CN103064531B - Terminal and input method - Google Patents

Terminal and input method Download PDF

Info

Publication number
CN103064531B
CN103064531B CN201310026097.9A CN201310026097A CN103064531B CN 103064531 B CN103064531 B CN 103064531B CN 201310026097 A CN201310026097 A CN 201310026097A CN 103064531 B CN103064531 B CN 103064531B
Authority
CN
China
Prior art keywords
character
input
reverse recommendation
reverse
recommendation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201310026097.9A
Other languages
Chinese (zh)
Other versions
CN103064531A (en
Inventor
梁维伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yulong Computer Telecommunication Scientific Shenzhen Co Ltd
Dongguan Yulong Telecommunication Technology Co Ltd
Original Assignee
Yulong Computer Telecommunication Scientific Shenzhen Co Ltd
Dongguan Yulong Telecommunication Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Yulong Computer Telecommunication Scientific Shenzhen Co Ltd, Dongguan Yulong Telecommunication Technology Co Ltd filed Critical Yulong Computer Telecommunication Scientific Shenzhen Co Ltd
Priority to CN201310026097.9A priority Critical patent/CN103064531B/en
Publication of CN103064531A publication Critical patent/CN103064531A/en
Application granted granted Critical
Publication of CN103064531B publication Critical patent/CN103064531B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Document Processing Apparatus (AREA)
  • Machine Translation (AREA)

Abstract

The invention provides a kind of terminal and a kind of input method, wherein, terminal comprises: character input unit, for receiving the character of input; Retrieval unit, for retrieving the reverse recommendation character be associated with the character of described input in basic dictionary according to the character of described input; Reverse character recommendation unit, for exporting described reverse recommendation character, wherein, the character of described reverse recommendation character for inputting before being corrected in the character of described input.By technical scheme of the present invention, error correction can be carried out by follow-up input character to the character inputted, need not perform when making user input Wrongly-written or mispronounced character and delete and the troublesome operation such as to re-enter, can further improve the intelligent of input method.

Description

Terminal and input method
Technical field
The present invention relates to communication technical field, in particular to a kind of terminal and a kind of input method
Background technology
In existing input method, association is the main method of input method of Chinese character, and existing association method is commonly forward association method, has both supposed that user inputs character is correct, has carried out the deduction of character below on this basis.When by this input method, if user inputs character has deviation, then need user to delete this character, and re-enter correct characters.On mobile terminals during input character, if there is above-mentioned situation, then need to carry out backspace and delete processing, operation steps is comparatively loaded down with trivial details, both have impact on input efficiency, have impact on input accuracy again.
Therefore need a kind of new input method, error correction can be carried out by follow-up input character to the character inputted, need not perform when making user input Wrongly-written or mispronounced character and delete and the troublesome operation such as to re-enter, can further improve the intelligent of input method.
Summary of the invention
The present invention, just based on the problems referred to above, proposes a kind of new character input scheme, retroactive association mode can be adopted to be that the character inputted carries out error correction, avoid the troublesome operation steps such as deletion.
In view of this, the present invention proposes a kind of terminal, comprising: character input unit, for receiving the character of input; Retrieval unit, for retrieving the reverse recommendation character be associated with the character of described input in basic dictionary according to the character of described input; Reverse character recommendation unit, for exporting described reverse recommendation character, wherein, the character of described reverse recommendation character for inputting before being corrected in the character of described input.
In the basic dictionary of this programme, the all corresponding one or more reverse recommendation character of each character, therefore, whether the character no matter inputted is correct, all can go out the reverse recommendation character corresponding with the character of current input according to the character machining of current input and be shown to user, during the character errors inputted before the character of current input, user can select certain oppositely to recommend character, character is oppositely recommended to correct the character inputted with this, without the need to moving the cursor to the character place of mistake and deleting by delete key, the operation steps such as then to re-enter, therefore, this programme can avoid the correction operation steps that current forward input method is loaded down with trivial details, improve input efficiency, and also improve the accuracy of input character.When the character inputted is accurate, also can not start retroactive association retrieval, can think that having have passed through a retroactive association confirms, thus constantly carrying out of retroactive association can be avoided, affect the normal use of user.
In technique scheme, preferably, described basic dictionary is used for the character of described input and described reverse recommendation character to carry out corresponding stored; Described retrieval unit is used for when the quantity of the character of the last time input is multiple, according to last character in the character of described the last time input, retrieves the reverse recommendation character be associated with last character described from described basic dictionary.
In once input, can disposable input character or multiple character, when only inputting a character, fairly simple, directly corresponding oppositely searching character can be gone out based on this character retrieval, and when the character of disposable input is multiple, oppositely character can be recommended by last character.Should be understood that the character described by book can be a phrase or a word, such as, when disposable input three words, retrieve based on the last character, when disposable input two phrases, retrieve based on last phrase.Except this retrieval base case, other retrieval bases can also be set, such as, do not consider the number of the character of disposable input, carry out retrieving oppositely recommending character based on all characters of current input, identical object can be reached equally.
In technique scheme, preferably, also comprise: sequencing unit, when exporting described reverse recommendation character, based on inputted before the character of described input and the character adjacent with the character of described input, the reverse recommendation character retrieved is sorted.
Because the reverse recommendation character of character is a lot, is not easy to navigate to the character meeting current context, views the reverse recommendation character meeting current context or meet user's request for the ease of user fast, can the reverse recommendation character retrieved be sorted.
In technique scheme, preferably, described sequencing unit comprises: coupling subelement, for by the phonetic of described reverse recommendation character, character shape and/or quantity, carry out matching primitives with the phonetic of described adjacent character, character shape and/or quantity respectively, according to matching value size, described reverse recommendation character is sorted; Described terminal also comprises: character error correction unit, when inputting the reverse recommendation character that user selects, described adjacent character is replaced to the reverse recommendation character of described selection.
When sorting, can sort to the reverse searching character retrieved according to the rule of setting, in a preferred version, the position reverse recommendation character arrangements of the number comparatively meeting the phonetic of the character that need correct, character shape and character easily can noticed user, the front end of such as list.Replace the character of the mistake inputted with the reverse recommendation character that user selects, thus achieve the object of error correction.
In technique scheme, preferably, also comprise: updating block, according to the use record of the reverse recommendation character that user selects, the reverse recommendation character be associated with the character of described input in described basic dictionary is upgraded.
In this technical scheme, according to the reverse recommendation character of user-selected use, statistics oppositely recommends the frequency of utilization of character, reverse recommendation character according to frequency of utilization and new choice for use upgrades basic dictionary, so as user again input character time user can be provided desired reverse recommendation character more accurately.
According to another aspect of the invention, also proposed a kind of input method, comprising: terminal receives the character of input; In basic dictionary, the reverse recommendation character be associated with the character of described input is retrieved according to the character of described input; Export described reverse recommendation character; Wherein, the character of described reverse recommendation character for inputting before being corrected in the character of described input.
In the basic dictionary of this programme, the all corresponding one or more reverse recommendation character of each character, therefore, whether the character no matter inputted is correct, all can go out the reverse recommendation character corresponding with the character of current input according to the character machining of current input and be shown to user, during the character errors inputted before the character of current input, user can select certain oppositely to recommend character, character is oppositely recommended to correct the character inputted with this, without the need to moving the cursor to the character place of mistake and deleting by delete key, the operation steps such as then to re-enter, therefore, this programme can avoid the correction operation steps that current forward input method is loaded down with trivial details, improve input efficiency, and also improve the accuracy of input character.When the character inputted is accurate, also can not start retroactive association retrieval, can think that having have passed through a retroactive association confirms, thus constantly carrying out of retroactive association can be avoided, affect the normal use of user.
In technique scheme, preferably, described basic dictionary is used for the character of described input and described reverse recommendation character to carry out corresponding stored; The step that the described character according to described input retrieves the reverse recommendation character be associated with the character of described input in basic dictionary specifically comprises: when the quantity of the character inputted in the last time is multiple, according to last character in the character of described the last time input, from described basic dictionary, retrieve the reverse recommendation character be associated with last character described.
In once input, can disposable input character or multiple character, when only inputting a character, fairly simple, directly corresponding oppositely searching character can be gone out based on this character retrieval, and when the character of disposable input is multiple, oppositely character can be recommended by last character.Should be understood that the character described by book can be a phrase or a word, such as, when disposable input three words, retrieve based on the last character, when disposable input two phrases, retrieve based on last phrase.Except this retrieval base case, other retrieval bases can also be set, such as, do not consider the number of the character of disposable input, carry out retrieving oppositely recommending character based on all characters of current input, identical object can be reached equally.
In technique scheme, preferably, when exporting described reverse recommendation character, based on inputted before the character of described input and the character adjacent with the character of described input, the reverse recommendation character retrieved is sorted.
Because the reverse recommendation character of character is a lot, is not easy to navigate to the character meeting current context, views the reverse recommendation character meeting current context or meet user's request for the ease of user fast, can the reverse recommendation character retrieved be sorted.
In technique scheme, preferably, by the phonetic of described reverse recommendation character, character shape and/or quantity, carry out matching primitives with the phonetic of described adjacent character, character shape and/or quantity respectively, according to matching value size, described reverse recommendation character is sorted; When inputting the reverse recommendation character that user selects, described adjacent character is replaced to the reverse recommendation character of described selection.
When sorting, can sort to the reverse searching character retrieved according to the rule of setting, in a preferred version, the position reverse recommendation character arrangements of the number comparatively meeting the phonetic of the character that need correct, character shape and character easily can noticed user, the front end of such as list.Replace the character of the mistake inputted with the reverse recommendation character that user selects, thus achieve the object of error correction.
In technique scheme, preferably, according to the use record of the reverse recommendation character that user selects, the reverse recommendation character be associated with the character of described input in described basic dictionary is upgraded.
In this technical scheme, according to the reverse recommendation character of user-selected use, statistics oppositely recommends the frequency of utilization of character, reverse recommendation character according to frequency of utilization and new choice for use upgrades basic dictionary, so as user again input character time user can be provided desired reverse recommendation character more accurately.
By reverse character association of the present invention scheme, error correction can be carried out to the character inputted by follow-up input character, make user when inputting Wrongly-written or mispronounced character, the character of deletion error need not be performed and re-enter the troublesome operation steps such as correct character, both improve the efficiency of input character, also improve the accuracy of input character, and then improve Consumer's Experience.
Accompanying drawing explanation
Fig. 1 shows the block diagram of terminal according to an embodiment of the invention;
Fig. 2 shows the process flow diagram of input method according to an embodiment of the invention;
Fig. 3 A to Fig. 3 F shows the schematic diagram realizing retroactive association error correction character according to an embodiment of the invention;
Fig. 4 shows the particular flow sheet of input method according to an embodiment of the invention.
Embodiment
In order to more clearly understand above-mentioned purpose of the present invention, feature and advantage, below in conjunction with the drawings and specific embodiments, the present invention is further described in detail.It should be noted that, when not conflicting, the feature in the embodiment of the application and embodiment can combine mutually.
Set forth a lot of detail in the following description so that fully understand the present invention; but; the present invention can also adopt other to be different from other modes described here and implement, and therefore, protection scope of the present invention is not by the restriction of following public specific embodiment.
Fig. 1 shows the block diagram of terminal according to an embodiment of the invention.
As shown in Figure 1, terminal 100 according to an embodiment of the invention, comprising: character input unit 102, for receiving the character of input; Retrieval unit 104, for retrieving the reverse recommendation character be associated with the character of described input in basic dictionary according to the character of described input; Reverse character recommendation unit 106, for exporting described reverse recommendation character, wherein, the character of described reverse recommendation character for inputting before being corrected in the character of described input.
In the basic dictionary of this programme, the all corresponding one or more reverse recommendation character of each character, therefore, whether the character no matter inputted is correct, all can go out the reverse recommendation character corresponding with the character of current input according to the character machining of current input and be shown to user, during the character errors inputted before the character of current input, user can select certain oppositely to recommend character, character is oppositely recommended to correct the character inputted with this, without the need to moving the cursor to the character place of mistake and deleting by delete key, the operation steps such as then to re-enter, therefore, this programme can avoid the correction operation steps that current forward input method is loaded down with trivial details, improve input efficiency, and also improve the accuracy of input character.When the character inputted is accurate, also can not start retroactive association retrieval, can think that having have passed through a retroactive association confirms, thus constantly carrying out of retroactive association can be avoided, affect the normal use of user.
In technique scheme, preferably, described basic dictionary is used for the character of described input and described reverse recommendation character to carry out corresponding stored; When described retrieval unit 104 is multiple for the quantity of the character inputted in the last time, according to last character in the character of described the last time input, from described basic dictionary, retrieve the reverse recommendation character be associated with last character described.
In once input, can disposable input character or multiple character, when only inputting a character, fairly simple, directly corresponding oppositely searching character can be gone out based on this character retrieval, and when the character of disposable input is multiple, oppositely character can be recommended by last character.Should be understood that the character described by book can be a phrase or a word, such as, when disposable input three words, retrieve based on the last character, when disposable input two phrases, retrieve based on last phrase.Except this retrieval base case, other retrieval bases can also be set, such as, do not consider the number of the character of disposable input, carry out retrieving oppositely recommending character based on all characters of current input, identical object can be reached equally.
In technique scheme, preferably, also comprise: sequencing unit 110, when exporting described reverse recommendation character, based on inputted before the character of described input and the character adjacent with the character of described input, the reverse recommendation character retrieved is sorted.
Because the reverse recommendation character of character is a lot, is not easy to navigate to the character meeting current context, views the reverse recommendation character meeting current context or meet user's request for the ease of user fast, can the reverse recommendation character retrieved be sorted.
In technique scheme, preferably, described sequencing unit 108 comprises: coupling subelement 1082, for by the phonetic of described reverse recommendation character, character shape and/or quantity, carry out matching primitives with the phonetic of described adjacent character, character shape and/or quantity respectively, according to matching value size, described reverse recommendation character is sorted; Described terminal also comprises: character error correction unit 110, when inputting the reverse recommendation character that user selects, described adjacent character is replaced to the reverse recommendation character of described selection.
When sorting, can sort to the reverse searching character retrieved according to the rule of setting, in a preferred version, the position reverse recommendation character arrangements of the number comparatively meeting the phonetic of the character that need correct, character shape and character easily can noticed user, the front end of such as list.Replace the character of the mistake inputted with the reverse recommendation character that user selects, thus achieve the object of error correction.
In technique scheme, preferably, also comprise: updating block 112, according to the use record of the reverse recommendation character that user selects, the reverse recommendation character be associated with the character of described input in described basic dictionary is upgraded.
In this technical scheme, according to the reverse recommendation character of user-selected use, statistics oppositely recommends the frequency of utilization of character, reverse recommendation character according to frequency of utilization and new choice for use upgrades basic dictionary, so as user again input character time user can be provided desired reverse recommendation character more accurately.
Fig. 2 shows the process flow diagram of input method according to an embodiment of the invention;
As shown in Figure 2, input method according to an embodiment of the invention, comprising: step 202, and terminal receives the character of input; Step 204, retrieves the reverse recommendation character be associated with the character of described input in basic dictionary according to the character of described input; Step 206, exports described reverse recommendation character, wherein, and the character of described reverse recommendation character for inputting before being corrected in the character of described input.
In the basic dictionary of this programme, the all corresponding one or more reverse recommendation character of each character, therefore, whether the character no matter inputted is correct, all can go out the reverse recommendation character corresponding with the character of current input according to the character machining of current input and be shown to user, during the character errors inputted before the character of current input, user can select certain oppositely to recommend character, character is oppositely recommended to correct the character inputted with this, without the need to moving the cursor to the character place of mistake and deleting by delete key, the operation steps such as then to re-enter, therefore, this programme can avoid the correction operation steps that current forward input method is loaded down with trivial details, improve input efficiency, and also improve the accuracy of input character.When the character inputted is accurate, also can not start retroactive association retrieval, can think that having have passed through a retroactive association confirms, thus constantly carrying out of retroactive association can be avoided, affect the normal use of user.
In technique scheme, preferably, described basic dictionary is used for the character of described input and described reverse recommendation character to carry out corresponding stored; The step that the described character according to described input retrieves the reverse recommendation character be associated with the character of described input in basic dictionary specifically comprises: when the quantity of the character inputted in the last time is multiple, according to last character in the character of described the last time input, from described basic dictionary, retrieve the reverse recommendation character be associated with last character described.
In once input, can disposable input character or multiple character, when only inputting a character, fairly simple, directly corresponding oppositely searching character can be gone out based on this character retrieval, and when the character of disposable input is multiple, oppositely character can be recommended by last character.Should be understood that the character described by book can be a phrase or a word, such as, when disposable input three words, retrieve based on the last character, when disposable input two phrases, retrieve based on last phrase.Except this retrieval base case, other retrieval bases can also be set, such as, do not consider the number of the character of disposable input, carry out retrieving oppositely recommending character based on all characters of current input, identical object can be reached equally.
In technique scheme, preferably, when exporting described reverse recommendation character, based on inputted before the character of described input and the character adjacent with the character of described input, the reverse recommendation character retrieved is sorted.
Because the reverse recommendation character of character is a lot, is not easy to navigate to the character meeting current context, views the reverse recommendation character meeting current context or meet user's request for the ease of user fast, can the reverse recommendation character retrieved be sorted.
In technique scheme, preferably, by the phonetic of described reverse recommendation character, character shape and/or quantity, carry out matching primitives with the phonetic of described adjacent character, character shape and/or quantity respectively, according to matching value size, described reverse recommendation character is sorted; When inputting the reverse recommendation character that user selects, described adjacent character is replaced to the reverse recommendation character of described selection.
When sorting, can sort to the reverse searching character retrieved according to the rule of setting, in a preferred version, the position reverse recommendation character arrangements of the number comparatively meeting the phonetic of the character that need correct, character shape and character easily can noticed user, the front end of such as list.Replace the character of the mistake inputted with the reverse recommendation character that user selects, thus achieve the object of error correction.
In technique scheme, preferably, according to the use record of the reverse recommendation character that user selects, the reverse recommendation character be associated with the character of described input in described basic dictionary is upgraded.
In this technical scheme, according to the reverse recommendation character of user-selected use, statistics oppositely recommends the frequency of utilization of character, reverse recommendation character according to frequency of utilization and new choice for use upgrades basic dictionary, so as user again input character time user can be provided desired reverse recommendation character more accurately.
Fig. 3 A to Fig. 3 F shows the schematic diagram realizing retroactive association error correction character according to an embodiment of the invention.
Below in conjunction with 3A to Fig. 3 F, specifically describe the retroactive association error correction how realizing input character.
First step: the character receiving user's input.
The character of user's input has two kinds of situations (with reference to figure 3A): situation 1, and the content of user's input is only a character, directly exports the reverse recommendation character of this character; Situation 2, the content of user's input is multiple characters, just based on last character, exports oppositely recommend character to the input character before this character.Situation 1 is applied more when user uses hand-writing input method, situation 2 is applied more when user uses spelling input method.
The input character of the user that system acceptance arrives might not as shown in Figure 3A, and the character namely inputted always is correct, in a lot of situation, may the character of input error, and such as user wants to input " contact person ", and the result that may input is as shown in Figure 3 B.Situation 1 represents, " connection " has been write as " listening " by user when hand-written; Situation 2 represents, " contact " has been write as " exercise " when phonetic by user.
Second step: the character according to user's input retrieves the reverse recommendation character corresponding with this character in preset basic dictionary.
The content of basic dictionary and the content of existing Chinese character input method basically identical, by the extensive character repertoire such as history People's Daily data, the Chinese character word of training out according to relational models such as Markov models and relational model storehouse.What the relational model of existing input method all described is positive relationship, namely based on input character before input character, sets up the relation list of successive character related with it.Character in relation list all has different matching degrees, according to user to this character using frequency adjustment matching degree.
And basic dictionary of the present invention and involved relational model are compared with conventional model, difference is to comprise a kind of inverse relationship, namely based on current character, sets up the relation list (with reference to figure 3F) of forward direction character (oppositely recommend character) related with it.Character in relation list all has different matching degrees, according to user to this character using frequency adjustment matching degree.
A character and multiple character two kinds of situations are divided to describe principle and the flow process of this step below.
First for a character situation as shown in Figure 3 C, the situation 1 that namely Fig. 3 A and Fig. 3 B is corresponding.As shown in Figure 3 C, in one case, when first time inputs, input " connection ", when second time inputs, input " being ", this situation is the situation that input is correct; In another case, when first time inputs, input " listening ", when second time inputs, input " being ", this situation is the situation of input error.For arbitrary situation, from the character of user's second time input, enter the treatment scheme of this programme.System retrieves the reverse recommendation character of its correspondence according to " being " character in basic dictionary.As shown in Figure 3 D, the reverse recommendation character retrieved is expressed with the word form comprising current character, and e.g., the words such as " contact " " relation " " direct line of descent ", are convenient to user and understand.When input error, user can select a character praised in the reverse recommendation character of display, such as, select " contact ", replace " listen and be " for " contact ", or the reverse recommendation character retrieved is " connection ", replaces " listening " with " connection ".
For user's input is the situation of multiple character, as shown in FIGURE 3 E." contact person " when first time inputs, may be entered as " amphibious people " by user; Also when first time inputs, " contact " may be entered as " exercise ", when second time input afterwards, have input again " people ".
In the case, the mode of character is oppositely recommended in retrieval, adopts last character of the current input phrase of user as benchmark, retrieves the reverse recommendation character of its correspondence in basic dictionary.As, user input is " contact person " or " amphibious people ", namely with " people " for benchmark, according to the reverse recommendation character retrieving its correspondence in dictionary.As shown in FIGURE 3 E, the reverse recommendation character retrieved is expressed with the word form comprising current character, and e.g., the word such as " contact person ", " exercise people ", " having pity on people ", is convenient to user and understands.
First user is inputted to the situation of " exercise "+" people ", can be understood as and first carried out once reverse correcting process for " exercise ", this flow process is the reverse correcting process carried out for " people " character.
Here for the ease of understanding, only list the reverse recommendation character that matching degree in dictionary is high, a series of relation character can may be there is for " people " in actual dictionary, as " good (people) ", " bad (people) ", " bully (people), " scolding (people) " etc. (with reference to figure 3F), this step does not illustrate one by one.
Third step: sort to oppositely recommending character.
After second step, retrieved a series of reverse recommendation character for current character, as shown in Figure 3 C for current " being " character retrieval to " connection (being) ", the word such as " closing (being) ", " closely related (being) ", and " contact (people) " of arriving for current " people " character retrieval shown in Fig. 3 E, " exercise (people) ", " having pity on people ", " good (people) ", " bad (people) ", " bullying (people) ", words such as " scoldings (people) ".This step can sort to the reverse recommendation character retrieved.
The rule of sequence is that the existing character word inputted before current character by user carries out association, as in Fig. 3 E, user have input " amphibious people ", the vocabulary that relation is more mated again is retrieved in the character lists that rule corresponding to " amphibious " retrieves from step 2, as " contact (people) ", " Jiangxi (people) ", " exercise (people) ", " having pity on (people) etc. ".If the degree of approximation of " contact (people) " is apparently higher than the degree of approximation of " scolding (people) ".
For the situation before user's current character being multiple character, as " amphibious (people) ", highlight quantity and pronunciation two rules that ordering rule is character in word.
Described character quantity rule refers to oppositely recommends character identical with the quantity of the character that user has inputted (treating the character of error correction).
Described pronunciation rule refers to corresponding character that user inputted multiple character and oppositely recommends the corresponding character of word to have identical or close initial consonant or simple or compound vowel of a Chinese syllable.The initial that initial character as " amphibious " is corresponding is " L ", so oppositely recommends the initial of word also to need for " L ", or close be " J " or " K " (the two in mobile phone key with " L " on same button).
For user's first time input is the situation (situation of handwriting input) of a character, and as " listening (being) ", the ordering rule highlighted is the degree of approximation of character stroke structure.The degree of approximation as corresponding in " connection (being) " will apparently higher than the degree of approximation of " closing (being) " correspondence.
4th step: the reverse recommendation character after display sequence.
For the result after the reverse recommendation character sequence that third step is retrieved further, first the character that first oppositely recommends character and user to input is compared, if the two is identical, then terminate checking, do not show reverse recommendation information, start downstream, that such as user has inputted is " contact person ", oppositely recommends the word that in sequence, the degree of approximation is the highest to be also that " contact (people) " just belongs to this situation.As corresponding in the dotted line in Fig. 3 C and Fig. 3 E is exactly this situation.
If the two is different, the highest several reverse recommendation character of matching degree can be shown to user, also character the highest for matching degree acquiescence can be substituted user's input character.As user's input is " listening "+" being ", so can recommend reverse error correction character lists " contact ", " direct line of descent ", " relation " etc. to user, also can adopt " contact " direct " listen and be " of replacing user and having inputted.
5th step: upgrade basic dictionary according to the reverse recommendation character that user selects.
After the 4th step, user can select to recommend error correction character to correct the character that mistake inputs accordingly.For a certain item character, to the corrigendum of the input character before it, the matching degree of this character and correct characters can be increased.Recommend error correction character to correct if user does not select to adopt, then the phrase of the character word connected mode that is not accordant to the old routine user inputted adds basic dictionary, but its matching degree can come finally.As illustrated in Figure 3 F, for character " people ", its corresponding recommendation list comprises " good (people) ", " bad (people) ", " contact (people) " etc., and each recommendation character all has different matching degrees.In the third step of this flow process, if certain recommends character to come first, and the character that user has inputted is identical with it, or user adopts it, and certain recommends character to carry out error correction, as " contact ", then the matching degree that " contact " is corresponding can increase, and the mode of increase does not limit.
Fig. 4 shows input method particular flow sheet according to an embodiment of the invention.
Step 402, terminal receives input character.Judge whether to open retroactive association error correction flow process, if open, then enter step 404, otherwise process ends.
Step 404, enters basic dictionary according to the character of user's input and oppositely recommends character retrieval.
Step 406, sorts to oppositely recommending character according to the character inputted before current input character (namely treating the character of error correction).
Step 408, the character the highest by matching degree in the reverse recommendation character after sequence and the character inputted before the character of current input compare, if the same, then do not provide and oppositely recommend character, if different, then provide and oppositely recommend character.
Step 410, provides to user and oppositely recommends character to allow user select the correct character that will input.
Step 412, replaces input character (treating error correction character) with oppositely recommending character.
Step 414, upgrades basic dictionary according to the reverse recommendation character that user uses.
More than be described with reference to the accompanying drawings technical scheme of the present invention, by reverse character association of the present invention scheme, error correction can be carried out to the character inputted by follow-up input character, make user when inputting Wrongly-written or mispronounced character, the character of deletion error need not be performed and re-enter the troublesome operation steps such as correct character, both improve the efficiency of input character, also improved the accuracy of input character, and then improve Consumer's Experience.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (8)

1. a terminal, is characterized in that, comprising:
Character input unit, for receiving the character of input;
Retrieval unit, for retrieving the reverse recommendation character be associated with the character of described input in basic dictionary according to the character of described input;
Reverse character recommendation unit, for exporting described reverse recommendation character,
Wherein, the character of described reverse recommendation character for inputting before being corrected in the character of described input;
Described basic dictionary is used for the character of described input and described reverse recommendation character to carry out corresponding stored;
Described retrieval unit is used for when the quantity of the character of the last time input is multiple, according to last character in the character of described the last time input, retrieves the reverse recommendation character be associated with last character described from described basic dictionary.
2. terminal according to claim 1, is characterized in that, also comprises:
Sequencing unit, when exporting described reverse recommendation character, based on inputted before the character of described input and the character adjacent with the character of described input, the reverse recommendation character retrieved is sorted.
3. terminal according to claim 2, it is characterized in that, described sequencing unit comprises: coupling subelement, for by the phonetic of described reverse recommendation character, character shape and/or quantity, carry out matching primitives with the phonetic of described adjacent character, character shape and/or quantity respectively, according to matching value size, described reverse recommendation character is sorted;
Described terminal also comprises: character error correction unit, when inputting the reverse recommendation character that user selects, described adjacent character is replaced to the reverse recommendation character of described selection.
4. terminal according to any one of claim 1 to 3, is characterized in that, also comprises:
Updating block, according to the use record of the reverse recommendation character that user selects, upgrades the reverse recommendation character be associated with the character of described input in described basic dictionary.
5. an input method, is characterized in that, comprising:
Terminal receives the character of input;
In basic dictionary, the reverse recommendation character be associated with the character of described input is retrieved according to the character of described input;
Export described reverse recommendation character,
Wherein, the character of described reverse recommendation character for inputting before being corrected in the character of described input;
Described basic dictionary is used for the character of described input and described reverse recommendation character to carry out corresponding stored;
The step that the described character according to described input retrieves the reverse recommendation character be associated with the character of described input in basic dictionary specifically comprises:
When the quantity of the character of the last time input is multiple, according to last character in the character of described the last time input, from described basic dictionary, retrieve the reverse recommendation character be associated with last character described.
6. input method according to claim 5, is characterized in that, when exporting described reverse recommendation character, based on inputted before the character of described input and the character adjacent with the character of described input, the reverse recommendation character retrieved is sorted.
7. input method according to claim 6, it is characterized in that, by the phonetic of described reverse recommendation character, character shape and/or quantity, carry out matching primitives with the phonetic of described adjacent character, character shape and/or quantity respectively, according to matching value size, described reverse recommendation character is sorted;
When inputting the reverse recommendation character that user selects, described adjacent character is replaced to the reverse recommendation character of described selection.
8. the input method according to any one of claim 5 to 7, is characterized in that, according to the use record of the reverse recommendation character that user selects, upgrades the reverse recommendation character be associated with the character of described input in described basic dictionary.
CN201310026097.9A 2013-01-18 2013-01-18 Terminal and input method Active CN103064531B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310026097.9A CN103064531B (en) 2013-01-18 2013-01-18 Terminal and input method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310026097.9A CN103064531B (en) 2013-01-18 2013-01-18 Terminal and input method

Publications (2)

Publication Number Publication Date
CN103064531A CN103064531A (en) 2013-04-24
CN103064531B true CN103064531B (en) 2016-04-06

Family

ID=48107182

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310026097.9A Active CN103064531B (en) 2013-01-18 2013-01-18 Terminal and input method

Country Status (1)

Country Link
CN (1) CN103064531B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106774975B (en) * 2016-11-30 2020-03-31 百度在线网络技术(北京)有限公司 Input method and device
CN106774978A (en) * 2016-12-16 2017-05-31 百度在线网络技术(北京)有限公司 Method and device for correcting entry
TWI771720B (en) 2020-07-24 2022-07-21 華碩電腦股份有限公司 Identification method with multiple type input and electronic device using the same

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN85100732A (en) * 1985-04-01 1986-08-06 扶良文 The optional compatible intelligent keyboard system in Sino-Japan west
CN1272755A (en) * 1999-05-03 2000-11-08 电话通有限公司 Spelling correcting system for bidirectional mobile communication equipment
CN1782974A (en) * 2004-11-01 2006-06-07 诺基亚有限公司 Improved mobile phone and method
CN1952856A (en) * 2005-10-21 2007-04-25 钟巨航 Input method
CN101639830A (en) * 2009-09-08 2010-02-03 西安交通大学 Chinese term automatic correction method in input process
CN101788855A (en) * 2010-03-09 2010-07-28 华为终端有限公司 Method, device and communication terminal for obtaining user input information
CN102156551A (en) * 2011-03-30 2011-08-17 北京搜狗科技发展有限公司 Method and system for correcting error of word input

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN85100732A (en) * 1985-04-01 1986-08-06 扶良文 The optional compatible intelligent keyboard system in Sino-Japan west
CN1272755A (en) * 1999-05-03 2000-11-08 电话通有限公司 Spelling correcting system for bidirectional mobile communication equipment
CN1782974A (en) * 2004-11-01 2006-06-07 诺基亚有限公司 Improved mobile phone and method
CN1952856A (en) * 2005-10-21 2007-04-25 钟巨航 Input method
CN101639830A (en) * 2009-09-08 2010-02-03 西安交通大学 Chinese term automatic correction method in input process
CN101788855A (en) * 2010-03-09 2010-07-28 华为终端有限公司 Method, device and communication terminal for obtaining user input information
CN102156551A (en) * 2011-03-30 2011-08-17 北京搜狗科技发展有限公司 Method and system for correcting error of word input

Also Published As

Publication number Publication date
CN103064531A (en) 2013-04-24

Similar Documents

Publication Publication Date Title
CN106598939B (en) A kind of text error correction method and device, server, storage medium
KR102668530B1 (en) Speech recognition methods, devices and devices, and storage media
US20190087084A1 (en) User-centric soft keyboard predictive technologies
CN104718544B (en) The method of part gesture text input calculates equipment and system
CN107704102B (en) Text input method and device
CN103502915B (en) System and method for implementing sliding input of text based upon on-screen soft keyboard on electronic equipment
CN103903619B (en) A kind of method and system improving speech recognition accuracy
CN104020943B (en) Character string is replaced
CN103777774B (en) The word error correction method of terminal installation and input method
CN111310440B (en) Text error correction method, device and system
CN102830924B (en) A kind of method and device adjusting entering method keyboard
TW200842613A (en) Spell-check for a keyboard system with automatic correction
EP2765473A1 (en) Input processing method and apparatus
CN105094368A (en) Control method and control device for frequency modulation ordering of input method candidate item
CN104007836B (en) A kind of processing method and terminal device of handwritten word input
CN101149806B (en) Method and device for hand writing identification post treatment using context information
CN102063482A (en) High-efficiency contact searching method of handheld device
US10152473B2 (en) English input method and input device
CN102117171A (en) Spelling input system and method
CN101287026A (en) System and method for executing quick dialing by hand-write recognition function
CN103064531B (en) Terminal and input method
CN108803890A (en) A kind of input method, input unit and the device for input
CN113342948A (en) Intelligent question and answer method and device
CN101405693A (en) Personal synergic filtering of multimodal inputs
CN105164669A (en) Information processing apparatus, information processing method, and program

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant