that the input value “22˜” can be interpreted as “b˜” by using the Chinese restriction that the Chinese pin sound does not start with “aa˜”. Therefore, when "2233 ~" is pressed, the index is searched only for the two combinations of "bdd ~" and "be ~", and there are more words in the index that begin with "bdd ~". If the system confirms that it does not, the combination starting with “bdd˜” is not considered any more, and only “be˜” may be processed according to the next input value. Again, it is determined that the input value “2233˜” is processed as “be˜” by using the Chinese restriction that “bdd˜” is impossible in the Chinese pin sound. The input value “22334˜” can be interpreted as “beg˜”. In addition, the input value “223344˜” can be interpreted as “begg˜” or “beh˜”, but the pin tone index is searched to confirm that there are no more words starting with “begg˜”. At this moment, the system processes the input value “223344˜” to “beh˜”. This is a Chinese restriction input method (for example, an input method using a Chinese restriction in which “beh˜” is not a valid syllable based on a Chinese pin sound generation rule but only “beh˜” is a valid pin syllable. ) Is judged to be similar. If "4" is pressed once more and the input value becomes "2233444 ~", it can be interpreted as "behg ~" or "bei ~", but after searching the index, there are more words starting with "behg ~" At the moment when the system confirms that it does not exist, the input value can be processed into “bei˜”, and since this is a pin sound combination that cannot be “behg˜” in the Chinese restriction input method, the input value “2233444˜ “Is interpreted as“ bei˜ ”and processed in a similar manner.

  In a system that stores all the words to be input and full codes, the number of candidate character combinations is simply 2 * 2 * 4 * 1 * 4 * 2 * 1 = 128 in “beijing = 223344445444664”. . Of course, it is also included in the scope of the present invention to process all the 128 candidate letter combinations by comparing them with the index and processing the matching words. However, in a system in which the system can recognize an input value from the initial stage of input, a combination of letters not existing in the index as the input value increases (in the above example, 'aa ~', 'bdd ~', 'behg ~ It is useful to remove 'etc.) at the beginning of input.

  As described above, the system can search only a part of candidate character combinations existing in the index among a large number of candidate character combinations generated as the same input value in the iterative selection method continuously increases. If it is confirmed that all candidate character combinations based on the input values do not exist in the index, the system can regard the input value as a promised type of simple code and process it.

6.16 Application of non-hidden control processing method In the above consonant separation keypad, any button (button in 3 * 4 keypad or external button) is defined as a control button, and only representative characters are applicable buttons. The remaining characters can be input by the control processing method. For example, in the keypad of FIG. 10-6, “w, y, v” assigned to the [0] button is assigned to an arbitrary consonant button, the [0] button is used as a control button, and “p = b + 0 = 10 "can be input (when input is applied after control). This is the content already mentioned in the previous application.

7). Parallel input method judging by the length of input value The length of the shortened code stored in the index may be less than a specific number in some cases. Also, in specific cases, types of abbreviated codes input by the user or by the system (for example, type 1 (city name), type 2 (bank name),..., Type 1 and type 2 in the previous application are grouped in a tree form. Specified). Assuming that the type of abbreviated code entered in type 1 (city name) is specified and the length of the abbreviated code of the city name is 3 or less, when applying the parallel input method using the abbreviated input method as the basic mode, At the moment when the length of the input value exceeds 3, the input value of the system can be regarded as a full code and processed. Conversely, when the full priority parallel input method is applied, if three input values are input and a word end function (for example, blank) is input, the system input value can be regarded as a simple code and processed. .

  This is particularly useful when the length of a shortened code of a specific type (city name-Beijing, company name-four-way group,...) Such as Korean or Chinese is less than a certain length (Korean) In many cases, certain types of nouns in Chinese, etc. are formed with a certain number of syllables or less, and it is natural to use them as syllable-based initial code shortening codes. Since the part is formed of two syllables, the length of the syllable standard initial code of the bank name is 2).

  This means that it is not necessary to determine whether the input value is an abbreviated code or a full code at the initial stage of input, that is, whether the length of the input value is greater than or less than a specific length. This is useful when it is possible to determine whether the input value is simple code or full code. That is, each time an input value is input, it is only necessary to check whether the length of the input value is equal to or longer than the specified length without searching the abbreviated code index. Performance can be improved.

8). Language-restricted parallel input method The language-restricted parallel input method is the system that inputs when the input value violates the "word generation rule" of a specific language (that is, the moment when the language restriction is violated) when the full priority parallel input method is applied. This is a method of processing by regarding the value as a simple code. If it is confirmed that there is no value in the simple code index that matches the input value, the input value can be regarded as a full code and processed again.

8.1 Language Restriction Parallel Input Method Applying Language Restriction Iterative Selection Method as Full Input Method 8.1.1 Chinese In particular, Chinese syllable reference initials based on FIGS. 10-1 to 10-4 are consonants. It has only the numerical values [1] to [9] which are buttons (the vowels a, o, e may be used alone, but it is about an exclamation, and there is almost no vowel 1, u, e. When u is used alone, y, w, and y are respectively preceded). In most cases, the full code to which the iterative selection method is applied in the examples of FIGS. 10-1 to 10-4 is [*], [0] for vowel input as the second or third input value. ] Or [#] button needs to be input. As mentioned in the previous Korean example (method using vowel elements), this property is used when the “parallel input method” is applied and the input value forms a full code at the beginning of input. Provide good characteristics that can be grasped.

  For example, as a full input method, an iterative selection method based on FIGS. 10-1 to 10-4 is used, and a parallel input method (full priority parallel input method) in which the full input mode is applied as the basic input mode is used. When the value “693... = Shk...” Is input, the input value cannot form a full chord at the moment when the third value [3] is input (can be up to “sh”, but the subsequent consonant “k” The system can recognize (because it can't), so the system can process the input value as a simple code. Similarly, when “112... = Pd ...” is input, the input value cannot form a full code at the moment when the third input value [2] is input (“11” is “bb” in Chinese input). It must be interpreted as 'p', and the system recognizes that the syllable consonant 'd' or 't' after 'p' cannot form a Chinese syllable). The system can process the input value as a simple code. Similarly, when “7771... = Wb...” Is input, the system can regard the input value as a simple code and process it at the moment when the fourth input value [1] is input. Finally, in most cases, when the second or third input value is input, the system can recognize whether the input value is a full code.

  When “14... = Bj...” Is input, the moment when the second input value [4] is input, the input value cannot be used to form a valid Chinese syllable. The system recognizes that the sound is not coupled) and can treat the input value as a simple code. That is, in FIG. 10-5, the system can provide the user with “Beijing” or “Beijing”, which is the corresponding phrase of the simple code “14”. If the simple code index does not have a simple code starting with “14” or “14”, the system assumes that the input value is a full code again and provides “bj” to the user.

をユーザに提供する。 This is the same as in the case of Korean using vowel elements (Fig. 4-5). The full input method in FIG. 4-5 already described is considered to be a Korean restricted repeat selection method. When the full priority parallel input method is applied, if “12” is input, a Korean syllable whose input value is valid is selected. Since it cannot be formed, the system regards the input value as a simple code, refers to the simple code index, and provides the user with a phrase corresponding to the simple code. However, if it is determined that there is no longer a simple code that matches “12” in the simple code index, the system again considers the input value as a full code,

To the user.

  In order to support understanding, the simple codes in the simple code index of FIGS. 5-4 and 10-5 are in an aligned state, but may actually be stored in any form within the system. However, during a simple code index search, the system may sort the simple code if necessary and check whether the input value still exists in the simple code index.

  Further, as a basic input mode, a shortened input code (assuming that only a syllable reference initial code consisting of [1] to [9] is stored in the simple code index to be searched with reference to FIG. 10-1) In the parallel input method (short-priority parallel input method) to which the input is applied, at the moment when the second value [*] of the input value “1 *. The value can be recognized and processed as a full code. This is also the same as the case of Korean in FIG. 4-5.

  5-4 and 10-5, the pin sound / kanji index and the simple code index have been described as one index. However, as in FIG. 5-1, a pin sound / kanji index (referred to simply as “pin sound index”. An index storing “pin sound + pin sound corresponding kanji”) — for searching for the corresponding kanji from the input pin sound. ) And simple chord / pin tone index (simply called “simple chord index”. “Simple chord + pin chord for simple chord + pinned kanji” or “simple chord + pin chord for simple chord”) It is obvious that a stored index) can exist separately.

  As shown in FIGS. 5-4 and 10-4, it is not always necessary to store only the syllable reference initial code in the simple code index, and other types of simple codes such as consonant reference initial codes are stored together. It is also possible. Similarly, the syllable reference initial code index, the consonant reference initial code index, and the overall associative simple code index can be implemented according to the types of simple codes.

  As shown in FIG. 10-10, it is assumed that “Beijing” has a syllable reference initial code “14”, a consonant reference simple code “1473”, and an overall associated simple code “1 * # 4 # 73”. In the keypad of FIG. 10-10, when the Chinese restricted full priority parallel input method is applied, the second input value “7” in the case of the syllable reference initial code “14” and the consonant reference simple code “1473”. At the moment when is pressed, the system regards the input as a simple code and can search and process the index. By the way, in the case of the entire associated simple code “1 * # 4 # 73” of “beijing”, when the input value “1 * # 4...” Is interpreted as a full code, it becomes “baiz. Does not violate pin tone generation rules. However, “baiz...” (And “baij...” —Because “baij... Can be made according to the next input value) is not a pin sound that exists in the pin sound index of FIG. Therefore, the system interprets the input value as a full code based on the input of each input value, and each time the input value is recognized as “b... Ba... Bai. Look up the phonetic index to find a matching phrase. As shown in FIG. 10-10, a matching pin sound can be searched until “bai˜”, but at the moment when the input value is recognized as “baiz˜” (and potential “baij˜”), the system It can be confirmed that there are no more matching words in the pin tone index (see (1) in FIG. 10-10). Therefore, the input values “1 * # 4...” Up to “baiz˜” are regarded as simple codes, and matching phrases can be searched with reference to the simple code index ((2) − (in FIG. 10-11). A) and (2)-(B)). Therefore, the input value “1 * # 4...” Up to “baiz˜” is regarded as a simple code, and a matching word / phrase can be searched with reference to the simple code index ((2) − (in FIG. 10-11). A) and (2)-(B)). Here, even in the simple code index, if a phrase that matches the input value “1 * # 4 #...” Cannot be found, the system further regards the input value as a full code and processes it to “bai…”. Can do.

  Due to the characteristics of Chinese ideographs, all pin sounds and kanji index (pin / kanji index) must be stored in the system, so even if the input value forms a valid pin syllable, the pin sound / At the moment when it is confirmed that it does not exist in the kanji index, this means that the system treats the input value as a simple code. This can be seen from the same context that an input value is treated as a simple code at the moment the system recognizes that the input value does not form a full code.

  If there are other types of simple chords for the same word, they should have simple chords in one of multiple catalogs for one pin sound as shown in Fig. 10-10 or Fig. 10-11 Is all possible. When processing the input value as a simple code (that is, processing the input value as a shortened input), the processing of (2)-(A) or (2)-(B) in FIG. 10-11 is selected. Is possible. If the processing steps (2) to (B) in FIGS. 10-11 are used, it is not necessary to have a simple code in the index. In the process (2)-(B), a combination of pin sounds that can be input at each input value is compared with a pin sound of an index, and a matching pin sound is searched. In the combination of pin sounds that can be entered, two letters are assigned to the [1] button, and three letters are assigned to the [*] button. It can be easily understood that six combinations (= 2 × 3) of letters assigned to each button are possible. For convenience, this is referred to as “simple codeable character combination” or “possible character combination”. In FIG. 10-11, since two letters are assigned to the [#] button, when 12 buttons are pressed up to “1 * #”, 24 combinations are pressed up to “1 * # 4”. Combinations of these are possible. By the way, when implemented in the system, since there is actually no pin sound beginning with “poi ...” in the Chinese pin sound dictionary, combinations below “poi ...” can be excluded from consideration. That is, “poiz...” And “poij...” That are combinations of “poi. Similarly, it is possible to search for a pin sound that matches the input value while excluding not only “poi...” But also combinations that do not exist in the pin sound index among possible character combinations.

  To use the processing steps (2) to (A) in FIG. 10-11, a simple code must be included in the index, but processing may be performed even when an irregularly defined simple code is used. it can. The use of the process of (2)-(B) means that simple codes do not need to be stored in the index, but simple codes having a certain rule (for example, totally related simple codes) are input. A premise is necessary, and the system needs to know in advance the promised simple code type (for example, the entire associated simple code).

  When the value interpreted as a full code does not exist in the index as shown in FIGS. The value can be processed as a simple code (ie, a shortened input value). When applying the full-priority parallel input method, FIGS. 10-12 show procedures that can be generally applied to Chinese input. This can be equally applied to a full priority parallel input method using a syllable reference initial code or a consonant reference simple code. For example, when a syllable reference initial code is input, the system interprets this as a full code and searches the pin tone index, which in most cases is the moment when the second or third input value is input. This is because the system can recognize that there are no more input values in the pin tone index. FIGS. 10-13 show that a procedure has been added to check if the input value constitutes a valid pin syllable before searching the pin tone index using Chinese linguistic characteristics.

  Not only in Chinese, but also in other language input systems, when the system stores an index of all words to be input, similarly as in the procedures in FIGS. 10-12 and 10-13, Each time an input value is entered, it interprets the input value as a full code and searches for words in the index to see if they match, and if there are no more matching words, the input value is It can be processed as a shortened code.

8.1.2 English, etc. Starting with “st ~” or “sp ~”, when the input of the fourth consonant from the start of the first word is recognized, and not starting with “st ~” or “sp ~” At the moment when the input of the third consonant from the start of the first word is recognized, the system can treat the input value as a simple code and process it.

  For example, when the simple iterative selection method is applied with the keypad of FIG. 1-1, when “467... = Gms...” Is input, the moment when the third input value [7] is input, the input value is an English word. Since the generation rule is violated, the system can process the input value as a simple code. If the entire associated simple code “4678255 = install” is stored in the simple code index, the system can provide the user with install, which is a simple code corresponding phrase. If there is no simple code that matches “467...” In the simple code index of the search range, the input value can be regarded as a full code and processed again.

  And, as described above, when using the English restricted repeated selection method as the full input method with the consonant separation keypad, the input value does not start with an English input rule (for example, a word does not start with “st˜” or “sp˜”). 3 consonants cannot appear as long as it is easy to know whether or not they violate, so the parallel input method can be applied more efficiently.

8.2 Language Restriction Parallel Input Method Applying Representative Character Exclusion Control Processing Method as Full Input Method In Figure 1-1, enter the first character of each button as the representative character with one stroke of the corresponding button, and the remaining characters When using the control processing method (that is, applying the representative character exclusion control processing method), use the word generation rules described in the case of English, and treat the input values as simple codes when the word generation rules are violated. A parallel input method can be applied.

  For example, in FIG. 1-1, when the first character is a representative character and the remaining characters are input by the control processing method, for example, when “467... = Gms...” Is input, the third input value [7] Since the input value violates the English word generation rules at the moment when is input, the system can process the input value as a simple code.

  Similarly, the present invention can be similarly applied to languages other than English, and the remaining items are similar to a language-limited parallel input method in which an iterative selection method is applied as a full input method.

8.3 Third-order ambiguity in the language-limited parallel input method When the entire associated simple code “72673” of “scope” is input, if the system first recognizes it as a full code, “72673 = pampd ”So that the input value does not violate the English word generation rules, so the system will first recognize the input value as a full code, and secondarily interpret the input value as a simple code. A “third-order ambiguity” that can be recognized as a scope corresponding to a code is generated. This is a case where a simple iterative selection method based on FIG. 1-1 is applied as a full input method.

  In this case, the system primarily provides the user with pampd. However, when the user desires to input a scope, a scope can be selected by pressing a specific button (for example, the ↓ [v] button). Alternatively, after inputting “72673”, when the word input is completed (for example, when a blank or the like is input), “pampd” in which the input value is primarily interpreted as a full code and “scope” in which it is interpreted as a simple code. "Is provided to the user in a list so that the user can select.

  In this case, the system primarily provides the user with pampd. However, when the user desires to input a scope, the user can press the specific button to select the scope. Alternatively, after the input of “72673”, when the word input is completed (for example, when a blank or the like is input), “pampd” in which the input value is primarily interpreted as a full code and “scope” in which it is interpreted as a simple code. To the user in a list so that the user can select.

  The above case can be similarly applied to a parallel input method that uses the representative character exclusion control processing method as a full input method. Also, in the shortened priority parallel input method that is not the full priority parallel input method, even if the input value exists in the simple code index after the input value is input and the input of the word is completed, the input value is converted to the full code. If the language restriction is not violated (when tertiary ambiguity occurs), the user can select it (by repeatedly pressing a specific button or selecting from a list).

  The same can be applied to languages other than English (particularly in the case of Japanese where the representative character exclusion control processing method is applied).

  In order to select one from a large number of candidates, any one of frequently used buttons among the movement buttons can be used to select a target word / phrase among candidate words / phrases output by the system. It has already been explained that this can be applied not only in the case where tertiary ambiguity occurs, but also in all cases where ambiguity occurs.

9. Parallel input method using example characters with different language restrictions Explained that the input value (full code or simple code) can be interpreted by the client side and also by the server side. FIGS. 11A and 11B are drawings of the prior application showing that they are interpreted on the client side and the server side, respectively. That is, the “system” here is a concept including both the client side system and the server side system. If the input value is a simple code, the client side interprets it, and when the corresponding words of the simple code are transferred to the server side as a text, the server side can use them to provide various services. Become. Similarly, when the input value is transferred numerically to the server side, the server side can interpret the simple code and use it for various services. Sending a DTMF tone from the client side and receiving and utilizing the input value on the server side is one example of transferring numerical values from the client side to the server side.

を念頭において“７７９９”を入力するとき、クライアント側システム内のシンプルコード索引に“７７９９”と“７７９９”に対応する

が記憶されていなければ、システムは単に

をユーザの入力とみなすことになる。万一、証券情報システムの場合、クライアント側が入力値“７７９９”をサーバ側に転送するか、又は入力値“７７９９”に対応する

をサーバ側に転送すると、証券情報サーバが“７７９９”を解釈することができるので、別に問題はないであろう（証券情報サーバはシンプルコード“７７９９”とシンプルコード対応語句である

を有していると仮定する）。しかし、クライアント側の文字入力モードでユーザが“７７９９”を入力した後、クライアント側が前記入力値をシンプルコードとみなして処理するが、この際、クライアント側のシンプルコード索引に“７７７９”が存在しなければ、それを再びフルコードと見なし、文字列

をサーバ側に転送する場合、証券情報サーバではそれを活用することが困難になる。 In the language restriction parallel input method, if the input value violates the language restriction (that is, if the input value cannot form a valid syllable of a specific language), the system treats the input value as a simple code. The simple code index may not have a simple code that matches the input value. In this case, the input value is processed again as a full code. For example, the user uses FIG.

When inputting “7799” with the above in mind, it corresponds to “7799” and “7799” in the simple code index in the client side system.

Is not remembered, the system simply

Is regarded as a user input. In the case of a securities information system, the client side transfers the input value “7799” to the server side or corresponds to the input value “7799”.

Is transferred to the server side, the securities information server can interpret “7799”, so there will be no problem (the securities information server is a simple code “7799” and a simple code corresponding phrase.

). However, after the user inputs “7799” in the character input mode on the client side, the client side treats the input value as a simple code, and at this time, “7779” exists in the simple code index on the client side. If not, consider it as a full code again and string

Is transferred to the server side, it is difficult for the securities information server to utilize it.

が意味なしの値であれば、

の入力に使用された“７７９９”を抽出し、シンプルコード“７７９９”の対応語句である

を追跡して、それをシステムで活用できるようにする。または、韓国語でシンプルコードとして音節基準イニシャルコードを使用する場合にのみ適用できる他の方法があり、すなわち、システムは、シンプルコードの対応語句のうち

と音節の初子音が一致する

を追跡し、それを活用してサービスを提供することができる。 Therefore, in the securities information server, the transferred character string

If is a meaningless value,

"7799" used for the input is extracted and is the corresponding phrase of the simple code "7799"

And make it available to the system. Or there is another method that can only be applied when using syllable reference initial codes as simple codes in Korean, ie, the system

Matches the first consonant of the syllable

Can be tracked and utilized to provide services.

の株価を照会するために

を入力しなければならなかったのに対して、ここでは、

のみを入力してサーバ側に転送しても、証券情報サーバシステムが、

が一文字の羅列であって、上場会社名に存在していないと認識し、前述した２つの方法（“７７９９”を抽出した後、

を追跡するか、又は“

と音節の初子音が一致する

を追跡する）によってサービスを提供することができる。 This is equally applicable to a personal computer environment that uses a keyboard that is not a keypad. In other words, in securities information systems that use existing personal computers as clients,

To query the stock price of

Whereas we had to enter

Even if you enter only and transfer to the server side, the securities information server system

Is an enumeration of one character and does not exist in the listed company name, and after extracting the above two methods ("7799"

Track or “

Matches the first consonant of the syllable

Service).

（シンプルコードの対応語句）”のみならず

を登録しておき、クライアント側から転送された値

が有効な韓国語音節を形成できないことを確認し、記憶されたシンプルコード対応文字と比較することで、ユーザが所望するのが

であることを把握することができる。 Other methods include “7799 (simple code) in the server side index,

(Simple code equivalent)

Is registered and the value transferred from the client side

Confirms that it cannot form a valid Korean syllable and compares it with the stored simple code-corresponding characters, the user wants

Can be understood.

  The above contents are summarized in the figure as shown in FIG. 11-3, and (A), (B), and (C) are selectively possible.

を入力しようとする場合、ユーザが

を入力すると、システムは索引を検索し、音節の初子音が

である

をユーザに提供することができる。前述した他の方法（約束された特定キーパッドを基準とした入力値がシンプルコード抽出及び索引検索、シンプルコード対応文字を索引に記憶）も、適用することができる。図１１−５を参考することができる。（Ａ），（Ｂ），（Ｃ）が選択的に可能である。 Furthermore, when inputting characters using a device such as a keyboard on an existing client such as a personal computer (for example, inputting characters using a word processor of a personal computer, etc.), the syllable reference initial code is used. By inputting only the first consonant and converting it into a target word / phrase in the system (client or server), high-speed character input is possible, or utilization in various information systems is possible. For example,

When trying to enter

The system searches the index and the first consonant of the syllable is

Is

Can be provided to the user. Other methods described above (input values based on a promised specific keypad are simple code extraction and index search, and simple code corresponding characters are stored in the index) can also be applied. Reference can be made to FIG. (A), (B) and (C) are selectively possible.

  Similarly, on the basis of FIG. 10-2, “Satongjipdan: Since the applicant does not know Chinese pronunciation, for the sake of convenience, Korean pronunciation is regarded as a Chinese pin sound. The syllable reference initial code “appropriately applied” is “6242” corresponding to s, t, j, and d. When the user inputs "6242", if 6242 and four groups (satongjipdan) are not registered in the simple code index of the client side system, the client side system only provides "sdzd" to the user. Become. When the user transfers “sdzd” to the server side, the server side receiving it recognizes that “sdzd” is not a valid Chinese syllable, and extracts “6242” which is the user input value from “sdjd”. The “four groups” corresponding to the simple code (here, the syllable reference initial code) “6242” can be searched (assuming that the server-side system stores the simple code and the corresponding phrase).

と各音節の初子音とが一致する語句を検索することができるように、図１０−２を基準として“ｓｄｊｄ”及びその他の変形可能な子音の組み合わせ（例えば、“ｘｄｊｄ”、“ｓｔｊｔ”，“ｓｄｊｔ”，…１６個の場合が存在）と各音節の初子音とが一致する単語をユーザに提供することができる。 In the case of Korean,

And a combination of “sdjd” and other deformable consonants (for example, “xdjd”, “stjt”, “Sdjt”,... 16 cases exist) and a word in which the first consonant of each syllable matches can be provided to the user.

  Also, if the corresponding character “sdzd” of the simple code is stored together on the server side, the value “sdzd” transferred from the client side is recognized as not a valid Chinese syllable, and the transfer value and index By comparing the corresponding characters of the simple code, the system can grasp that the user desires “four groups”.

  The above contents can be summarized as shown in FIG. 11-4, and (A), (B), and (C) can be selectively performed.

  The processing steps in FIGS. 11-3 and 11-4 can be performed on the server side or on the client side.

  When the user inputs “stjd”, an information communication device such as a personal computer can also search for a word (“four groups”) that matches the first consonant of a Chinese pin syllable and provide it to the user. Alternatively, if the first consonant “stjd” of each syllable is registered as a shortened code (corresponding character of a simple code) in the system (ie, bypass capacitor), when the user inputs “stjd”, the second consonant is At the moment of input, the Chinese language restriction is violated (when two Roman consonants that are not sh, ch, zh, etc. appear), so the input value is regarded as an abbreviated code (the user performs abbreviated input) Can be processed). Extracting the sample code from the input value “stjd” can extract a simple code (eg, 6242) using a keypad (eg, 10-2) promised in advance. 11A can be referred to, and (A), (B), and (C) are selectively possible.

  When a user inputs “stjd”, not only on a device equipped with a keypad but also on an information communication device such as a PC, a word (“four groups”) that matches the first consonant of a Chinese pin syllable is searched. Can be provided to the user. Alternatively, if the first consonant “stjd” of each syllable is registered as a shortened code (corresponding character of a simple code) in the system (ie, bypass capacitor), when the user inputs “stjd”, the second consonant is At the moment of input, the Chinese language restriction is violated (when two Roman consonants that are not sh, ch, zh, etc. appear), so the input value is regarded as an abbreviated input value (the user enters the abbreviated input) Can be processed). Extracting the sample code from the input value “stjd” can extract a simple code (eg, 6242) using a keypad (eg, 10-2) promised in advance. 11A can be referred to, and (A), (B), and (C) are selectively possible.

  (B) in FIG. 11-6 shows a word that generates an input value and does not generate a valid pin syllable (for example, “stjd” as “syllable reference initial value”) and each syllable of the phrase stored in the index. The first word (“sa-tong-jip-dan → stjd”) is searched for a matching phrase. It can be determined that the syllable reference initial value is used as a shortened input value. On the keypad, the syllable reference initial value corresponding to “stjd” is “622442.” As can be easily understood from FIG. It is also possible to provide an index of pin tones for the same, and similarly, when the input value is “consonant reference input value”, the system uses the word (character string) and the index to generate the input value. Memory Phrase consonant that can search for words that match.

  In summary, the type of input value to be shortened and can be regularly recognized is “syllable reference initial value” (eg, “beijing” syllable reference initial value is “144 = bj”), similar to the type of simple chord. ), “Consonant reference input value” (the consonant reference input value of “beijing” is “144773 = bjng”), “first vowel + syllable reference initial value”, “word reference initial value”, and the like. The system treats the input value as a shortened input value at the moment when the input value violates the language restriction.

  The types of “abbreviated input values” presented so far are summarized in the table of FIGS. 11-11. In FIG. 11-11, the shortened input value (A) includes (B) to (L), which are the short input values that have been presented so far that the system can recognize based on certain rules. Indicates the type. Also included are partially associated simple codes defined in association with any partial character and abbreviated input values consisting of any partial character. It can be seen that the “total input value” (Z) is a full coat of the full input method.

  As already indicated, if a certain condition is met (for example, the input value violates the full-code generation rule or does not constitute a valid syllable of a specific language), the input value is regarded as an abbreviated input value. When processing (ie, processing assuming that the input value is not the full code of the promised full input method) is the core content of the parallel input method, a shortened input value not presented in the present invention is used. Obviously, it is included in the category of the parallel input method of the present invention. 11-11, the shortened input values (H) to (L) can be used not only in devices equipped with keypads but also in devices such as personal computers as shown in FIGS. 11-5 and 11-6. FIG. 11B shows an example in which “syllable reference initial value” is used as a shortened input value. Although the term simple code or abbreviated code is used as a term for the abbreviated input value, the abbreviated input values (H) to (L) can be said to be simple codes having a comprehensive meaning. That is, (I) “syllable reference initial value” in FIGS. 11-11 can be said to be “syllable reference initial value simple code”, and the shortened input values of (H) to (L) can be said similarly.

  In FIG. 10, when “14 = bz” is input as the syllable reference initial value, the system searches the index for a phrase that matches “bz” with the consonant of each syllable. Here, when “4” is input again, “144 = bj”, so that a phrase (for example, “beijing”) in which “bj” matches the initial voice consonant of each syllable is searched from the index. Become.

  Furthermore, in the present invention, when any shortened input value among the shortened input values shown above is inputted, it is shown that it can be analyzed as several kinds of shortened input values. For example, a certain abbreviated input value is primarily treated as a predetermined abbreviated input value (for example, a syllable reference initial value), and secondarily abbreviated to another abbreviated input value (for example, a consonant reference input value). It is presented that it can be processed as third-order and other abbreviated input values, and can be analyzed as several kinds of abbreviated input values for one abbreviated input value. When the abbreviated input value “144 = bj” is input with reference to FIG. 10- *, the system first indexes a phrase (for example, “beijing”) where “bj” and the first letter of each syllable coincide with each other. Thus, regardless of whether or not there is a primary search result, it is possible to search for a word or phrase with a consonant match that is secondarily indexed.

  This indicates that the system can analyze any abbreviated input value presented in FIGS. 11-11 by the user. When this is applied, a large number of search results are obtained for one shortened input value. At this time, the output priority order depends on the type of the shortened input value to be applied. For example, first, each word searched by analyzing as a syllable reference initial value is output, and then each word searched by analyzing a consonant reference input value is output. If a user mainly uses a syllable reference initial value as an abbreviated input value, it is preferable to set so that each word analyzed based on the syllable reference initial value can be primarily output. Therefore, it is preferable that the user can specify the priority order according to the type of the shortened input value. If a user desires to use only the consonant reference initial value, the shortened input value should be set to be analyzed only as the consonant reference initial value. The output order of each word obtained by analyzing the shortened input value of the same type may be determined by factors such as the frequency of use as described above, and the output priority is applied to the analysis of the shortened input value. Regardless of the type of the abbreviated input, it may be determined by the frequency of use of each searched phrase.

の場合、これらはそれぞれ音節基準イニシャル値の“ｊｚ”，“ｚｊ”を有し、該音節基準イニシャル値はいずれも“４４４”である。この場合は、フル優先並行入力方法を適用するに当たり、システムは３番目の入力値“４”が入力される瞬間、入力値がピン音音節を形成できないことを認識することになる。図１０を基準として音節基準イニシャル値が同じボタンに割り当てられたそれぞれ別々の字母である場合は、できる限りすべての場合を検索する。すなわち、短縮入力値“４４４”に対して、“ｊｚ”と“ｚｊ”と一致する各音節を有する語句をすべて検索する。

In this case, these have syllable reference initial values “jz” and “zj”, respectively, and the syllable reference initial values are both “444”. In this case, in applying the full priority parallel input method, the system recognizes that the input value cannot form a pin syllable at the moment when the third input value “4” is input. When the syllable reference initial value is a separate letter assigned to the same button with reference to FIG. 10, all cases are searched as much as possible. That is, for the abbreviated input value “444”, all the phrases having each syllable that matches “jz” and “zj” are searched.

  Similarly, when the shortened priority parallel input method is applied, the input value can be regarded as the shortened input value of FIGS. 11-11 (B) to (L) and processed. Further, the input value is first regarded as a syllable reference initial value, processed secondarily as a syllable reference initial code, processed as a consonant reference initial value, and further processed as follows. Similarly, it can be processed as another shortened input value. The moment the input value is processed as a shortened input value and no more matching words are found in the index, the system treats the input value as a full code. Preferably, the type of abbreviated input value used mainly by the user can be designated in advance.

の場合は、音節基準イニシャル値を使用するときの問題を解決するために、捲舌音（そり舌音）の音節基準イニシャル値シンプルコードおよび音節基準イニシャルコードを次のように設定することができる。中国語における捲舌音“ｃｈ”,“ｓｈ”,“ｚｈ”は、実際に単一声母であって、注音符号では１つの記号で表記できるが、英語字母では２つの字母の組み合わせで表記できる。図１０−６によるｚｈｕｉｊｉａｎの音節基準イニシャル値シンプルコードも“ｈ”を含み、“ｚｈ＋ｊ”に対応する“４９４４”とする。音節基準イニシャルコードは“ｚｈ＋ｊ”に対応する“４９４”となる。図１０のキーパッドにおいて、捲舌音“ｃｈ”,“ｓｈ”,ｚｈ”を形成する初めの字母“ｃ”,“ｓ”,“ｚ”が、２番目の字母“ｈ”が割り当てられたボタンと異なるボタンに割り当てられているため、

の音節基準イニシャル値シンプルコードをどちらも“４４４”とした場合に発生した曖昧さがなくなる。

In order to solve the problem when using the syllable reference initial value, the syllable reference initial value simple code and syllable reference initial code of the tongue sound (sled tongue sound) can be set as follows: . The Chinese tongue sounds “ch”, “sh”, and “zh” are actually single phonetics, and can be expressed by one symbol in the phonetic code, but can be expressed by a combination of two characters in the English alphabet. . The Zhuijian syllable reference initial value simple code according to FIG. 10-6 also includes “h” and is “4944” corresponding to “zh + j”. The syllable reference initial code is “494” corresponding to “zh + j”. In the keypad of FIG. 10, the first letter “c”, “s”, “z” forming the tongue sound “ch”, “sh”, zh ”is assigned the second letter“ h ”. Because it is assigned to a button different from the button,

The ambiguity that occurs when both simple syllable reference initial value simple codes are “444” is eliminated.

の次に薄い色で表示された[. . .]は、“ｚ＿ｈ＿”である単語が多くある場合、後順位の他の単語が出力されることを意味する。後順位が“ｚｈ＿”である単語が出力できることはもとより、ユーザの好みによって“ｚｈ＿”である単語が先に出力できるように設定することもできるが、通常“ｚ＿ｈ＿”形態の単語が先に出力できるようにすることが便利である。 In the full priority parallel input method using the syllable reference initial value simple code as a shortened input value, when the input value “4˜” is recognized as “z˜” and the input value “49˜” is inputted, “zh˜” It is recognized and processed. However, since there are many words that begin with “zh˜”, there is no great meaning even if the system processes the input value “49˜” as “zh˜” and outputs the corresponding phrase to the user. Therefore, the system can analyze the input value “49” as a syllable reference initial value simple code and output and provide a phrase whose syllable structure is “z_h_” (the underlined portion indicates the position of the vowel). Means). If “49” is analyzed only as “zh”, it is difficult to abbreviate a word whose syllable reference initial value is “z_h_”. In the case of Chinese, most of the words consist of one syllable (letter) or two syllables (letters). Therefore, in order to shorten the input of the two syllable words, it is necessary to process the input of the tongue sound as a shortened input is there. Refer to FIG. 10-17 on the liquid crystal

[...] Displayed in a lighter color next to ”means that when there are many words with“ z_h_ ”, other words in the subsequent rank are output. Although it is possible to set the word “zh_” to be output first depending on the user's preference, the word “z_h_” type is output first, as well as being able to output the word “z__” in the rear rank. It is convenient to be able to do so.

  When a vowel input is recognized after “49” (for example, when a vowel button is pressed), the system processes the input value as “zh_” (ie, processes the input value as a full code). When a consonant input is recognized next to “49” (eg, when a consonant button is pressed), the system will treat the input value as “zh_X_” (capital letter “X” is an English word for any phoneme) consonant). That is, even in the full-priority parallel input method, the system uses a simple code (syllable reference initial value simple code or syllable reference) for the input of “ch”, “sh”, “zh”, etc. Initial code) and a full code or a simple code depending on whether the next input value is a vowel or a consonant.

  The above contents are similar to those obtained by considering the simple code of the consonant reference initial value as a shortened input value and applying the shortened priority parallel input method. However, the simple code of the syllable reference initial value is applied as the abbreviated input value, and the full priority parallel input method is applied, except that the word start state (a confirmation button is pressed or a word in the form promised by the input system) When a tongue-and-mouth sound is input at the end of processing, the input value is regarded as a shortened input value and processed (for example, the initial voice of the word is “z_h_” (another tongue-and-mouth sound such as “ch” or “sh”). In the case of, it searches for a word that matches).

  As described above, the description has been made with reference to FIG. 10, but it goes without saying that the above concept is applied to other keypads.

  Furthermore, the present invention can be applied not only to a personal computer keyboard and a keypad, but also to all modified keyboards similar to these (the keyboard forms embodied on the screen are extremely diverse).

“ｓｄ…”, “ｘｄ…”,“ｘｔ…”のリスト（２＊＊４＝１６種類が可能である）を有するとともに、入力値“ｓｄ…”に対して各レコードのこの多重値と比較してユーザに対して四通集團を提供することもできる。これも同じ概念であるが、具現が難しく、かつメリットがないので、薦めるものではない。図１１−９を参照する。両方法はどちらも入力値によって表現できる可能字母の組み合わせ（available character combination）と索引の語句（図１１−９ではピン音）との比較を含む。 10. Parallel input method using combinations of simple code-capable characters When using the syllable-based initial code as a simple code among the methods described above, comparing the first consonant of each syllable means that the client side has a simple code index (for example, The present invention can be usefully applied to a case where there is no 6242 / satangjipdan / four-way group) and only an index for a specific phrase (for example, satangjipdan / four-way group). For example, at the moment when the second [2] of “62...” Is input, the system recognizes that the input value violates the Chinese language restriction, and the input values “62... Sd..., St. xt... (referred to as “possible alphabet combinations” for convenience) ”and a phrase that matches the first consonant of each syllable in the index words can be searched and provided to the user. As the number of button pushes increases, the number of possible character combinations also increases, and this possible character combination is used to compare / search with an index phrase. Reference can be made to FIG. In the index of FIG. 11-7, the “four-way collection” line (record) is “stjp” in a multriple manner.

It has a list of “sd...”, “Xd...”, “Xt. Thus, it is possible to provide a four-way collection to the user. This is the same concept, but it is not recommended because it is difficult to implement and has no merit. Refer to FIG. Both methods involve a comparison between an available character combination that can be represented by an input value and an index phrase (pinned in FIG. 11-9).

を有せずただ特定の語句の索引

のみを有した場合、ユーザが

を念頭において“７７…”を入力するとき、システムは、２番目の［７］が入力される瞬間、入力値が韓国語言語制限を違背すると認識し、入力値

（便宜上‘可能字母の組み合わせ’と称する）”と、索引単語のうち各音節の初子音とが一致する語句を検索してユーザに提供することができる。図１１−８を参考することができる。 Similarly, the simple code index on the client side

Index of specific words without having

If the user has only

When inputting "77 ..." with the above in mind, the system recognizes that the input value violates the Korean language restriction at the moment when the second [7] is input, and the input value

(For convenience, it is referred to as “possible character combination”) ”and a phrase that matches the first consonant of each syllable in the index words can be searched and provided to the user. FIG. 11-8 can be referred to. .

  If it is confirmed that there is no longer a phrase that matches the input value in the index, the input value can be regarded as a full code and provided to the user. This is effective under the assumption that the user inputs only the value of the syllable reference initial code, and does not include a simple code in the index, and has a similar effect to the parallel input method with only the word index. That's what it means. However, this method is not limited to the moment when the system detects the moment when the input value violates the full code generation rule in the full priority parallel input method, or the moment when the input value violates the language restriction in the full priority language restriction parallel input method. Are identical and only differ in the search for the word corresponding to the input value. In other words, at the moment when the full code generation rule is violated or the language restriction is violated, the system considers the input value not to be a full code, but instead uses a simple code instead of the input value. Using the initial code, the phrase corresponding to the input value is interpreted (or searched) and provided to the user.

  The core content of the parallel input method is that the system automatically detects the moment when the full code generation rule is violated, or the moment when the language restriction is violated, and the input value is regarded as a simple code (that is, the user does not enter the full input). It is considered that the abbreviated input was performed). In the end, only the case where the phrase corresponding to the input value is searched by the parallel input method already presented is limited to a special case (use of a syllable reference initial code).

  As in the case of syllable reference initial codes, if there is regularity such as whole related simple codes, consonant related simple codes, first vowel + consonant related simple codes, and simple codes can be automatically extracted, The target phrase can be retrieved from the index by comparing "possible character combinations" with the index phrase. 10-11, the system analyzes the input value as a full code each time an input value is input one by one, and confirms that the pin code analyzed as a full code does not exist in the pin sound index. The target word / phrase can be searched by regarding the input value as an abbreviated input value and comparing the “possible character combination” with the pin sound. This is shown in FIG. 11-10, and FIG. 11-10 represents a part (part (2)-(B)) of FIG. 10-11. It is.

11. Full and parallel input methods on input systems that store full codes Specific language restrictions on specific keypads Full input methods can make it somewhat complicated to identify the correct words for input values This is as described above (for example, English input and Chinese pin sound input by a repetitive selection method in a standard English keypad). As described above for Chinese, if the system has an index of all words that are to be input not only in Chinese but also in other languages, the input value is interpreted as a full code primarily, If it does not exist, it can be interpreted as a secondary code type of simple code.

  Hereinafter, the case of Chinese will be described as an example for convenience. As shown in FIG. 10 to FIG. 14, if all of the full codes are stored in a system that stores all Chinese pin tones (words that are possible in other languages), the system simply inputs values. The target character can be recognized by searching for a phrase that matches. If the full-priority parallel input method is applied, the system will treat the input value as a promised type of simple code at the moment of confirming that there are no more matching full-code values in the index. become. In FIG. 10-14, the process of inputting “Beijing” can be regarded as a process of interpreting and processing the input value as a full code. However, since the full code is stored in the index, the full code is interpreted through a search of the index.

  When applying a full input method in which words can be identified without ambiguity by the language restriction input method, it is not always effective to store the full code in the index. However, it may be effective when an ambiguous full input method such as an iterative selection method is applied and all indexes of the words to be input are stored. When applying the full input method that can cause ambiguity, as described above in the section “Limited input method of language with incomplete character separation keypad”, all words to be input are stored in the index. If so, this full-code index can be used to process input values.

  As described above, the case of Chinese has been described as an example, but it goes without saying that the present invention can also be applied to cases other than Chinese. Further, not only the keypad of FIG. 1-1 but also the keypad of FIG. 10 and other keypads can be similarly adopted.

12 Parallel input method for processing input values as many types of codes (input values) at the same time A parallel input method for processing input values as a number of predetermined types of codes (input values) can be applied. For example, in FIG. 10-14, if the input value is considered as a full code (A) and an overall associated simple code (B) at the same time and processed, the system converts the input value “1 * #” into a full code. It is possible to recognize “bai” when interpreted and “bei” when interpreted as a totally associated simple code. In this case, the system can simultaneously give “bai” and “bei” to the user.

  In the case of the full priority parallel input method, only “bei” is output when interpreted as a full code. In the case of the shortened priority parallel input method, only “bai” is output as a result of interpretation as a promised shortened code. There is a difference in doing. Even in such a parallel input method, at the moment when it is confirmed that the input value does not exist any more in a specific type of code, it is possible to process the input value only as another type of code, After all, it can be said that this is included in the category of the parallel input method described above. The system index does not necessarily have to store the full code and the entire associative simple code as shown in FIG. 10-14, and the same applies when the input value is interpreted as several types of code at the same time. It is applicable to.

13. Priorities of words that cause ambiguity When a certain input method (for example, a specific full input method or a specific type of abbreviated input method) has ambiguity for the same input value, FIG. As in -14 and other figures, the target words can be output in the order based on the priority when ambiguity arises when applying a particular input method. For example, if ambiguity occurs when the full input method is applied in FIGS. 10-14, the target word / phrase is output to the user based on “(A) priority” which is the priority between the full codes (A) and recommended to the user. Can do.

  Hereinafter, a case where the input value is interpreted as several predetermined types of codes will be described. As described above, in FIG. 10-14, if the input value “1 * #” is interpreted as a full code, it becomes “bai”, and if it is interpreted as an entire associated simple code, it becomes “bei”. When ambiguity arises when interpreted as different codes in this way, when the system interprets simultaneously as a full code ((A) in FIG. 10-14) and a totally associated simple code ((B) in FIG. 10-14). A target word / phrase can be given to the user based on the priority order.

  For example, the input value is primarily treated as a full code, but the full code generation rules are not observed, the syllable generation rules of a specific language are not observed, or there are more full codes. In the parallel input method in which the input value is processed as if it were abbreviated code in the parallel input method in which the input value is immediately processed as a syllable reference initial code and a syllable reference initial value simple code. ”And“ Syllable reference initial value simple code ”can be processed based on the priority when ambiguity occurs (that is, the syllable reference initial code is (C) and the syllable reference initial value is simple). If the code is (D), output the target phrase based on “(C) + (D) priority”).

  In the above description, Chinese has been described as an example. Needless to say, the present invention can be similarly applied to other languages when a pin sound is used as a target word instead of inputting a Chinese character.

14 Parallel input method in which the control processing method is a full input method As shown in FIGS. 2-1 and 2-2, when the control processing method excluding the representative character is applied with the control buttons as [*] and [#] buttons. The parallel input method is applied, but if the shortened priority parallel input method is applied by storing all the words consisting only of the representative characters in the index, the system will instantly press “*” or “#” as the input value. As already pointed out, input values can be processed as full codes.

  In the case of Japanese shown in FIGS. 2-1 and 2-2, it is assumed that the abbreviated code consists of only numeric values. In this case, prepare an index (all may be included in the abbreviated code index or separate from the abbreviated code index) that stores all words consisting only of the representative characters, and input in parallel Applying the method, while only the numeric button is pressed as an input value, the result of interpreting the input value as a full code and a word index consisting of only representative characters (or a simple code index containing it) Any result that is referred to and interpreted as an abbreviated code can be given to the user as a target word / phrase based on a predetermined priority.

  For example, if the input value, that is, the input value “111” in FIGS. 2-1 and 2-2 is processed as a full code (that is, if the full priority parallel input method is applied), it is interpreted as “Oh”. However, if there is no word starting with “ah ~” and only words starting with “ai” or “ai ~” with reference to the index of all words consisting only of representative characters, “ “Ai” or “Ai ~” can be given to the user in preference to “Oh ~”.

  Please refer to FIG. 10-15 and FIG. 10-16. In FIG. 10-15, the input value “11” is interpreted as a full code, but since this is a word consisting only of the representative character, it is further retrieved by searching for “index of all words consisting only of the representative character”. To see if it ’s a bad word. Here, if “oh”, which interprets “11” as a full code, exists in the index, it may be recognized as an effective word for the input value. At the moment when it is confirmed that there are no more words corresponding to the index of all words consisting only of representative characters, the system further interprets the input value as a simple code and the system recognizes the corresponding “ai”. It will be. In FIG. 10-15, the system recognizes the input value “11” as “ai” with the highest priority. Depending on the next input value of “11”, there may be “Ai…” (“…” is the case where “Ai” is followed by another letter). Can be output as a ranking. Similarly, even if it does not exist in the index, the target word / phrase may be “Oh”, so this can also be output as the next rank after “Ai”. In FIG. 10-15, “Ai ...” and “Oh” are light-colored, and the second and third recommended orders are shown as the system can also output in this way. "" Can also be output.

  10-15, (1), (2), and (3) may be performed sequentially, as described in the column “Parallel input method for simultaneously recognizing input values as many types of codes”. , May be done at the same time. Similarly, the full code and simple code shown in the index are merely for the purpose of improving understanding, and it is clear that the same procedure can be performed even if the full code and simple code are not stored. (If the simple code is a simple code that can be interpreted regularly, such as a syllable reference initial code or a totally related simple code). As shown in FIG. 10-16, the present invention can be similarly applied to the case where the index of all words consisting of only representative characters and the simple code index exist as a single index. The material structure can be implemented in various forms in consideration of implementation.

  The operation result of the system is almost the same as the parallel input method described above, but there is a difference in that the full priority parallel input method or the general parallel input method is applied instead of the shortened priority parallel input method. .

  This is applicable not only to the parallel input method in which the control processing method excluding the representative character is the full input method but also to any parallel input method in which the control processing method is the full input method.

15. Kanji conversion Korean and Chinese kanji pin sounds and kanji have a 1: many correspondence. For example, there are many kanji characters corresponding to “courtesy” such as courtesy, courtesy, keenness, and so on.

  In the case of Chinese, there are Beijing, background, and the like as kanji corresponding to “beijing” (when no tone code is attached). Accordingly, it is possible to select the subsequent kanji (for example, Beijing (2nd) and background (3rd)) repeatedly by pressing the kanji conversion button. For example, after inputting “beijing”, the kanji conversion function is pressed once. Press again to get the background, or select from the list.

  In the case of Chinese, in order to input kanji, all kanji to be input (referred to as “target kanji” for convenience) are stored in the index in advance, so the target kanji is fixed even while inputting the Chinese pin sound. If so, it can be provided to the user. For example, if the word starting with “beij ...” is only “Beijing”, if only “beij” is entered, “Beijing” can be provided to the user. If only one exists, the system will convert “beijing” to “Beijing” at the moment when all of “beijing” is deterministically recognized even if the user does not determine the target kanji in the middle of input. be able to.

  In Japanese, hiragana and katagana have a 1: 1 correspondence, and hiragana or katagana and kanji have a 1: many relationship. Therefore, as described above, after inputting hiragana, Katakana conversion control (for example, “A / A” control in FIG. 2A) is input and converted to Katakana. In FIG. 2A, the deformed character can be input by the cross control processing method, and the “A / A” control can be selected with two hits of the [0] button. Here, each kanji corresponding to a specific hiragana or katakana can be selected by repeatedly pressing the [0] button. For example, if “A / A” control is specified as post-input, pressing [0] + [0] after inputting a hiragana word (or letter) will convert it to Katakana corresponding to the input hiragana, and again When the [0] button is pressed, it is converted to the first kanji, and when the [0] button is pressed again, it is converted to the next kanji. Conversely, in Katakana input mode, after entering Katakana, pressing [0] + [0] converts it to the corresponding hiragana, and pressing the [0] button once again converts it to Kanji.

16. Construction of simple code index Among the simple codes presented by the applicant, all-related simple codes, syllable reference initial codes, consonant related simple codes, initial vowel + consonant related simple codes, word reference initial codes, etc. are mechanically (ie, It has the good property that it can automatically extract simple codes from specific words (words or phrases). Therefore, it is possible to automatically construct a simple code index (an index made up of simple code and simple code-corresponding phrases) for words that the user has entered once by the full input method, and use it for the parallel input method. The simple code may be stored on the client side or the server side in an environment similar to FIG. 5-5, or may be stored on both the client side and the server side.

  Different types of simple chords (for example, globally related simple chords and syllable reference initial chords) can be generated for the same phrase. However, it is generally convenient to use a specific type of simple code, and it is fine if the user is aware of the type of simple code that is automatically generated and is used when applying the parallel input method. can do. When this is applied, there is an effect that the simple code index becomes rich as the user's character input increases.

  In order to extract a syllable reference initial code from a specific phrase of a specific language, it is necessary to separate the syllable from the specific word. Separation of syllables from specific words in a specific language is possible according to the specific language separation (syllable separation) rules. Since each language has a separation rule, it is easy to automatically separate a syllable from a specific word by using the separation rule. Therefore, detailed description is omitted here.

For example, “Zhongguo” in a Chinese pin sound can be syllable separated into “Zhong” and “guo” around the vowels “o” and “uo”, which is much more than the recognition process of the pin sound in the present invention. Easy to do. In separating a syllable from the front of a word by forward scanning, for example, the system does not know whether the consonant “n” next to “zho” is a consonant of the next syllable. It can be seen that the consonant “g” is a consonant belonging to the first syllable. Similarly, the system can also say that “g” can be a consonant of the next syllable, but looking at the next “g”, the first “g” is a consonant belonging to the first syllable, and the second “ It can be seen that g ″ is a consonant belonging to the next syllable (a syllable in a pin sound can end with “˜ng”). The first syllable “zhong” can be separated by scanning to “zhongg”. Next, the system can recognize that “u” is naturally a vowel belonging to the second syllable, and that the continuously displayed vowel “o” is also a vowel belonging to the second syllable (pin). English vowel “uo” can be displayed continuously by sound). Since there are no more letters to be displayed after “o”, the system can separate the syllable of “zhongguo” into “zhong” and “guo”, so that the initial value of syllable is “zg” Can be recognized, and it can be identified that the syllable reference initial code of “zhongguo” is “43” (reference to FIG. 10-6). Of course, for other languages, the system can automatically separate syllables from specific words or phrases based on the syllable separation rules of the language.

17. Target phrase analysis from words without special codes (subscripts) In the case of beijing, both “Beijing” and “Background” are supported simultaneously if no tone code is added to the vowel. In this case, a specific button can be repeatedly pressed to select a target word or phrase, or any one can be selected from a list.

  Similarly, in languages other than Chinese, you must enter a vowel without subscripts or subscripts, but if you enter only basic vowels, press a specific button repeatedly to search the index. Providing the user with a target phrase or searching the index and providing the user with a list of corresponding phrases with subscripted vowels allows the user to make a selection.

  This is useful for languages such as Vietnamese, which have a lot of deformed characters and complex tone codes. However, the system must have an index of words with subscripts attached to the basic characters. Similarly, such an index can be constructed from words already entered by the user. For example, when a character is input and the user inputs the word “^ + abc” (when the subscript “^” is added to “a”), the system stores “^ + abc” in the system (system In this case, both “abc” and “^ + abc” corresponding to “abc” may be stored.) After the user inputs “abc” again and presses a specific button, “^ + abc” is displayed to the user. It is possible to provide the user with a list of phrases corresponding to “abc” such as “^ + abc” when the user inputs “abc”. If "abc" is the target phrase, the user will not press a particular button or will select 'abc' when the system provides a list of related phrases. The list is provided when the input of the word is completed or at an arbitrary time during the input.

18. A parallel input method in which a part of a word (word or phrase) is input using the full input method and a part is input using the abbreviated input method. A method for outputting and allowing the user to make a selection is widely used. For example, it is similar to the “automatic completion function” of a window. Even at the time of full input (input by full code), an index (for example, an index for parallel input) is referred to, and a matching word / phrase (word or phrase) stored in the index is output and selected / selected by the user. It goes without saying that the decision is made.

と

などの単語が記憶されていると想定する。

まで入力されると、システムは

と一致する単語（すなわち、

から始まる単語として、

…）を適切な形態に出力することができる。携帯電話と類似した環境では入力値によって生成される内容は入力ライン（図１３−１において液晶の上端）に示され、索引を参照して出力する候補語句（単語又は句節）は下端にリスト又はリストと類似した形態に出力することができる。ユーザは移動ボタン（ネビゲーションボタン）などを利用して所望の単語を選択したし上確定して入力することができる。携帯電話液晶を例にとって図１３−１を参考する。図１３−１は、フル入力によって形成される音節と一致する単語を液晶の下端に出力する事例を示す。 Next, this will be described with reference to the input system of FIGS. 4-5 to 4-8. For example, in the index

When

Assume that words such as are stored.

The system will enter

Words that match (ie

As words starting with

...) can be output in an appropriate form. In an environment similar to a mobile phone, the content generated by the input value is shown in the input line (upper end of the liquid crystal in FIG. 13-1), and candidate words (words or phrases) to be output with reference to the index are listed at the lower end. Alternatively, it can be output in a form similar to a list. The user can select and input a desired word using a movement button (negotiation button) or the like. Reference is made to FIG. FIG. 13A shows an example in which a word that matches the syllable formed by full input is output to the lower end of the liquid crystal.

を入力するに当たり、

まで入力した後、残りの音節を短縮入力（例示において音節基準イニシャル値シンプルコードを利用した短縮入力）によって入力することができる。すなわち、

を入力する。ここで、最初の音節が

であり、残りの音節はイニシャル値が

に対応する単語（すなわち、

）をユーザに提供することができる。 here,

When entering

The remaining syllables can be input by shortening input (shortening input using the syllable reference initial value simple code in the example). That is,

Enter. Where the first syllable is

And the remaining syllables have initial values

The word corresponding to (i.e.

) Can be provided to the user.

の場合に“１＊０５７２３”を連続的に入力すると、

となる。この場合、システムは

までは正常なフル入力と認識するが、“２”が入力される瞬間、入力値の一部が正常な韓国語音節を形成しないと認識し、このため、音節を形成しなかった残りの部分（例えば、

）を短縮入力と認識することができる。これも先出願で述べられた“並行入力方法（フル（入力）優先並行入力方法）”の範ちゅうに属するものであり、フル入力によって認識された音節

は当然フルコードと認識し、残りの入力値（例えば、

）はシンプルコードと認識する。 However, about half of Korean syllables are syllables with finality (final voice), and the other half are syllables without final voice.

In the case of "1 * 05723" continuously input,

It becomes. In this case, the system

Is recognized as normal full input, but at the moment when “2” is input, it is recognized that some of the input values do not form normal Korean syllables, so the remaining part that did not form syllables (For example,

) Can be recognized as an abbreviated input. This also belongs to the category of “parallel input method (full (input) priority parallel input method)” described in the previous application, and syllables recognized by full input.

Is of course recognized as a full code and the remaining input values (eg

) Is recognized as a simple code.

が記憶されている場合）を、索引から検索してユーザに提供でき、シンプルコードと認識される入力値に対応する単語（例えば、索引に

が記憶されている場合）をユーザに提供でき、両者を一緒に出力することもできる。両者を出力する場合は、所定の約束（rule；例えば、フルコードと認識された部分を含む単語を優先的に、又は逆に、又は単語の優先順位に基づいて、又は特定のグループに属する単語を優先にして）により、特定の単語を液晶（ＬＣＤ）の上端（top）に出力することができる。様々な手段が可能であるので、好ましくは、ユーザにこのような出力順序を設定させる。 Where the corresponding word containing the syllables already formed by the full input (eg in the index

Can be retrieved from the index and provided to the user, and the word corresponding to the input value recognized as a simple code (eg, in the index)

Can be provided to the user, and both can be output together. When outputting both, a word that belongs to a predetermined rule (rule; for example, a word including a part recognized as a full code is given priority, or vice versa, or based on the priority of the word, or Can give a specific word to the top of the liquid crystal (LCD). Since various means are possible, it is preferable to have the user set such an output order.

に対応する

が液晶の下端(bottom)に出力されたとき、ユーザが

を選択すると、

全体が

に取り替えられ、

を選択すると、これに対応する

が

に取り替えられ、さらに

となる。 When outputting both,

Corresponding to

Is output to the bottom of the LCD, the user

If you select

Whole

Replaced by

Corresponding to this

But

Replaced by

It becomes.

だけが記憶されていると、入力値“１＊０５７２３”の連続入力に対応する

を出力することができない。したがって、システムは追加的に“１＊０５７２３”を

に同時に解析し、索引を参照して一致する単語を出力（ユーザに提供）することができる。

に解析することは、短縮入力の開始として認識された

の前の音節が終声子音に終わった場合、その音節を前の子音から除き（すなわち、

）、終声子音を一子音（unit consonant）として

との間に位置させる。これは少し複雑であるが、ハングル構造の基本的な知識を有するものであれば、容易に理解できる。終声が二重子音である場合も上記のことを同様に適用することができる。図１３−２は出力の優先順位として、“優先出力として指定された特定グループの単語−フルコードと解析された部分を含む単語−フルコードと解析された部分を含まない単語−…”の順序を適用して候補単語を出力した例を示す。ここで、

は優先出力として指定されたグループに属するもので、最上端（uppermost top）に出力されている。 Where in the index

Corresponds to the continuous input of the input value “1 * 05723”.

Cannot be output. Therefore, the system additionally adds “1 * 05723”.

Can be simultaneously analyzed, and the matching word can be output (provided to the user) by referring to the index.

Was recognized as the start of abbreviated input

If the previous syllable ends in a final consonant, remove that syllable from the previous consonant (ie,

), The final consonant as a unit consonant

Position between. This is a little complicated, but can be easily understood by those who have basic knowledge of the Hangul structure. The same applies to the case where the final voice is a double consonant. FIG. 13-2 shows an order of “a word of a specific group designated as a priority output—a word including a full code and an analyzed part—a word not including a full code and an analyzed part—... An example in which candidate words are output by applying is shown. here,

Belongs to the group designated as the priority output and is output to the uppermost top.

だけと解析しようとするなら、

まで入力した後、音節確定のための約束の操作（例えば、約束された所定の時間中には無入力、又は右矢印［＞］ボタンを１回押すなど）後、“５７２３”を押すことにより、

だけと解析され、それに対応する単語だけを検索することができる。この場合も既に認識された音節である

を含む

に対応する単語

を出力することもでき、

を除いた

に対応する単語（例えば、

が索引に記憶されている場合）を出力することもでき、両者を一緒に出力することもできる。図１３−３を参照する。 If the user just

If you just want to analyze

By pressing “5723” after a promised operation for confirming the syllable (for example, no input during the promised predetermined time or pressing the right arrow [>] button once). ,

Only the word corresponding to it is searched. In this case, it is already recognized syllable

including

Words corresponding to

Can also be output,

Excluding

The word corresponding to (for example,

Can be output), or both can be output together. Refer to FIG.

を入力した後に［＞］ボタンを１回押すと、音節

が確定され、［＞］ボタンをもう１回押すと、

となる。すなわち、“１０*>５７２３”が入力されると、システムはそれを

と認識し、それに対応する単語である

をユーザに提供する。もちろん、システムが図１３−３に示した索引を参照する場合、

、までのみ入力されてもシステムは

を出力することは言うまでもない。 Here, a syllable can be determined by pressing a right arrow button (hereinafter referred to as “[>] button”) once as an operation button for intentionally determining a syllable. Furthermore, the [>] button can be used as a space button. When the syllable is confirmed, the button is applied to blank input. That is, if the user

If you press the [>] button once after entering

Is confirmed and the [>] button is pressed once again.

It becomes. That is, if “10 *> 5723” is entered, the system

And the corresponding word

To the user. Of course, if the system references the index shown in Figure 13-3,

, Even if only entered until

Needless to say, is output.

、シンプルコードと解釈する入力値）にのみ対応する候補単語を出力する方法は、頻繁に使用される接尾語又は助詞などを短縮入力で迅速に入力するのに役に立つ。例えば、

まで入力後、（所定の手段によって入力内容を確定する）、“

（索引に記憶されているという仮定下で）”を出力するために

を入力する。前述の如く、

のように頻繁に使用される接尾語は、索引において別途のグループ（例えば、接尾語グループ）に記憶し、並行入力における候補単語の出力時、該当グループの単語を優先的に出力することができる。もし

の入力時、索引に

と

の両者が記憶されているなら、

と

が候補単語リストとして出力できる。 In such a parallel input method, processing including syllables formed by full input is useful for determining a target word from a list of candidate words when a matching candidate word is provided during full input. . Also, input values that are processed by abbreviated input, excluding already formed syllables (

The method of outputting a candidate word corresponding only to an input value that is interpreted as a simple code is useful for quickly inputting a frequently used suffix, particle, or the like with abbreviated input. For example,

(After confirming the input contents by a predetermined means)

To output "(under the assumption that it is stored in the index)

Enter. As mentioned above,

As described above, suffixes that are frequently used can be stored in a separate group (for example, suffix group) in the index, and when candidate words are output in parallel input, the words of the corresponding group can be output preferentially. . if

When entering

When

If both are remembered,

When

Can be output as a candidate word list.

の入力時、

まではフル入力として処理し、適切な音節を形成できない“３８”は短縮入力値として処理することができる。同様に索引に

のみがあれば、

が出力される。逆に索引に

のみがあれば、

のみが出力される。もし

の両者があれば、両者が一緒に出力できる。しかしながら、約束（regulations）によって１つ（フル入力として処理した部分を含む／含まない）のみを出力させることが好ましい。また、

の場合、もし

までフル入力として入力した後、

を連続的に入力すると、

となる。したがって、

までフル入力として入力し、

のみを短縮入力で入力するときは、

を入力し、入力された音節

を確定（例えば、音節の終了のために右矢印［＞］ボタンを１回押す）した後、

を入力する。ここで、

に対応する索引の他の単語

よりも低い優先順位を有するとしても、索引に

が記憶されているならば、システムは、入力値“５＊０２５＊０＞３８”に対して他の候補単語

よりも優先して

を出力することができる。 Another example is

When entering

Are processed as full input, and “38” which cannot form an appropriate syllable can be processed as a shortened input value. Similarly to the index

If there is only

Is output. Conversely to the index

If there is only

Is output only. if

If both are available, both can output together. However, it is preferable to output only one (including / not including the part processed as a full input) according to the regulations. Also,

If

After typing as full input,

Is input continuously,

It becomes. Therefore,

Enter as full input until

When entering only with abbreviated input,

And input the syllable

After confirming (for example, pressing the right arrow [>] button once to end the syllable)

Enter. here,

Other words in the index corresponding to

Even if it has a lower priority than

Is stored for the input value “5 * 025 * 0> 38”.

In preference to

Can be output.

の如く短縮入力として入力し、一部の音節をフル入力として入力する場合には、システムはそれを

と処理することができる。

まで入力されると、システムはシンプルコード（例えば、音節基準イニシャル値シンプルコード又は音節基準イニシャルコード）“１５７”を解析し、各音節が

に対応する単語を検索する。さらに“＊６”が入力されると、

となり、システムは

という音節の形成を認識することができる。すなわち、システムが既入力の“１５”が短縮入力値であることを認識し、さらに“２３”が入力されると、適切な音節が形成されないため、システムはそれを短縮入力値として処理できる。また、この事例においてもシステムは、

を一部音節ではフルコードと解析し、一部音節では短縮コードと解析して、それに対応する

をユーザに提供することができる。 In addition, some syllables

If you enter a short input as shown below and some syllables are input as full input, the system

And can be processed.

The system analyzes a simple chord (for example, a syllable reference initial value simple code or a syllable reference initial code) “157” and each syllable is

Search for the word corresponding to. If “* 6” is entered,

And the system is

Can be recognized. That is, when the system recognizes that the already input “15” is a shortened input value and further inputs “23”, an appropriate syllable is not formed, and the system can process it as a shortened input value. In this case, the system

Is analyzed as a full chord for some syllables and abbreviated chord for some syllables

Can be provided to the user.

  As mentioned above, one syllable of Chinese consists of “vowel + final”. When displaying a Chinese pin sound, the “vowel” is a consonant of a Roman vowel, and the “spoken vowel” is a vowel of a Roman vowel, or “vowel + n” or “vowel + ng.” That is, Chinese represented by a pin sound. One kanji character (one syllable) consists of “consonant + vowel”, or “consonant + vowel + n”, or “consonant + vowel + ng” of the roman vowel. In rare cases, the vowels “a”, “e”, Or, there can be a syllable that starts with “o.” Here, as mentioned above, there are only “n” or “ng” in the Romaji consonant that can be located at the end of one syllable of a pin sound. That is, only “n” or “ng” is possible as the final voice (last consonant of one syllable).

（ｚｈｏｎｇｈｕａ）”、“

（ｍｉｎｇｕｏ）”、“

（ｚｈｏｎｇｈｕａｍｉｎｇｕｏ）”、…などの単語が記憶されていると仮定すると、“

（ｚｈｏｎｇｈｕａ＝４９＊＊＊７７３９＃＃＊、図１０−６基準）”まではフル入力によってピン音を入力し、“

（ｍｉｎｇｕｏ）”のみは短縮入力（例示において短縮入力値として音節基準イニシャル値シンプルコードを利用した短縮入力）によって“ｍｇ＝７３（図１０−６基準）”を入力する。“

”の入力に当たり、該当ピン音（“ｚｈｏｎｇｈｕａ＝４９＊＊＊７７３９＃＃＊”）の入力後、リストに載っている候補漢字のうちいずれか１つを選択して確定し、その後“ｍｇ＝７３”を入力して良く、ピン音“ｚｈｏｎｇｈｕａ＝４９＊＊＊７７３９＃＃＊”入力後、連続的に“ｍｇ＝７３”を入力しても良い。前者の場合は、“７３＝ｍｇ”に対応する“

（ｍｉｎｇｕｏ）”、

（ｍｉｎｇｅ）、…などの単語のみを候補単語として出力することが当然である。また、後者の場合は、入力値“４９＊＊＊７７３９＃＃＊”を“

”と解析し、短縮入力値として認識された“７３”に対応する“

”と“

”を出力することが当然である。入力値“４９＊＊＊７７３９＃＃＊７３＝ｚｈｏｎｇｈｕａ＋ｍｇ”に対して“４９＊＊＊７７３９＃＃＊７”まで入力されるとき、システムは最後の“７”を正常なフル入力として入力されたピン音声母と認識するが、最後の“３”が入力される瞬間、システムは最後に入力された２つの入力値“７３”を短縮入力値と認識する。これは、前述した韓国語の場合と同一である。したがって、システムは入力値“４９＊＊＊７７３９＃＃＊７３”に対して、フル入力値“４９＊＊＊７７３９＃＃＊”に対応する“

”と短縮入力値“７３”に対応する“

”とからなる単語である“

”を出力することができる。図１３−４を参照する。 For example, “

(Zhonghua) ","

(Minguo) ","

(Zhonghuaminguo) Assuming that words such as “,... Are stored,“

(Zhonghua = 49 *** 77739 ##, reference to FIG. 10-6) “Pin sound is input by full input,

Only “(minguo)” inputs “mg = 73 (reference to FIG. 10-6)” by a shortening input (shortening input using a syllable reference initial value simple code as a shortened input value in the example).

”Is input, and after inputting the corresponding pin sound (“ zhonghua = 49 *** 77739 ## ”), one of the candidate kanji characters in the list is selected and confirmed, and then“ mg = 73 ”may be input, and“ mg = 73 ”may be continuously input after inputting the pin tone“ zhonghua = 49 *** 77739 ## ”. In the former case,“ 73 = mg ” Corresponding to “

(Minguo) ",

Of course, only words such as (minge),... Are output as candidate words. In the latter case, the input value “49 *** 77739 ##” is changed to “

"And corresponding to" 73 "recognized as a shortened input value"

"When"

It is natural to output “49 *** 77739 ## 73 = zhonghua + mg” up to “49 *** 77739 ## 7” for the input value “49 *** 77739 ## 73”. 7 ”is recognized as a pinned voice mother input as a normal full input, but at the moment when the last“ 3 ”is input, the system recognizes the last two input values“ 73 ”as shortened input values. This is the same as in the case of the Korean language described above, so the system inputs the full input value “49 *** 77739 ##” against the input value “49 *** 77739 ## 73”. Corresponding to “

"And" corresponding to the abbreviated input value "73"

"Is a word consisting of"

"Can be output. See FIG. 13-4.

”に対して短縮入力として“ｚｈ＝４９”を入力した後、候補単語のうち“

”を選択して入力せずに、連続的に“

”に対してフル入力として“７＃７７３＃＃＊＊＊”を入力するとき（すなわち、“４９７＃７７３＃＃＊＊＊＝ｚｈｍｉｎｇｕｏ”）でも、“ｚ＿ｈ＿ｍｉｎｇｕｏ”に対応する“

”を出力することができる。同様に“

（ｍｉｎｇｕｏ）”を入力するために“７３＃＃＊＊＊”を入力するとき、“７３＃＃＊＊＊＝ｍｇｕｏ＝ｍ＿ｇｕｏ”と解析して“

”を出力することができる。“７３”まで入力する瞬間、システムはそれを短縮入力と見なし、“ｍ＿ｇ＿”に対応する単語を検索して処理するが、次に来る“＃＃＊＊＊＝ｕｏ”を２番目の音節を形成する母音として処理する。“ｍｇ”の次の“ｕｏ”の入力は、“ｍ＿ｇ＿”に対応する各候補単語を“

”に限定させる役割をする。 In contrast, “

After inputting “zh = 49” as an abbreviated input, “

Without selecting and entering “

Even when “7 # 773 ## ***” is input as a full input to “” (that is, “497 # 773 #### = zmminguo”), “

"Can be output. Similarly,"

When “73 ## ***” is input to input “(minguo)”, “73 # *** = mguo = m_guo” is analyzed and “

At the moment of inputting up to “73”, the system regards it as a shortened input, searches for the word corresponding to “m_g_”, and processes it, but the next “## *** = “uo” is processed as a vowel that forms the second syllable.The input of “uo” next to “mg” is to select each candidate word corresponding to “m_g_” as “

It plays the role of restricting to “.

”を入力し、次いで“

”に対して短縮入力として“７３”を入力するとき、システムが単に入力値“７３”をシンプルコードと解釈して“ｍ__ｇ__”に対応する単語をユーザに提供するならば、多数の候補単語（例えば、“

（ｍｉｎｇｕｏ）”、

（ｍｉｎｇｅ）、…）のリストの形態に出力する。しかし、“

”が索引に記憶されていれば、“

”が索引に記憶されていることは別にして、“

＋ｍｇ＝

＋７３”を“

”と解析することが可能である。。この場合、“

”は既に漢字で入力されているので、システムは入力値“ｍｇ＝７３”に対応する“

”を他の候補単語（例えば、“

（ｍｉｎｇｅ）”、…）よりも優先的に出力することができる。 Kanji “

"And then"

When “73” is input as an abbreviated input to “”, if the system simply interprets the input value “73” as a simple code and provides the user with a word corresponding to “m__g__”, a large number of candidate words ( For example, “

(Minguo) ",

(Minge),... However,"

"Is stored in the index,

Aside from being stored in the index,

+ Mg =

+73 ”to“

Can be parsed as ". In this case,"

"" Has already been entered in Chinese characters, the system corresponds to the input value "mg = 73"

"To other candidate words (for example,"

(Minge) ", ...) can be preferentially output.

  Here, when “zhong...” Is input, the system can interpret it as “zho” + “n_g_” or “zhon” + “g_”. However, since there is actually no syllable consisting of “zho” or “zhon” in Chinese pin tones (ie, naturally not included in the pin tone index), the system looks up the index for “zhong…” Then, it is analyzed as “medium”. If included in the index, the system can output candidate words according to a predetermined output order when "zhong" is entered. According to FIG. 10-6, the system can also analyze the input of “zhongg... = 49 *** 7733” as “zhonk...” = “Zho” + “n_k_” or “zhon” + “k_”. As described above, this is similar to the parallel input method for simultaneously analyzing an input value into a full code and a shortened input value (simple code).

  As mentioned above, the example of using “syllable reference initial value simple code” as a shortened input value for convenience is presented. However, when “syllable reference initial code” is applied as a shortened input value, simple code analysis is performed. Therefore, the contents shown in FIGS. 11-3 to 11-10 can be applied.

19. Character input method using long press and control processing method using long press Generally, only one button is pressed once, and only one target (for example, an English letter in a language that does not use English as a native language, Symbols (special characters), functions, control of various uses, etc. can be comprehensively referred to as “alphabet”. Here, a specific button is pressed once, but “long press” can represent an object different from the object represented by simply pressing it once. Hereinafter, a long press of a specific button is referred to as “long press” or “long hit”. In the following, what is not represented as “long press” means an ordinary button press, and in particular, when compared with a long press, it is referred to as “short press”. In the following, we will explain the usual contents in English examples and explain application examples for each language as an example, but it goes without saying that the explanation contents in one language can be applied to other languages as well .

19.1 English (and common)
19.1.1 Entering Characters Using Long Press Currently, a method is used in which a corresponding number is entered by pressing a specific number button for a long time. However, it is not always possible to input only numbers. For example, pressing each number button once in FIG. 1-1 inputs the first letter among the letters assigned to the button (for example, [2] When it is assumed that “A” is input by pressing the number button once, the second letter (for example, “B”) can be input by long pressing. However, since long press is a factor that hinders the natural flow of input, it is not preferable to apply it to the input of frequently used characters. For this reason, it is preferable to apply the input by long press (long hit) to input an object that is not frequently used. Moreover, it is preferable that the user sets the time recognized as long press as much as possible according to the skill level of each user.

  In the present invention, long press is appropriately used, but this simplifies the input rule of a specific input method, widens the range of expression, and further supports that ambiguity can be removed. Here, for the sake of convenience, the long press of the [2] button is denoted as “2” by adding “˜” after “2”. As described above, when the existing iterative selection method is applied and the [2] button is pressed three times in FIG. 1-1, “222” is “C”, “AAA”, “AB” "Or" BA "is created.

  In FIG. 1-1, “A” is a long press of the [2] button or a single press of the corresponding button, “B” is a single press of the corresponding button after “A” is input, and “C” is a “B” input. An example of defining such that the corresponding button is pressed once (or the same as pressing twice after inputting “A”) is taken. When characters in the same button are continuously input, it is possible to input without ambiguity when starting with a long press in the input of the second and subsequent characters. For example, when “DACB” is input, it can be input without ambiguity as “322 to 222-2”, and “ACB” on the same button appears continuously, but “C” and “B” are input. Sometimes a long press is pinched so that it can be identified without ambiguity. This case can be input with almost the same feeling as the existing iterative selection method, and ambiguity is removed using long press only when the letters assigned to the same button are input continuously. There is an effect that it is possible. Of course, in the case of “DACB = 322 to 222 to 2”, it is also possible to use a long press of the [2] button (that is, “2 to”) to input “A”.

  The above example shows that any of the letters assigned to a specific button (including both explicit and implicit assignments) (for example, “A”) is the representative letter and the remaining letters follow. As the character to be used (for example, “B”, “C”), the subsequent character is input by pressing a button to which the representative character and the subsequent character are assigned a predetermined number of times. That is, a relationship of “A (representative character) -B (second succeeding character) -C (third succeeding character)” is set, “A = 2 to”, and “B = A + 2 =“ 2 ”. ˜2 ”,“ C = B + 2 = A + 2 + 2 = 2-22 ”In addition, it is possible to input“ A ”not by long pressing but by normal pressing (that is, short hitting) to the same button. If the entered characters are not input continuously, it is possible to input the representative characters by pressing once (that is, short strokes) as in the existing iterative selection method. Hereinafter, this is referred to as a “repetitive selection method by long-pressing a representative character” for convenience.

  If you enter only two letters using one button, define one letter as a normal one press (ie, short stroke) and the other as a long press (ie, long stroke) They can be defined and input (for example, “B = 1”, “P = 1˜” in FIG. 10A). However, when three letters are assigned per button (explicit assignment or implicit assignment) as shown in FIG. 1-1 (for example, [2] button), the first letter is pressed once. Define (for example, “A = 2”), define the second letter as long press (for example, “B = 2”), and press the third letter twice (for example, “C = 22”) Alternatively, if it is defined as long press and single press (for example, “C = 2 to 2”), ambiguity exists even when the third character is input. This is because it can be interpreted as “AA” when defined as “C = 22”, and can be interpreted as “BA” when defined as “C = 2 to 2”. In this case, the first, second and third letters are arbitrarily defined. Again, the number of times that ambiguity can occur is much less than with the usual simple iterative selection method.

  The merit of the iterative selection method using long press is that it can be input without ambiguity. In addition, it is more effective when inputting not only the letter assigned to each button but also numerals and symbols related to each button. For example, a specific letter “A1, A2, A3,...” Is associated with any “button x” (explicit assignment or implied assignment), and a specific number “N1” is associated, Assume that specific symbols “S1, S2, S3,...” Are related. Here, it can be defined that an arbitrary letter related to “button x” is input by long press. For example, it is defined that the character “A1” is defined as a representative character and “button x” is input by long pressing. If the input of “A1, A2, A3, N1, S1, S2, S3” related to “button x” is represented by a graph, it is as shown in FIG. From FIG. 14C, it is understood that “A1S1 = xx to xxx” or “A1S1 = x to x to xxx” is input. Here, in a language in which English is not a native language, English letters “E1, E2, E3,...” May be input in association with each other.

  Furthermore, when inputting an object related to “button x” (for example, “A1, A2, A3, N1, S1, S2, S3”), it is not an iterative selection method by pressing a representative character long, but an ordinary iterative selection. It is also possible to input related numbers and various symbols (special characters) by applying the method. However, since ambiguity arises when objects related to the same button are input continuously, after inputting one object, "delay of a predetermined time" or a numerator (for example, "complementary items in the parallel input method") It is necessary to suppress the occurrence of ambiguity by inputting the [>] button once pressed) described in the column. For example, when “A1, A2, A3, N1, S1, S2, S3” are related to “button x”, and “S1A1” is input by an iterative selection method, “S1A1 = xxxxxxxx> x” is input. Can be. This is a technique used to overcome the ambiguity when applying the iterative selection method, but there is a difference in that it is possible to input a specific button in association with a specific number and symbol group.

  What has been developed so far is that if there is a native letter “A1, A2, A3” related to (assigned) “button x”, “x = A”, “xx = B”, “xxx = If it is “C” or the button is pressed beyond the predetermined maximum number of repeated pressings (for example, “button x” is pressed four times), “xxxx = A”, “xxxx = B”, etc. It was to the extent that it was cyclically toggled. The circular toggle is considered as a consideration that allows the target character to be input further by repeatedly pressing the pressed button when an input error occurs when the repeated selection method is applied. However, according to the applicant's proposal, if the button input is not deleted of the entire generated character, but instead of deleting the “recent input value”, it is not necessary to adopt a cyclic toggle method. good. Only “xx = B” must be pressed, but if “xxx = C” is entered by mistake, “XX” is further pressed and the delete button is pressed once without entering “B”. This is because "" can be input.

  FIG. 14-4 shows a process of inputting a character by a long press after pressing a specific button. 14-4, lines A1, A2, A3,... Are input processes by the existing iterative selection method. It can be understood that B2 is input by combining a long press of the same button after A1 is input by pressing a certain button (for example, button “x”) shortly. That is, “A1 = x”, “B2 = xx˜”. Here, if no input of any character is defined only by long pressing the button x, and if only the characters A1 and B2 are input using the button x, it can be recognized without ambiguity. Even if the long press of the button x is defined as any character (for example, the character E1), if there is no case where “A1” and “E1” appear at the same time, such a language restriction (A1 and E2 are A1, B2, and E1 can be input without ambiguity using a language restriction that does not appear.

  Similarly, in FIG. 14-4, when the long press of “button x” is not defined as any character input, A1, B2, and C3 can be recognized without ambiguity. Furthermore, in the case of A1, A2, and D3, “A1 = x” and “A2 = xx = A1A1” may be ambiguous, but if A1 does not come out continuously in a certain language. For example, A1, A2, and D3 can be input without ambiguity using such language restrictions (language restrictions that A1 does not appear continuously). The method in FIG. 14-4 is referred to as a “repetitive selection method by long press after short press” for convenience. For example, in FIG. 10-6, it is determined that “i = #” and “u = # ˜” are input, and any vowel (for example, “..u”-“u” is superscripted by “## ˜”). (Characters with “...”) can be input without ambiguity (“i” and “u” do not appear consecutively in Chinese pin tones). Of course, since “i” does not come out continuously, another target may be input by “##”. This can be adopted in other languages as well. In FIG. 14-4, it can be used to input the last letter of the A1, A2, A3 lines (for example, S1, S2, S3,...).

19.1.2 Control processing method using long press Next, an example of application to control selection will be shown. In the following application example, it is assumed that the control is input later. In the above example, both the short stroke and long stroke of the [2] button are set to represent “A” simply to give an example in which input can be made with almost the same feeling as the existing iterative selection method. However, a specific object can be represented by a single press of a specific button, and another object can be represented by a long press of the same button. For example, a single press of a particular button (eg, [1] button) may cause a particular object (eg, any “character x” —where “x” is not an actual “x”, but any character Meaning long) and long presses can represent other specific objects (eg, any “letter y”). That is, “character letter x = 1”, “character letter y = 1”. Here, since the “character x” and “character y” given as examples are conceptual, “A = 2 to 2” or “A =” in the “repetitive selection method by long press of representative characters” described above. Even if it is defined as “2”, “A” by ordinary pressing (that is, short stroke) corresponds to the letter “x”, and “A” by long pressing corresponds to “character y”.

  This is equally applicable when selecting a control. For example, a single press of an arbitrary button (eg, [*] button as a control button) can represent a specific object (eg, an arbitrary “control a1”), and a long press of the same button can indicate another specific (For example, an arbitrary “control b1”). “A2 control”, “a3 control”,... Can be selected by repeatedly pressing the control button. What is important here is that after selecting “b1 control” by long-pressing the control button, “b2 control”, “b3 control”, “b4 control”,. It can be done. Similarly, no ambiguity arises. In this example, it is possible to refer to an example in which only the representative character “A” is input by long pressing and short strokes are combined when selecting the remaining subsequent characters.

  For convenience, a target input by a combination of a specific button (for example, [1] button) and “b1 control” is expressed as “B1” as follows.

“A1 = x + {a1 control} = 1 *”, “A2 = x + {a2 control} = 1 **”, “A3 = x + {a3 control} = 1 ***”,...
“B1 = x + {b1 control} = 1 * ˜”, “B2 = x + {b2 control} = 1 * ˜ *”, “B3 = x + {b3 control} = 1 * ˜ **”,.

  Of course, it is also possible to apply one long stroke, two long strokes, and three long strokes to select the b1, b2, and b3 controls. However, as a matter of course, it is not preferable to use long press frequently. As in the case of B1, B2, B3,..., Input by ordinary pressing after one long pressing of the control button is referred to as “repetitive pressing after long pressing” for convenience, and such a control processing method is referred to as “long "Control processing method by repeated pressing after pressing". The A1, A2,... System means the control processing method already presented, and the B1, B2,... System means “control processing method by repeated pressing after long pressing”. The above contents are shown in a graph in FIG. The above is an example in which one long press is applied only to the first control selection as in the repeated selection method by long press of the representative character. In addition, various further expansions (for example, C system and D system) are possible by inserting a long press as shown in FIG. 14-2.

  Here, it goes without saying that other objects can be expressed by applying a long press of a button ([1] button in the example) instead of the control button (for example, “character y”). Furthermore, the alphabet (FIG. 2) displayed on each button of the keypad by the repeated selection method using long press (for example, “A = 2 to”, “B = 2 to 2”, “C = 2 to 22”). (3 characters in 1-1) can be identified without ambiguity, so that different groups are selected by repeatedly pressing the control button after selecting three characters that can be selected by operating the numeric buttons, or repeatedly pressing after a long press. Can be entered. That is, in FIG. 1-1, the character of a specific group (for example, group 1) (meaning including symbols, numbers, native language, English character, etc.) is input by repeatedly pressing the control button after inputting "A". Alternatively, the character of another specific group (for example, group 2) may be input by “repetitive pressing after long pressing”. Similarly, the group 3 character can be input by a combination of “B” input and repeated pressing of the control button, and the group 4 character of another group can be input by repeated pressing after a long press. In addition to the three letters shown on the keypad buttons of FIG. 1-1, an arbitrary rule (for example, “2 to 222”) is defined, and the control button is repeatedly pressed. You may make it input the character of an arbitrary group by the repeated press after a long press.

  The control processing method by repeated pressing after long pressing is controlled as shown in Korean examples (Figs. 4-5 to 4-8) and Japanese examples (Figs. 2-1 and 2-2). When a button is necessary for inputting a basic native language, it is extremely useful for inputting numbers used intermittently in a sentence and inputting English alphabets (in a language whose native language is not English). Hereinafter, an application example for each language will be briefly described, but the present invention can be applied to languages not mentioned here.

または

）する場合にも同様に適用可能である。すなわち、同じボタンに属する字母が連続して出てくる場合、２番目以降に出てくる字母の入力に際し、長押しによって入力するのである。例えば、図４及び図５において、［１］ボタンの連続押しを

として定義すれば、

が連続して出てくる場合、

ではなく、

として認識されることがあるが、このときにも、２番目の

の入力時に長押しにより入力すれば、

として認識できる。すなわち、

において２番目の音節の

の入力時に長押しを用いるのである。

の入力時にも２番目の音節の

の入力に長押しを用いて

として入力するのである。これは、変形字母をコントロール処理方法によって入力することと選択的に（２通りのうち１つのみ、又は、両方とも）適用可能である。 19.2 Korean In the examples of FIGS. 4-5 to 4-8, a modified alphabet (for example, a grit consonant / hard consonant and a drop consonant) is input by repeatedly pressing a button to which a basic consonant belongs (for example,

Or

), The same applies. That is, when characters belonging to the same button appear continuously, the characters that appear after the second are entered by long pressing. For example, in FIG. 4 and FIG.

If defined as

Is coming out continuously,

not,

It may be recognized as, but at this time also the second

If you enter a long press to enter

Can be recognized as. That is,

In the second syllable

A long press is used when inputting.

Of the second syllable

Use long press to enter

As input. This can be applied selectively (only one of the two or both) by inputting the modified letter by the control processing method.

の場合、

となる曖昧さがある。 As an example in which a part of the characters is input by long press instead of the representative character, one of the consonant sounds or the hard consonants can be input by long press. For example, if it is defined as inputting a consonant sound by long press, a hard consonant (a twin sound) can be input by a combination of basic consonants. Of course, as disclosed in the earlier application,

in the case of,

There is ambiguity to become.

）の入力に長押しを応用することができる。すなわち、

のようになる。出願人の私見によって、先出願に既に提案されているように、図４−５を基準として

、

又は

として入力することが、長押しを用いた方よりも入力効率面で良好であると思われる。既に提案されている脱落子音（例えば、

）の入力方法及び長押しによる入力方法が選択的に適用可能である。 Similarly, consonants assigned with vowels (for example, assigned to the [0] button in FIGS. 4-5 and 4-6)

) Can be applied to long press. That is,

become that way. Based on the personal opinion of the applicant, as already proposed in the earlier application,

,

Or

It seems that the input efficiency is better in terms of input efficiency than the long press. Already proposed drop consonants (for example,

) And an input method by long press can be selectively applied.

  Next, an application example of “control processing method by repeated pressing after long pressing” will be shown. A representative Korean language input system by the applicant can be found on the homepage http: // www. simplecode. It has been introduced to the net together with simulators (introducing old words, symbol input and parallel input technology), and it is disclosed that numbers and English letters can also be input without switching modes.

  There is a request from a person in the industry to provide input technology for numbers and English without switching modes. The company that made such a request did not adopt the applicant's Korean input technology).

を表すが、［＊］ボタンの１回の長押しは任意の対象（例示においては、“数字コントロール”）を表すことができるからである。 The normal frequency of use in the input is in the order of “local language letter-various symbols (special characters) -numbers-English letter”. As shown in FIG. 4-5 and the like, there is no control button used for inputting numbers and English. Here, it is possible to input a number by combining a number button and a long press of the [*] button (that is, “* ˜ = number control”). That is, “number 1 = 1 * ˜”. The [*] button is used primarily as a vowel button and also as a control button. Pressing the [*] button once is a vowel

This is because one long press of the [*] button can represent an arbitrary object (in the example, “numerical control”).

として認識されず、“Ｂ”として認識されるのは、韓国語において母音

で始まる音節がない性質（一種の韓国語制限）を用いるからである。もし、［＃］ボタンの長押しおよび反復押しを他の用途として用いるならば、英語字母を数字の後続する字母としてみなして“Ａ＝２＊〜＊”又は“Ａ＝２＊〜＃（クロスコントロール処理方法）”として、“Ｂ＝２＊〜＊＊”又は“Ｂ＝２＊〜＃＃”などのように各種に定義できる。 The input of English letters can be done by various methods. For example, “A = 2 # ˜”, “B = 2 # ˜ #”, and “C = 2 # ˜ ##” can be defined. “A” and vowels for “2 # to #” input

Is not recognized as "B" but is recognized as "B" in Korean

This is because it uses the property that there is no syllable starting with (a kind of Korean restriction). If long press and repeat press of the [#] button are used for other purposes, the English letter is regarded as the letter following the number, and "A = 2 *-*" or "A = 2 *-# (cross Control processing method) ”can be defined in various ways such as“ B = 2 * to ** ”or“ B = 2 * to ## ”.

が割り当てられたボタンの反復押しによって

を入力する

と定義すれば、字母の入力に最大３打ち（すなわち、最大可能な反復回数）の［＊］ボタンの連続押しが必要となる。この場合、激音コントロールは“最大可能な反復回数＋１”回押し（すなわち、［＊］ボタン４回）によって選択できる。“すなわち、

である。この場合、連続して４回押す負担を軽減するために、“激音コントロール＝＊〜”として

として定義することができる。これは、硬子音および変形字母として見なす脱落子音に対しても同様に適用可能である。また、コントロール処理方法において、激音コントロールを反復選択方法によって選択するあらゆる場合にも適用することができ、長押しによる変形字母の入力及び既存の反復押しによる変形字母の入力が選択的（両者のうち一方のみ、又は両方とも）に行えることは言うまでもない。この場合、数字の入力は他のボタン（例えば、［＃］ボタン）を用いれば良い。 An example of the control processing method by long press is further proposed. In FIGS. 4-5 to 4-8, the vowels

By repeatedly pressing the assigned button

Enter

Is defined, it is necessary to continuously press the [*] button of up to three strokes (that is, the maximum possible number of repetitions) to input the character. In this case, the intense sound control can be selected by pressing “the maximum possible number of repetitions + 1” times (that is, four times of the [*] button). “That is,

It is. In this case, in order to reduce the burden of pressing 4 times in succession, "Geki sound control = * ~"

Can be defined as This is equally applicable to hard consonants and dropped consonants that are considered as deformed bases. Also, in the control processing method, it can be applied to any case in which the intense sound control is selected by the repetitive selection method, and the input of the modified letter by long press and the input of the modified letter by the existing repeated press are selective (both It goes without saying that only one or both of them can be done. In this case, other buttons (for example, [#] button) may be used to input numbers.

19.3 Japanese As in the Japanese examples shown in FIGS. 2A and 2B, there are a large number of subsequent characters, but the subsequent characters are effective at the time of input by the iterative selection method. That is, “A = 1” or “A = 1” is defined as “I = 1-1”, “U = 1-11”, “E = 1-111”, “O = 1-1111”. It is. As described above, when characters assigned to the same button come out continuously, only the second character out of the characters assigned to the same button is pressed and held (ie, “1” to “1”). If it is not, ambiguity does not occur even if it is input by simply pressing once. However, in the Japanese cases shown in FIGS. 2-1 and 2-2, it is possible to input without ambiguity with a small number of input strokes by applying the subsequent control processing method. This method can be applied together with an iterative selection method by long-pressing a representative character.

  In the case of inputting the subsequent character by the control processing method without applying the repeated selection method by the long press of the representative character, the deformed character is input by using the long press. In FIG. 2-2, when a subsequent character is input by the control processing method, a modified character (for example, “ba”, “pa”) of a representative character (for example, “ha”) is input as shown in FIG. It can be seen that the number of input strokes is somewhat large and unnatural. In this case, the long press of the corresponding button ([6] button in FIG. 2-2) can be defined as a modified character (for example, “BA”) of the representative character. The remaining deformed characters can be defined as “Pa = ba + predetermined control button (for example, [*] button) = 6 to *”.

  Here, if a deformed character is not input using a long press of a button to which a character is assigned, iterative selection method by long press of a representative character and input of subsequent characters by a control processing method can be applied simultaneously. The input of deformed characters (for example, “BA”, “PA”) can be set to input “PA” by combining the long press of any control button (for example, [*] button) and the representative character “HA”. . That is, “B (arbitrary deformed character) == + * ˜ = 6 * ˜” and “Pa (any other deformed character) = can be defined as + * ˜ * = 6 * ˜ *”. . Even if there are a large number of deformed letters of “ha”, there is no ambiguity using repeated pressing after long pressing like “6 * to **”, “6 * to ***”, etc. Needless to say, it can be set to input. This is because “**” and “* to *” can represent different objects (in this case, arbitrary controls), as “*” and “* to” represent different objects.

  What is important here is that not only can any control be selected by a long press, but also other objects (eg, a control button) without any ambiguity by repeatedly pressing a button (for example, a control button) used for a long press after a long press. Any control) can be represented.

  Similarly, what is explained in the Korean column can be applied to input of numbers and English letters. For example, it can be defined as “number 2 = 2 # ˜”, and can be defined as “A = 2 # ˜ #”, “B = 2 # ˜ ##”,.

19.4 Chinese It has been described that a Chinese pin sound can be input without ambiguity (or almost without ambiguity) on keypads such as FIGS. 10-1 to 10-6. However, in order to abbreviate a word consisting of “b_b_” (underlined pin syllables consisting of vowels), when “bb = 11” is entered, “bb” does not appear continuously in Chinese pin tones. If the language restriction is applied, it can be recognized as “11 = p”. In this case, as described above, after “b = 1” is input, the input value “1” is intentionally determined as “b” by a predetermined time delay or a specific operation (for example, pressing the [>] button). , "B = 1" must be entered. Similarly, it can be seen as a problem of ambiguity that occurs when characters assigned to the same button are continuously input. Also in this case, it is possible to solve the problem by applying a control processing method by long pressing of the representative character (the representative character “b” of the [1] button is changed to “b = 1 to”). For example, when characters “bb = 1 to 1” or characters assigned to the same button appear successively, long press can be applied to the input of the second and subsequent characters to “bb = 11 to 11”. It is.

  Next, inputting numbers and symbols will be described with reference to FIG. In the case of English input in Chinese, it is not necessary to refer to it because it is not necessary to convert it to Kanji after inputting the English letter. Based on FIG. 10-6, it can be defined as “number 1 = 1 * ˜”. In FIG. 10-6, a single press of the [*] button is recognized as “a”, but a long press (“* ˜”) can represent another object (in the example, “numerical control”). It is. In FIG. 10-6, it is preferable to apply a control processing method using long press and repeated press of the [#] button to input various symbols. For example, a dot (.), A comma (,), a colon (:), a semicolon (;), etc. are connected to the [2] button, and “point (.) = 2 # ˜”, “comma (,) = 2 # ˜ “#”, “Colon (:) = 2 # to ##”, “semicolon (;) = 2 # to ##”. The reason why “2 # to #” is not recognized as “dot (.) + I” but can be recognized as “comma (,)” is because a Chinese pin tone does not start with “i”. (See the case study in Korean). Similarly, “2 # to ##” is not recognized as “point (.) + I + i” or “comma (,) + i”, but can be recognized as “colon (:)”. If it is desired to input “point (.) + I”, as described above, after the input of “2 # ˜”, after a predetermined time has elapsed, after the determination of “point (.)”, Or a specific operation (for example, It is only necessary to input “# = i” after the “point (.)” Is confirmed by pressing the “>” button.

  In FIG. 10-6, it is rare that the [*] button is not used for inputting a symbol of a specific group, but since there are pin sounds beginning with “a”, “e”, “o”, etc., “ This is because an ambiguity may occur when “repetitive press after long press” is used. Since numbers exist one by one, long presses of the [*] button need only be combined once. In the case of FIGS. 10-1 to 10-5, since there is no pin sound beginning with “i” and “u” assigned to the [*] button and the [#] button, the [*] button and the [#] button are It is always possible to input numbers, symbols, etc. by a control processing method using repeated pressing after using and pressing for a long time.

  Needless to say, the above contents can be applied not only to the keypads of FIGS. 10-1 to 10-6 but also to other keypads.

19.5 Languages that use Roman letters In European languages that use Roman letters, there are many variants of the basic letters with a superscript (or subscript). As described above, various long presses can be applied to the input of modified characters and the input of special characters.

  The present invention has shown that, following the applicant's previous application, a full input method is applied and a target word is efficiently input using an index. Furthermore, an example of applying the (Language restriction) parallel input method effectively was shown.

  In addition, the present invention inputs various characters by a combination of a long press of a specific button and a repetitive press of the specific button, and a combination of a long press of a specific button and a repetitive press of a control button. An input system was presented.

FIG. 1-1 is a diagram showing an English standard keypad. FIG. 1-2 is a diagram illustrating an example in which a subsequent keypad is configured with a central character as a representative character and a left and right character as a subsequent character when assigning approximately three characters to each button. FIG. 2-1 is a diagram illustrating an example (1) in which a Japanese subsequent keypad is configured. FIG. 2-2 is a diagram illustrating an example (2) of a Japanese subsequent keypad. FIG. 2-3 is a graph of the two-dimensional reuse control processing method. FIG. 2-4 is a graph of the two-dimensional cross control processing method. None FIG. 4A is a diagram illustrating an example 1 (Korean) in which a basic consonant and a basic vowel are paired and assigned to each button. FIG. 4-2 is a diagram of an example 2 (Korean) in which a basic consonant and a basic vowel are paired and assigned to each button. FIG. 4-3 is a diagram of an example 3 (Korean) in which a basic consonant and a basic vowel are paired and assigned to each button. FIG. 4-4 is a flowchart (an example of Korean) when a basic consonant and a basic vowel are input by an iterative selection method. FIG. 4-5 is a diagram illustrating an example in which vowel elements are used in Korean. FIGS. 4-6 is a figure which shows the example which utilized the vowel element in Korean. FIG. 4-7 is a diagram illustrating an example in which vowel elements are used in Korean. FIG. 4-8 is a diagram illustrating an example in which vowel elements are used in Korean. FIG. 5A is a diagram illustrating an example of Chinese character input using a pin sound dictionary and pin sounds. FIG. 5B is a diagram illustrating an example of English text input using the entire association simple code. FIG. 5C is a diagram illustrating a case (1) in which the input value is processed as a promised simple code in the parallel input method. FIG. 5-4 is a diagram illustrating a case (2) in which the input value is processed as a promised simple code in the parallel input method. FIG. 5-5 is a diagram illustrating an application example of the parallel input method via the client and the server. FIGS. 5-6 is a figure which shows the grouping example of the phrase (word or phrase) memorize | stored in the index. FIG. 6A shows a case where the server (relay server) interprets the simple code transferred from the client, and transfers the information obtained by interpreting the simple code to the third server that needs the information. FIG. FIG. 7A is a diagram illustrating an example (English keypad) of a subsequent keypad that can input various symbols based on the control processing method. FIG. 7-2 is a diagram illustrating an example (Korean keypad) of a subsequent keypad that can input various symbols based on the control processing method. None None FIG. 10A is a diagram illustrating an example of a consonant separation keypad in a language using a Roman alphabet. FIG. 10-2 is a diagram illustrating an example of a consonant separation keypad in a language using Roman alphabet. FIG. 10C is a diagram illustrating an example of a consonant separation keypad in a language using a Roman alphabet. FIG. 10-4 is a diagram illustrating an example of a consonant separation keypad in a language using Roman alphabet. FIG. 10-5 is a diagram illustrating a short input method example in the consonant separation keypad and a case in which the input value is processed as a promised simple code in the parallel input method. FIG. 10-6 is a diagram illustrating an example of a consonant separation keypad in a language using Roman alphabet. FIG. 10-7 is a flowchart (1) in which the language restriction input method is applied to the consonant separation keypad. FIG. 10-8 is a flowchart (2) in which the language restriction input method is applied to the consonant separation keypad. FIG. 10-9 is a flowchart in which the language restriction input method is applied to an incomplete consonant separation keypad. FIG. 10-10 is a diagram showing an application example (1) of the parallel input method (in a system including all words). FIG. 10-11 is a diagram illustrating an application example (2) of the parallel input method (in a system including all words). FIG. 10-12 is an application flowchart (1) of the parallel input method (in a system including all words). FIG. 10-13 is an application flowchart (2) of the language-limited parallel input method (in a system including all words). FIGS. 10-14 is a figure which shows the application example of the full input method (and parallel input method) in the system which memorize | stored the full code. FIGS. 10-15 is a figure which shows the application example (1) of the parallel input method in the system provided with the index of the words (word or phrase) which consist only of a representative character. FIGS. 10-16 is a figure which shows the application example (2) of the parallel input method in the system provided with the index of the words (word or phrase) which consist only of a representative character. FIGS. 10-17 is a figure which shows the example which processes a tongue-sound sound input as a simple code | cord | chord (shortened input value) primarily. FIG. 11A is a diagram illustrating an example of a system configuration diagram when an input value is interpreted on the client side. FIG. 11B is a diagram illustrating an example of a system configuration diagram when an input value is interpreted on the server side. FIG. 11C is a diagram illustrating an example of processing a simple code (in a parallel input method). FIG. 11-4 is a diagram illustrating an example of processing a simple code (in a parallel input method). FIG. 11-5 is a diagram illustrating an example of processing a simple code (in a parallel input method). FIG. 11-6 is a diagram illustrating an example of processing a simple code (in a parallel input method). FIG. 11-7 is a diagram illustrating an example of processing a simple code (in a parallel input method). FIG. 11-8 is a diagram illustrating an example of processing a simple code (in a parallel input method). FIG. 11-9 is a diagram illustrating an example of processing a simple code (in a parallel input method). FIG. 11-10 is a diagram illustrating an example of processing a simple code (in a parallel input method). FIG. 11-11 is a diagram illustrating an example of a type of simple code (that is, a shortened code or a shortened input value). None FIG. 13A is a diagram illustrating an example in which a part of the input value is processed as a full code and the other part is processed as a simple code. FIG. 13B is a diagram illustrating an example in which a part of the input value is processed as a full code and the other part is processed as a simple code. FIG. 13C is a diagram illustrating an example in which a part of the input value is processed as a full code and the other part is processed as a simple code. FIG. 13-4 is a diagram illustrating an example in which a part of the input value is processed as a full code and the other part is processed as a simple code. FIG. 14A is a diagram illustrating an example of a control processing method using long press. FIG. 14B is a diagram illustrating an example of a control processing method using long press. FIG. 14C is a diagram illustrating an example in which a short press is used after a long press. FIG. 14-4 is a diagram illustrating an example in which long press is used after short press.

Claims (9)

In a keypad that has a large number of buttons and a number of letters assigned to the buttons,
(R-1) By pressing once a button to which one or more characters are assigned among the buttons, the first assigned character of the pressed button is recognized,
(R-2) By pressing twice a button to which two or more characters are assigned among the buttons, two characters recognized from (r-1) are recognized or pressed The second assigned letter of the assigned button is recognized,
(R-3) By pressing a button to which three or more characters are assigned among the buttons three times in succession, three characters recognized from (r-1) are recognized, or ( the character recognized from r-1) and the character to which the button is assigned second are recognized, or the character assigned to the third of the pressed button is recognized;
(R−n) By pressing a button to which n or more letters are assigned among the buttons, the letters assigned by the logics (r-1) to (r-3) are In a method of inputting a vowel (including a consonant, a vowel, a word or a phrase) by a known repetitive selection method (Repeat Selection Method),
The system index is provided, on the index, the phrase (word or Kubushi) button value of the iterative selection method corresponding to the phrase associated with the pair,
(A) recognizing an input value of the button in response to pressing of the button;
(B) comparing the button value of the iterative selection method stored in the index of the system with the input value in step (a) when the same button is pressed twice or more times ;
(C) recognizing a word / phrase in which the input value in step (a) matches the button value of the iterative selection method stored in the index as a target word;
The character input method with the keypad characterized by including. The system index stores a simple code that is a button value corresponding to all or part of a word (word or phrase) and a letter constituting the word,
(D-1) comparing the input value with the simple code of the index at the moment when it is confirmed that there are no more button values of the iterative selection method that match the input value;
(D-2) a step of recognizing a matching word / phrase as a target word / phrase as a result of the comparison in step (d-1);
The character input method with the keypad according to claim 1, wherein The system index stores a simple code that is a button value corresponding to all or part of a word (word or phrase) and a letter constituting the word,
(E-1) The button value of the iterative selection method stored in the index among the input values at the moment when it is confirmed that there are no more button values of the iterative selection method that match all the input values; Comparing the input value remaining except the matching part with the simple code;
(E-2) a step of recognizing a matching word / phrase as a target word / phrase as a comparison result of the step (e-1);
The character input method with the keypad according to claim 1, wherein   The method of claim 1, wherein the keypad is a minimum numeric keypad including 1 to 9 and 0 buttons. The method of claim 4, wherein the keypad is an international standard English keypad.   5. The character input method using a keypad according to claim 4, wherein the numeric keypad is a keypad of a telephone or a mobile terminal. The character input method using a keypad according to claim 1, wherein the keypad is a touch screen keypad displayed on a screen .   The method according to claim 1, wherein the index is stored in a machine-readable medium, and the input method is performed by a processor executing a program from the medium. In a keypad that has a large number of buttons, and the buttons are assigned a large number of letters,
(R-1) By pressing once a button to which one or more characters are assigned among the buttons, the first assigned character of the pressed button is recognized,
(R-2) By pressing twice a button to which two or more characters are assigned among the buttons, two characters recognized from (r-1) are recognized or pressed The second assigned letter of the assigned button is recognized,
(R-3) By pressing a button to which three or more characters are assigned among the buttons three times in succession, three characters recognized from (r-1) are recognized, or ( the character recognized from r-1) and the character to which the button is assigned second are recognized, or the character assigned to the third of the pressed button is recognized;
(R−n) By pressing a button to which n or more letters are assigned among the buttons, the letters assigned by the logics (r-1) to (r-3) are In a method of entering a vowel (including consonants, vowels, words or phrases) by a known iterative selection method recognized :
The system is provided with an index, in which the phrase (word or phrase) and the button value of the iterative selection method corresponding to the phrase are stored in association with each other ,
In the index, the phrase and a simple code that is a button value corresponding to all or part of the letters constituting the phrase are stored in association with each other ,
(A) recognizing an input value of the button in response to pressing of the button;
(B-1) comparing the input value in the step (a) with the button value of the iterative selection method stored in the index of the system;
(C-1) recognizing as a target phrase a phrase that matches the input value and the button value of the iterative selection method stored in the index;
(B-2) comparing the input value in the step (a) with the simple code stored in the index;
(C-2) the step of recognizing the matching word / phrase as a target word / phrase as a comparison result of the step (b-2);
Including
A character input method using a keypad, wherein the steps (b-2) and (c-2) are performed simultaneously with the steps (b-1) and (c-1).

Images