CN109976548A - A kind of input method and input unit - Google Patents

A kind of input method and input unit Download PDF

Info

Publication number
CN109976548A
CN109976548A CN201711463446.8A CN201711463446A CN109976548A CN 109976548 A CN109976548 A CN 109976548A CN 201711463446 A CN201711463446 A CN 201711463446A CN 109976548 A CN109976548 A CN 109976548A
Authority
CN
China
Prior art keywords
character
string
target character
target
character string
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.)
Granted
Application number
CN201711463446.8A
Other languages
Chinese (zh)
Other versions
CN109976548B (en
Inventor
侯柏岑
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.)
Beijing Sogou Technology Development Co Ltd
Original Assignee
Beijing Sogou Technology Development 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 Beijing Sogou Technology Development Co Ltd filed Critical Beijing Sogou Technology Development Co Ltd
Priority to CN201711463446.8A priority Critical patent/CN109976548B/en
Publication of CN109976548A publication Critical patent/CN109976548A/en
Application granted granted Critical
Publication of CN109976548B publication Critical patent/CN109976548B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
    • G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
    • G06F3/0233—Character input methods
    • G06F3/0236—Character input methods using selection techniques to select from displayed items
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
    • G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
    • G06F3/0233—Character input methods
    • G06F3/0237—Character input methods using prediction or retrieval techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Document Processing Apparatus (AREA)

Abstract

The embodiment of the invention discloses a kind of input methods, when using input method input, when user is not in the case where knowing the accurate representation of target character, target character can be described, obtain the character string for characterizing the description, then target character is determined according to the character string, and is shown target character as candidate item.Wherein, the character string can characterize the description to target character, therefore, the character string and target character both correspond to the same description, and character string can characterize the description, and the description can characterize the target character, even if user does not know when being inputted using input method, the accurate representation method of the target character, input method can automatically determine the target character according to the character string, it realizes the input to the target character, improves user experience.The embodiment of the invention also discloses a kind of input units.

Description

A kind of input method and input unit
Technical field
The present invention relates to Internet technical fields, more particularly to a kind of input method and input unit.
Background technique
Input method is the coding method that one kind can input characters into equipment (such as: computer, mobile phone etc.).User is using defeated When entering method input target character, need to know the accurate representation of the target character, then using coding provided by input method Method inputs the target character.By taking Chinese pinyin inputs as an example, user needs to know the pronunciation for the target Chinese to be inputted, The phonetic of the target Chinese is inputted in Chinese character coding input method, Chinese character coding input method can provide the candidate of the target Chinese according to the phonetic , upper screen is carried out to the target Chinese, realizes the input to the target Chinese.Again by taking English input as an example, user needs to know Target English how is spelt, English input method is directly used, according to the spelling sequence of target English, sequentially inputs English words Symbol realizes the input to target English.
It is accurate when user does not know target character using input method input but in many actual application scenarios Representation, such as: the pronunciation of target Chinese is not known, or, it is not known that how English is spelt.At this point, user can not make The input to the target character is realized with input method.
Summary of the invention
Present invention solves the technical problem that it is to provide a kind of input method and input unit, thus, even if user does not know Accurate representation when road target character is inputted using input method, also may be implemented the input to target character.
For this purpose, the technical solution that the present invention solves technical problem is:
To solve the above-mentioned problems, the embodiment of the invention provides a kind of input methods for first aspect, this method comprises:
The character string of input is obtained, character string is used to characterize the description to target character；
Target character is determined according to character string；
It is shown target character as candidate item.
Optionally, determine that target character includes: according to character string
From preset attribute library, the corresponding target character of character string is searched, preset attribute library includes characterization character The character string of attribute description and the corresponding relationship of character.
Optionally, character string includes pinyin string, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Semantic analysis is carried out to attribute description content and determines target character.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the charcter topology of target character；
Obtain the corresponding target character of charcter topology.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the character title of target character；
Obtain the corresponding target character of character title.
Optionally, character string includes pinyin string, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Search the search entry with attribute description content matching；
Target character is determined according to the corresponding search result of search entry.
Optionally, target character is carried out display as candidate item includes:
When target character is Chinese character, the phonetic of Chinese character is obtained；
It is shown the phonetic of Chinese character and Chinese character as the same candidate item.
Optionally, character string includes English string, determines that target character includes: according to character string
English is gone here and there and carries out semantic analysis；
Obtain the target character with English string semantic matches.
Second aspect, to solve the above-mentioned problems, the embodiment of the invention provides a kind of input unit, which includes:
Acquiring unit, for obtaining the character string of input, character string is used to characterize the description to target character；
Determination unit, for determining target character according to character string；
Display unit, for being shown target character as candidate item.
Optionally, determination unit is specifically used for:
From preset attribute library, the corresponding target character of character string is searched, preset attribute library includes characterization character The character string of attribute description and the corresponding relationship of character.
Optionally, determination unit includes:
Attribute description content determines subelement, for determining attribute description according to character string when character string is pinyin string Content；
Semantic analysis subelement determines target character for carrying out semantic analysis to attribute description content.
Optionally, semantic analysis subelement includes:
Charcter topology obtains subelement, for carrying out semantic analysis to attribute description content, obtains the character of target character Structure；
Target character obtains subelement, for obtaining the corresponding target character of charcter topology.
Optionally, semantic analysis subelement includes:
Character name acquiring subelement obtains the character of target character for carrying out semantic analysis to attribute description content Title；
Target character obtains subelement, for obtaining the corresponding target character of character title.
Optionally, determination unit includes:
Attribute description content determines subelement, for determining attribute description according to character string when character string is pinyin string Content；
Subelement is searched, for searching and the search entry of attribute description content matching；
Target character determines subelement, for determining target character according to the corresponding search result of search entry.
Optionally, display unit includes:
Phonetic obtains subelement, for obtaining the phonetic of Chinese character when target character is Chinese character；
Subelement is shown, for showing the phonetic of Chinese character and Chinese character as the same candidate item.
Optionally, determination unit includes:
Semantic analysis subelement, for being gone here and there to English and carrying out semantic analysis when character string is English string；
Target character obtains subelement, for obtaining the target character with English string semantic matches.
The third aspect, it to include memory and one or one that the embodiment of the invention provides a kind of electronic equipment Above program, one of them perhaps more than one program be stored in memory and be configured to by one or one with It includes the instruction for performing the following operation that upper processor, which executes the one or more programs:
The character string of input is obtained, character string is used to characterize the description to target character；
Target character is determined according to character string；
It is shown target character as candidate item.
Optionally, determine that target character includes: according to character string
From preset attribute library, the corresponding target character of character string is searched, preset attribute library includes characterization character The character string of attribute description and the corresponding relationship of character.
Optionally, character string includes pinyin string, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Semantic analysis is carried out to attribute description content and determines target character.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the charcter topology of target character；
Obtain the corresponding target character of charcter topology.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the character title of target character；
Obtain the corresponding target character of character title.
Optionally, character string includes pinyin string, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Search the search entry with attribute description content matching；
Target character is determined according to the corresponding search result of search entry.
Optionally, target character is carried out display as candidate item includes:
When target character is Chinese character, the phonetic of Chinese character is obtained；
It is shown the phonetic of Chinese character and Chinese character as the same candidate item.
Optionally, character string includes English string, determines that target character includes: according to character string
English is gone here and there and carries out semantic analysis；
Obtain the target character with English string semantic matches.
Fourth aspect, inventive embodiments provide a kind of non-transitorycomputer readable storage medium, when the storage is situated between When instruction in matter is executed by the processor of electronic equipment, so that electronic equipment is able to carry out a kind of input method, this method packet It includes:
The character string of input is obtained, character string is used to characterize the description to target character；
Target character is determined according to character string；
It is shown target character as candidate item.
Optionally, determine that target character includes: according to character string
From preset attribute library, the corresponding target character of character string is searched, preset attribute library includes characterization character The character string of attribute description and the corresponding relationship of character.
Optionally, character string includes pinyin string, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Semantic analysis is carried out to attribute description content and determines target character.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the charcter topology of target character；
Obtain the corresponding target character of charcter topology.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the character title of target character；
Obtain the corresponding target character of character title.
Optionally, character string includes pinyin string, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Search the search entry with attribute description content matching；
Target character is determined according to the corresponding search result of search entry.
Optionally, target character is carried out display as candidate item includes:
When target character is Chinese character, the phonetic of Chinese character is obtained；
It is shown the phonetic of Chinese character and Chinese character as the same candidate item.
Optionally, character string includes English string, determines that target character includes: according to character string
English is gone here and there and carries out semantic analysis；
Obtain the target character with English string semantic matches.
According to the above-mentioned technical solution, the method have the advantages that:
It is found that when using input method input, when user is not the case where knowing the accurate representation of target character Under, target character can be described, then the character string that acquisition characterizes the description determines target according to the character string Character, and shown target character as candidate item.Wherein, which can characterize the description to target character, because This, the character string and target character both correspond to the same description, and character string can characterize the description, and the description being capable of table The target character is levied, therefore, even if user does not know that the accurate representation method of the target character is defeated when inputting using input method The target character can be automatically determined according to the character string by entering method, realized the input to the target character, improved user's body It tests.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with It obtains other drawings based on these drawings.
Fig. 1 show a kind of flow diagram of input method provided in an embodiment of the present invention；
Fig. 2A show in a kind of input method provided in an embodiment of the present invention showing for the only candidate item of displaying target character It is intended to；
Fig. 2 B is shown in a kind of input method provided in an embodiment of the present invention to target character candidate item and character string pair The schematic diagram that the candidate item answered is shown in same display area；
Fig. 2 C is shown in a kind of input method provided in an embodiment of the present invention to target character candidate item and character string pair The schematic diagram that the candidate item answered is shown in same display area；
Fig. 2 D is shown in a kind of input method provided in an embodiment of the present invention to target character candidate item and character string pair The schematic diagram that the candidate item answered is shown respectively in different display areas；
Fig. 3 show a kind of pair of attribute description content progress semantic analysis provided in an embodiment of the present invention and determines target character The flow chart of method；
Fig. 4 show a kind of pair of attribute description content progress semantic analysis provided in an embodiment of the present invention and determines target character The flow chart of method；
Fig. 5 show a kind of schematic diagram that user is searched in a search engine for target character；
Fig. 6 show a kind of flow chart of method that target character is determined according to character string provided in an embodiment of the present invention；
Fig. 7 show the search result schematic diagram directly scanned for pinyin string；
Fig. 8 show the stream of the input method of target character when a kind of character string provided in an embodiment of the present invention is English string Cheng Tu；
Fig. 9 show the effect signal of target character input when a kind of character string provided in an embodiment of the present invention is English string Figure；
Figure 10 is a kind of structural schematic diagram of input unit in the embodiment of the present invention；
Figure 11 is input unit hardware structural diagram provided in an embodiment of the present invention.
Specific embodiment
In many actual application scenarios, user does not know the accurate representation of target character, such as: do not know The pronunciation of target Chinese, or, it is not known that how English is spelt.It realizes at this point, user is not available input method to the target The input of character.
For example, being directed to character " Yao ", user does not know the pronunciation of the character, when user wants to input the character, does not know How road uses input method to realize the input to the character.
It is not available input of the input method realization to target character for user caused by the pronunciation for not knowing target Chinese, The prior art provides a kind of input method of dividing by means of characters to realize the input of target character.The input method of dividing by means of characters is specially by mesh Marking-up symbol splits into multiple sub- characters according to charcter topology, and input splits the corresponding substring of the sub- character of obtain multiple and obtains Character string determines the corresponding character of the character string as target word according to the corresponding relationship of character string and character in dividing by means of characters dictionary Symbol, shows target character as candidate item, so that user can realize according to target character candidate item to the mesh The input of marking-up symbol.
For example, being directed to Chinese character " Yao " this character, user does not know its pronunciation, according to the input side to divide by means of characters in the prior art " Yao " first can be split into three sub- characters, respectively " soil ", " soil ", " soil ", these three sub- characters according to charcter topology by method Corresponding substring is respectively " tu ", " tu ", " tu ", inputs the corresponding substring of these three sub- characters, can obtain word Symbol string " tututu " can determine that character string " tututu " is corresponding according to the corresponding relationship of character string and character in dividing by means of characters dictionary Target character be " Yao ", " Yao " is shown as candidate item, user can be allowed according to candidate item realization pair The input of target character " Yao ".
However, the excessively sequencing of the input method of this dividing by means of characters, needs to split character, and clearly each fractionation The pronunciation of sub- character afterwards.It is difficult to split for some, cannot split, and the indefinite word of pronunciation of the sub- character after fractionation Symbol, it is clear that be also difficult to realize input using the input method of above-mentioned dividing by means of characters.For example, character " wrapping up in " is difficult to tear open according to charcter topology Point, thus the input method by dividing by means of characters is difficult to input.
In addition, the character string that above-mentioned dividing by means of characters method obtains, only to the corresponding substring of sub- character after fractionation into Row mechanically combine, the structural information of target character can not be embodied, i.e., can not determine target character be tiled configuration still on Flowering structure, or surround structure.Further, which sub- character can not be also determined on a left side, which sub- character is difficult on the right side Accurately determine target character.For this purpose, when input splits the corresponding substring of obtained multiple sub- characters, it is also necessary to according to The sequence that each sub- character is spelt in target character is inputted.For including the character string of identical substring, The input sequence of substring is different, can determine different target characters.If ignoring the input sequence of substring, meeting The target character for causing determining target character and user to be intended to input mismatches, and reduces the hit rate of target character.Having In a little situations, when as identical such as the sub- character pronunciation after splitting, even if inputting in sequence, the target character determined is also and different Fixed and user is intended to the character match of input.
For example, substring " jia " and substring " mu ", difference can obtain character string " jiamu " in sequence, also Character string " mujia " can be obtained.Input character string " jiamu " can determine that a target character is " frame ", and input word Symbol string " mujia " can determine that another target character is " cangue ".When user is intended to input character " cangue ", but the character inputted When string is " jiamu ", above-mentioned dividing by means of characters input method determines that target character is " frame ", without can determine that out " cangue ".Again for example: when with When family is intended to input character " returning ", character string " koukou " can be inputted, although " koukou " corresponding fractionation character can be true It is set to " mouth mouth ", but can not define " mouth mouth " by " koukou " is combined according to up-down structure, also according to encirclement Structure is combined.Therefore, the target character determined according to character string " koukou " may be " Lyu ", and the input with user is intended to And it mismatches.
It was found by the inventors of the present invention that if the character string of input can express semantic information, for example, to target character into Row description, inputs this and describes corresponding character string, then can not have to input character string according to the structural order of target character, can also To determine target character, the input of target character is realized.
By taking " cangue " word as an example, character string " zuobianyigemuyoubianyigejia " can be inputted, according to the character string The description " one, left side wood, one, the right add " of characterization, can determine that target character is " cangue ", rather than " frame ".Show at this In example, even if user does not input character string in sequence, e.g., input character string " youbianyigejiazuobianyigemu ", It can also determine that a target character is according to the description " one, the right adds, one, left side wood " that the character string is characterized " cangue ".Due to containing structural information in the character string of input, user do not need in strict accordance with target character structural order i.e. The input of target character can be achieved.Or character string " zuoyoujiegou, mujia ", according to the character string can also be inputted The description " tiled configuration, wood add " characterized, determines that a target character is " cangue ".The character string of the input contains target The structural information " tiled configuration " of character further includes the composition " wood and plus " of the target character, therefore, user do not need strictly by The input of target character can also be realized according to the structural order of target character.
Since character string can characterize the description to target character, and describing may include the composition knot to target character Structure, the title of target character, target character express the incidence relation of meaning or target character and other characters, therein any One may be incorporated for determining target character.Therefore, character string target character being described by inputting characterization, not only may be used To realize the target character input when not knowing Chinese pronunciation, also may be implemented not knowing the mesh in the case of charcter topology Character input is marked, can also realize the word input that do not know in the case of how English word is spelt.The embodiment of the present invention provides Input method be not limited to Chinese, can also be other language such as English, Japanese.For example, input word can be intended to user It is described, obtains the English string for characterizing the description, gone here and there according to the English and determine the word namely target character that are intended to input.
By taking " Zhao " word as an example, when user knows the structure of the word without knowing phonetic, character string can be inputted " shangmianyigeriyigeyuexiamianyigekong ", according to corresponding semantic information " day, one month above, A sky below " can determine that the target character is " Zhao ".When user neither knows the phonetic of the character, to the knot of the character When structure impression is also relatively fuzzyyer, it can also be inputted using other modes, such as input character string " wuzetiandemingzi ", It can determine that the character is " Zhao " according to the semantic information " name of Wu Tse-tien " that the string table reaches.When user do not know it is a certain When how English word is spelt, the character string of description can be inputted, such as " huobandeyingwendanci ", according to the character string Semantic information " English word of partner ", can determine a target character be " fellow ", by " fellow " in candidate item On be shown, may be implemented not know because how word inputs the English word when how to spell.
In order to provide the implementation for inputting target character in the case of the accurate representation for not knowing target character, this Inventive embodiments provide a kind of input method and input unit, below in conjunction with Figure of description to the preferred embodiment of the present invention It is illustrated, it should be understood that preferred embodiments described herein are only used to illustrate and explain the present invention, is not used to limit this Invention.And in the absence of conflict, the feature in the embodiment and embodiment in the present invention can be combined with each other.
Fig. 1 show a kind of flow chart of input method provided in an embodiment of the present invention, please refers to Fig. 1, this method comprises:
S101: obtaining the character string of input, and the character string is used to characterize the description to target character.
Target character can be understood as the character that user is intended to input.In embodiments of the present invention, user does not know how It is target character using the character that input method inputs.Wherein, target character can be any one in Chinese character, symbol or number Kind, target character can also be the word in English, and the text etc. in Japanese, it is not limited in the embodiment of the present invention.Make For an example, user wants input Chinese character " Yao ", but does not know the pronunciation of Chinese character " Yao ", Chinese character " Yao " can be used as target Character.User wants to carry out integral operation, but when not knowing how input integral symbol, symbol " ∫ " can be determined as mesh Marking-up symbol.Area or volume operation are carried out when user wants input pi, when specific value without knowing pi, may be used also Using by digital " 3.14 " as target character.
In embodiments of the present invention, when user wants to input target character using input method, and does not know how input, Such as do not know the pronunciation of character, target character can be described, then input characterizes the character string of the description.Input word The mode for according with string can also be inputted, the embodiment of the present application is to the defeated of character string to be manually entered by keyboard by voice Enter mode to be not specifically limited.
In one possible implementation, target character is described, can be the attribute progress to target character Description, to obtain the character string of the attribute description of characterization target character.It should be noted that the attribute to target character into When row description, it can use natural language and the attribute of target character be described, so that obtaining characterization utilizes natural language pair The character string of the attribute description of target character.
Wherein, the description of target character may include the charcter topology of target character, character title, character meaning and/or Character relation.It is appreciated that constraining target character, so that target character is different from any constraint condition of other characters Description can be used as the description of target character.Wherein, after determining how target character is described, benefit can be inputted With natural language to the character string of the attribute description of target character.It should be noted that natural language, which can be considered as one kind, to be had The language of concrete meaning.A kind of description for having concrete meaning can be considered as using attribute description of the natural language to target character, Rather than simple character is piled up or is combined.In order to make it easy to understand, illustrating.For target character " Yao ", can incite somebody to action (meaning is " above one by " sangetu " (meaning is " three soil ") or " shangmianyigetuxiamianlianggetu " A soil, following two soil ") as the attribute description to the target character, and " tututu " is only after splitting to target character The corresponding pinyin character string mechanical combination of character after as a result, and do not have concrete meaning, thus cannot function as the mesh The attribute description of marking-up symbol.
Above-mentioned example is the character string using pinyin string as characterization to the description of target character.Character string not only can be the Chinese The pinyin string of word is also possible to the pinyin string of symbol, can also be it is digital pinyin string.Expansion as the embodiment of the present invention Exhibition, character string can also be English string.For example, being directed to word " familiar " (meaning " familiar "), user forgets how to spell, (" strange antisense can be meant by the character string " the opposite of strange " that input describes the target character Word ") it realizes.Character string is also possible in pinyin string, English string or symbol at least two combination.For example, it is directed to character " line in the Eight Diagrams ", The character string for characterizing description can be " lianggeX ", and for character " Gua ", the character string for characterizing description can be "shangmianyigekou+xiamianyigedao".Above example predominantly from composed structure etc. to target character into Row description, can also be described target character by semanteme, incidence relation etc..Such as, for character " lonely ", description is characterized Character string can be " xiangziyiyang ", mean " as son ".
S102: the target character is determined according to character string.
In this step, it since character string can characterize the description to target character, is retouched according to the characterization obtained in S101 The character string stated can determine target character.
Character string and target character both correspond to the same description, and character string can characterize the description, and the description energy The target character is enough characterized, therefore, according to the character string that the characterization describes, can determine the target word to match with the description Symbol.Determine that target character can be there are many implementation according to the character string of characterization description.It can be according to the character of characterization description At least one of string and corresponding relationship, semantic analysis or the search behavior analysis based on character string of character, determine target Character.
In one possible implementation, attribute library can be preset, includes characterization word in the preset attribute library The character string of the attribute description of symbol and the corresponding relationship of the character.It, can be from default after the character string for getting user's input Attribute library in, search characterization attribute description the corresponding target character of character string.Pass through the category of character in preset attribute library Property description character string and character corresponding relationship, search corresponding target character, can quickly to user input character String is responded, and relative to the method that other determine target character, this method has higher query rate and shorter response Time improves user experience.Wherein, which can be pinyin string employed in Chinese character coding input method, can also be English English string in literary input method.Certainly, character string is not limited in that other Languages can also be used using pinyin string and English string Used character string, which is not described herein again.
Pinyin string can be converted as the extension of the embodiment of the present application when character string is the pinyin string of attribute description For Chinese character string corresponding with pinyin string, attribute description content is determined according to the Chinese character string, semantic point is carried out to attribute description content It analyses and determines target character.Wherein, carrying out semantic analysis to attribute description content can be, to the Chinese character string for indicating attribute description (i.e. attribute description content) is segmented, and is labeled or is divided by each participle of the modes such as neural network model to Chinese character string Class realizes the semantic analysis to attribute description content according to mark or classification results.The corresponding attribute of this acquisition character string is retouched Content is stated, the mode that semantic analysis determines target character is carried out to attribute description content, application range is more extensive, can be adapted for The pinyin string of any attribute description, though the pinyin string not in preset attribute library, can also be by way of semantic analysis The corresponding target character of the pinyin string is determined, to realize the input of target character.
When character string is the pinyin string of attribute description, the data acquisition target character in internet can be combined with.It obtains Taking the process of target character can be, determine attribute description content according to character string, search and the attribute description content matching Search entry, wherein search entry can be include keyword, phrase or sentence etc. as attribute description content, according to The corresponding search result of search entry determines target character.It is appreciated that including mass data in internet, by a large amount of numbers The corresponding target character of character string of characterization attributes description can be determined according to analysis is carried out, and realize the input of target character.Due to The process for determining target character is to be based on internet rather than pre-set attribute library, therefore, for not in attribute library Character string can also determine target character according to the search result of internet, and application is more extensive, can will be Internet-based Search result determines supplement of the mode of target character as the mode for determining target character based on attribute library, has both ensured inquiry Rate has also ensured application range.
Character string in addition to can be attribute description pinyin string, can also for attribute description English string.When character string is When English string, which can be gone here and there and carry out semantic analysis, obtained and the semantic target character to match of English string.It is this to English Text string carries out semantic analysis, acquisition and the semantic mode to matched target character of English string, so that input provided by the invention Method not only can be adapted for Chinese, can also be applied to English, and so on, also can be applied to other language includes day Text, French etc., greatly improve the scope of application and scalability of input method, input method may be implemented in different language Between smoothly switch, improve user experience.
It is the corresponding relationship of the character string of above-mentioned implementation dependence description and character, semantic analysis, Internet-based Multiple angles such as search result determine that the process of target character is illustrated to the character string described according to characterization attributes.By The attribute description using natural language to target character can be characterized in character string, which can be the word of target character The description of structure, character title, character meaning or at least one attribute in character relation is accorded with, thus can be according to characterization attributes The character string of description determines target character, and the method for this determining target character can overcome the input method of traditional dividing by means of characters The indefinite disadvantage for causing hit rate not high of middle character structure information.Moreover, being described using natural language, more close to true Real interaction scenarios, user can be according to their own needs to the characters for being intended to input, from charcter topology, character title, character Any one angle such as meaning or character relation inputs the character string of characterization attributes description, does not need user according to dividing by means of characters sequence Character string is inputted, so that input method is more intelligent.
S103: it is shown using the target character as candidate item.
After determining target character, it can be shown target character as candidate item.In a kind of possible realization In mode, only the candidate item of target character can be shown, can also candidate item to target character and character string it is corresponding Candidate item shown.
It, can be in same display area when the corresponding candidate item of candidate item and character string to target character is shown It is shown.The sequence of the corresponding candidate item of candidate item and character string of target character can be set according to demand.In order to The interference that candidate item is corresponded to character string is reduced, it can be by the candidate item of target character in the laggard of the corresponding candidate item of character string Row display.In some cases, the candidate item of target character can also be shown before the corresponding candidate item of character string, To facilitate user to select and input.
It, can also be in different viewing areas when the corresponding candidate item of candidate item and character string to target character is shown Domain is shown respectively.By different display areas to the corresponding candidate item of candidate item and character string of target character respectively into Row display, can interference to avoid target character candidate item to the corresponding candidate item of character string, while being also convenient for user and selecting It selects, to realize that target character inputs.
Below with reference to specific example, display is carried out as candidate item to target character and is illustrated.
Fig. 2A shows a kind of schematic diagram of only candidate item of displaying target character.As shown in Figure 2 A, it is retouched in input characterization After the character string " sangetu " stated, according to the character string that the characterization describes, target character " Yao " can be determined, by target word It accords with " Yao " and is used as candidate item, and only the candidate item is shown.
The candidate item and the corresponding candidate item of character string that Fig. 2 B and Fig. 2 C show a kind of pair of target character are in same display The schematic diagram that region is shown.After the character string " sangetu " of input characterization description, according to the character of characterization description String, can determine target character " Yao ", and target character " Yao " is used as candidate item.The corresponding candidate of character string " sangetu " Item may include " three soil ", " three figures ", " umbrella trrellis diagram " and " three ".It as shown in Figure 2 B, can be by target character " Yao " Candidate item shown after the corresponding candidate item of character string " sangetu ".It as shown in Figure 2 C, can be by target character The candidate item of " Yao " is shown before the corresponding candidate item of character string " sangetu ".
The candidate item and the corresponding candidate item of character string that Fig. 2 D shows a kind of pair of target character are in different display areas point The schematic diagram not shown.For character string " sangetu ", the candidate item of target character can be " Yao ", and character string is corresponding Candidate item can be " three soil ", " three figure ", " umbrella trrellis diagram ", " three " and " three brothers ".It as shown in Figure 2 D, can be by word Candidate item display area of the corresponding candidate item of symbol string under character input region is shown, by the candidate item of target character Candidate item display area on the right of character input region is shown.
In order to which the subsequent input of user is convenient, when showing to target character, its representation can also be shown together. Specifically, when target character is Chinese character, the phonetic of available Chinese character is carried out Chinese characters and pinyin as the same candidate item Display.Fig. 2 D shows the schematic diagram of the display mode for target character.By Fig. 2 D it is found that candidate item 6 does not show only mesh Marking-up accords with " Yao ", also illustrates the phonetic of target character " Yao ", user can directly pass through phonetic in subsequent input character Input.
The above are a kind of specific implementations of input method provided in an embodiment of the present invention, are not knowing target character In the case where accurate representation, target character can be described, by obtaining the characterization of the user's input description Character string, since the character string and target character both correspond to the same description, character string can characterize the description, and the description The target character can be characterized, thus can determine target character according to character string, even if user does not know using input method When input, the accurate representation method of the target character can also automatically determine the target character, realization pair according to the character string Target character input, meets the demand of user, improves user experience.
In order to realize that target character inputs, key is to determine target character according to the character string of characterization description.Below from It characterizes the character string of characters' property description and the corresponding relationship of character, carry out semantic analysis to attribute description and based on interconnection The process for determining target character is introduced in multiple angles such as the search result of net.
In one possible implementation, it can be retouched including characters' property in preset attribute library with preset attribute library The corresponding relationship of the character string and the character stated can search character string pair from preset attribute library according to the corresponding relationship The target character answered.
In order to establish attribute library, the character string of characterization characters' property description can be first obtained, then establishes characterization character Attribute description character string and character between corresponding relationship.It is appreciated that after establishing attribute library, it can also be to attribute Library is updated, and can be realized to attribute library more by least one of modes such as addition, modification, deletion corresponding relationship Newly.
Wherein, the character string of characterization characters' property description can be pinyin string.The category of character can be characterized by obtaining Property description Chinese character string, which is converted into corresponding pinyin string, to establish the corresponding relationship of pinyin string and character.It obtains There are many implementations for the Chinese character string for taking characterization characters' property to describe.
In one possible implementation, the Chinese of characterization characters' property description can be artificially generated according to preset rules Word string.Preset rules may include preset structure rule, preset name rule, default meaning rule, preset relation rule isotactic At least one of then.By taking preset structure rule as an example, preset rules are illustrated.Preset structure rule can be include body The structural word of existing character structure information.For example, " one, one, the left side ... the right ... ", " one above ..., below two It is a ... " etc. similar statement.
It in alternatively possible implementation, can be analyzed based on the mass data in internet, extraction meets default Phrase, phrase or the sentence of rule describe the phrase, phrase or sentence that meet preset rules as characters' property.It is default Rule has hereinbefore been discussed in detail, and which is not described herein again.It illustrates how below with reference to specific example based on internet acquisition table Levy the Chinese character string of character attibute description.Here by preset relation rule for, as from internet extraction " as ... The phrase or sentence of word ", when extracting " word as family ", " word as son ", can by " word as family ", The Chinese character string of " word as son " as characterization characters' property description.
The above is only some examples for the Chinese character string for obtaining characterization characters' property description, and in the embodiment of the present invention, other can In the implementation of energy, the Chinese character string of characterization characters' property description can also be obtained by other means.
Due to Chinese character string characterization characters' property description, corresponding character can be determined according to the Chinese character string.And it characterizes The Chinese character string of characters' property description can be converted to pinyin string, therefore, can pass through the Chinese character of characterization characters' property description String establishes the corresponding relationship between the pinyin string and character of characterization characters' property description.
It is appreciated that the corresponding relationship can artificially carry out the pinyin string of attribute description and character according to preset rules It is corresponding.In order to make it easy to understand, illustrating.It, can be by the pinyin string of attribute description " three soil " according to charcter topology " sangetu " is corresponding with character " Yao ", establishes corresponding relationship.It can be by the spelling of attribute description " as son " according to character relation Sound string " xiangziyiyang " is corresponding with character " lonely ", establishes corresponding relationship.It, can be by attribute description " one according to character meaning Kind animal is used to describe gluttonous greediness " it is corresponding with character " being greedy for money or food ".As the extension of the embodiment of the present invention, preset rules be can also be Historical events or allusion etc., can be by the pinyin string " diyiweinvhuangdi " and word of attribute description " first empress " It is corresponding to accord with " Zhao ".
It should be noted that the corresponding relationship of pinyin string and character can be one-to-one in preset attribute library, it is also possible to Many-one.The hit rate of character can be improved using the corresponding character of the pinyin string of a variety of attribute descriptions.In some cases, The pinyin string of one attribute description can also be corresponded to multiple characters, that is to say, that the corresponding relationship of pinyin string and character can be with It is one-to-many.
Corresponding relationship can also be determined by analyzing network data according to network data analysis result.Specifically, After the Chinese character string for getting characterization character attibute description, it can be searched in internet and the matched search term of the Chinese character string Item extracts the corresponding character of Chinese character string according to the corresponding search result of search entry, so establish the pinyin string of Chinese character string with Corresponding relationship between the corresponding character of Chinese character string.
May exist a variety of attribute descriptions for same character, can will meet the pinyin string of the attribute description of preset condition Corresponding relationship is established with character.It, can also be by above-mentioned a variety of attribute descriptions and character in order to further increase the hit rate of character Establish corresponding relationship.Wherein, preset condition can be searching times condition, and searching times can be met to searching times condition Characters' property description carried out with character corresponding, corresponding relationship is stored in attribute library.
For example, being directed to target character " line in the Eight Diagrams ", there are many attribute descriptions, and such as " two X, one on the other ", " refuting for refuting is removed Horse " finds that more people is gone based on " horse is removed in refuting for refuting " this attribute description by carrying out analysis to network data Search, the corresponding pinyin string of the attribute description can be mapped with target character " line in the Eight Diagrams ".
The character string for characterizing characters' property description can also be English string.First obtain the English of characterization character attibute description String, it is in one possible implementation, similar with the pinyin string of attribute description is obtained, for example, can be according to preset rules people For the English string for generating characterization characters' property description；In alternatively possible implementation, it is also based on to internet In mass data analyzed, extract the English phrase for meeting preset rules, English phrase and/or english sentence etc. as table Levy the English string of attribute description.When establishing attribute library, the character for corresponding to the same attribute description content and English are gone here and there, built Vertical corresponding relationship.As a specific example, the category of characterization character can be generated from the length of character, the meaning of character etc. Property description English string, e.g., thelongestenglishword, theoppositeofstrange etc..Establish attribute library After, it is gone here and there according to English, that is, can determine the corresponding target character of English string.
As an example, its attribute description content can be determined according to English string " thelongestenglishword " For " longest English word ", it can determine that corresponding character is as follows according to the attribute description content:
Maisonpneumonoultramicroscopicsilicovolcanoconiosis。
Based on above procedure, English string " thelongestenglishword " and character can establish The corresponding relationship of " Maisonpneumonoultramicroscopicsilicovolcanoconiosis ".
As another specific example of the application, can determine to belong to according to English string " theoppositeofstrange " Property description content be " strange antonym ", according to the attribute description content it is found that strange antonym is familiar, that is, " familiar " is hence, it can be determined that the corresponding character of attribute description content is familiar, and then can establish English string The corresponding relationship of " theoppositeofstrange " and character " familiar ".
Above-described embodiment mainly determines target character according to the character string of attribute description and the corresponding relationship of character, makees For the extension of the embodiment of the present invention, attribute description content can also be determined according to character string, attribute description content is carried out semantic It analyzes and determines target character.
With reference to the accompanying drawing, it is described in detail and how semantic analysis is carried out to attribute description content, determine target character.
Fig. 3 show a kind of pair of attribute description content progress semantic analysis provided in an embodiment of the present invention and determines target character The flow chart of method, referring to figure 3., this method comprises:
S301: semantic analysis is carried out to the attribute description content, obtains the charcter topology of the target character.
It is appreciated that the attribute description for target character may include multiple types, it can be and charcter topology is retouched It states, can be the description to character meaning, be also possible to the description etc. to character title.Attribute description content is carried out semantic Analysis, can determine the type of attribute description, if attribute description includes the description to the charcter topology of target character, obtain mesh The charcter topology of marking-up symbol.
Wherein, charcter topology includes at least the component part information of character.In order to more accurately determine character, character is improved Hit rate, in some cases, charcter topology can also include each component part location information.By taking " three soil " as an example, The charcter topology of attribute description " three soil " includes the component part of character, by taking " soil above, following two soil " as an example, word According with structure can also include the positional relationship between each component part of character.
Illustrate below with reference to specific example.
If the character string of input is that " zuobianyigedanlirenjiayoubianyigemu " can according to the character string To obtain the attribute description content " one, left side list stands people and adds one wood in the right " for target character, to the attribute description content Semantic analysis is carried out, the similar statement such as " left side ... the right ... " and " above ... below ... ", Ke Yishi are passed through Not Chu the attribute description be the attribute description for including charcter topology, the charcter topology of target character can be obtained, in this example, Target character is tiled configuration, and left side radical " single vertical people ", the right is " wood ".
In the above examples, target character is Chinese character, when target character is symbol, can also carry out language to attribute description Justice analysis, obtains the structure of target character.If the character string of input is " zimuSzhongjianjiayishu ", according to the character The available attribute description content " among alphabetical S plus one is perpendicular " for the target character is gone here and there, by the attribute description content Semantic analysis is carried out, according to the similar statement of " centre ", " adding " or " subtracting ", "+", "-" etc., the attribute description can be identified for packet The attribute description for including charcter topology, it is specific semantic according to it is found that the character is neither tiled configuration nor up-down structure or encirclement Structure can be considered as absolute construction, be formed by alphabetical S and one is perpendicular.
It is to be appreciated that target character is not limited to Chinese character or symbol, it is also possible to the character of other forms, the knot of target character Structure is also not necessarily limited to structure cited by above-mentioned example, can also be that other charcter topology forms, the embodiment of the present invention do not do this It limits.
S302: the corresponding target character of the charcter topology is obtained.
Due to the component part of charcter topology and character it has been determined that each component part is carried out group according to charcter topology It closes, the corresponding target character of the available charcter topology.
In order to make it easy to understand, being illustrated below with the example in S301.
Target character is tiled configuration, and component part is respectively " single vertical people " and " wood ", by its according to tiled configuration into Row combination, available target character are " not ".
Target character is that absolute construction is combined when component part is respectively alphabetical " S " and " one is perpendicular ", can To obtain target character " $ ", i.e. dollar mark ().
The above are charcter topology is obtained by carrying out semantic analysis to attribute description content, is obtained and corresponded to according to charcter topology The implementation of target character can also be to attribute description content in the embodiment of the present invention in other possible implementations Semantic analysis is carried out to obtain character title perhaps character meaning according to character title or character meaning obtains corresponding target word Symbol.It is described in detail with reference to the accompanying drawing.
Fig. 4 show a kind of pair of attribute description content progress semantic analysis provided in an embodiment of the present invention and determines target character The flow chart of method, referring to figure 4., this method comprises:
S401: semantic analysis is carried out to the attribute description content, obtains the character title of the target character.
In some cases, user may know that character title, but not know how to input the character, or know word Title is accorded with, but does not know the corresponding specific numerical value of character or the corresponding manifestation mode of character, for this purpose, can input can The character string for characterizing character name attribute, determines corresponding attribute description according to character string, carries out semantic analysis to attribute description, Obtain the character title of target character.
In order to make it easy to understand, being exemplified below.
User wants to calculate the area of a certain geometric figure, it is desirable to which the numerical value for inputting pi is calculated, however user Forget circumference rate score, have input character string " yuanzhoulv " thus, obtains attribute description " pi ", it can be from the attribute The character title " pi " of target character is obtained in description.
User wants to input certain formula in some cases, but does not know to embody mode to a certain physical quantity, for example, When calculating magnetic flux, it is not known that how magnetic flux indicates, can input character string " citongliang ", can be with according to character string Obtaining attribute description is " magnetic flux ", and the character of available target character is entitled " magnetic flux " from the attribute description.
S402: the corresponding target character of the character title is obtained.
In this step, the corresponding relationship between being referred to as according to target character and character name, obtains and obtains in S401 The corresponding target character of character title.In the embodiment of the present invention in other possible implementations, other can also be used Mode obtains target character corresponding with character title.
In order to make it easy to understand, being still illustrated below with the example in S401.
When character entitled " pi ", can be determined according to the corresponding relationship between pi and its specific value " pi " corresponding target character is digital " 3.14 ".
When the entitled magnetic flux of character, according to the corresponding relationship between physical quantity and its symbol, magnetic flux can be determined Expression symbol be " Φ ", namely obtain target character " Φ ".
In the present embodiment, target character is not limited to number or symbol, is also possible to other targets with character title Character, it is not limited in the embodiment of the present invention.
The above are a kind of pair of attribute description contents provided in an embodiment of the present invention to carry out semantic analysis, obtains target character Character title obtains the specific implementation of the corresponding target character of character title.Since there are multiple types, examples for attribute description Such as character meaning, semantic analysis can be carried out to attribute description, obtain the character meaning of target character, it is corresponding to obtain character meaning Target character.Semantic analysis is carried out to attribute description content, obtains the character meaning of target character, it is corresponding to obtain character meaning The implementation of target character be referred to the realization side according to charcter topology or the corresponding target character of character name acquiring Formula is no longer described in detail here.
It is appreciated that in order to improve hit rate, it in some cases, can also be semantic according to being carried out to attribute description content Analysis obtains at least one of charcter topology, character title or character meaning etc., according to charcter topology, character title or word It accords at least one of meaning etc. and obtains corresponding target character.By taking attribute description " symbol of pi " as an example, in some feelings Under condition, user may want to input the expression symbol of pi, rather than specific value, if only according to character title " circumference Numerical value " 3.14 " is then probably determined as target character by rate ", and realistic objective character is " π ", is unable to satisfy user's input The demand of target character.For this purpose, can in conjunction with attribute description " symbol of pi " character title " pi " and " character contains Justice " determines target character jointly, to improve the hit rate of target character.
In addition to the corresponding relationship of the character string and target character that are described according to characterization attributes in attribute library, or attribute is retouched It states content progress semantic analysis to determine outside target character, can also determine target word according to the search behavior data in internet Symbol.Since user is not knowing the accurate manifestation mode of character, when leading to not input character, often sought by internet It helps, such as is searched by search engine, or post inquiry in interactive forum or space.Fig. 5 shows a kind of user and is directed to The schematic diagram that target character is searched in a search engine.As shown in Figure 5, have more user with " three soil read what ", " what three soil read ", " how three soil are read " implement search behavior, are based on this, can when user is intended to input target character To determine target character according to the corresponding search result of search entry according to the matched search entry of attribute description content search.
The method for obtaining target character according to the search result of internet is introduced with reference to the accompanying drawing.
Fig. 6 show a kind of flow chart of method that target character is determined according to character string provided in an embodiment of the present invention, Wherein, character string is the pinyin string of attribute description, please refers to Fig. 6, this method comprises:
S601: attribute description content is determined according to the character string.
Due to the pinyin string that character string is attribute description, which is grouped, each group of phonetic can be converted to Chinese character the Chinese character after conversion combines available attribute description.Such as pinyin string " shangmianyigetuxiamianlianggetu ", is grouped, available following pinyin string group " shang ' mian ' Yi ' ge ' tu ' xia ' mian ' liang ' ge ' tu ", the Chinese character group after conversion are combined into " above one native following two soil ", can be with As the attribute description of pinyin string " shangmianyigetuxiamianlianggetu ".
S602: the search entry with the attribute description content matching is searched.
After obtaining attribute description, it can search and the matched search entry of the attribute description.It is matched with attribute description Search entry, can for with attribute description same keyword, and can be used for search for phrase, phrase or sentence Deng.Wherein, keyword can be understood as to the helpful word of target character is determined, may include structural word, numeral-classifier compound, characterization group At word and proper noun etc..With the matched search entry of attribute description, have with attribute description same keyword, can be with It is interpreted as the keyword that the search entry includes and the keyword that attribute description includes corresponds, if having in search entry more Remaining keyword lacks keyword, then is not intended as matched search entry.
Matched search entry is illustrated by taking Fig. 5 as an example, for attribute description " three soil ", including keyword " three It is a " and " soil ", and search entry " what three soil read ", " how three soil are read ", " how three five, soil beaten ", " three soil What is read ", " how three soil are beaten ", " what is read ", " how reading " etc. can't be to determining character band in " what word three soil are " It helps, is not belonging to keyword, thus the keyword for including in above-mentioned search entry is identical as the keyword that attribute description includes, Can be considered as with the matched search entry of attribute description, and " π below three soil ", " three soil add π ", " three natives " also Including " following ", "one", the keywords such as " π " and " people ", thus cannot as with the matched search term of attribute description Item.
The search entry for searching attribute description content matching can be there are many implementation, can be in the input of search engine Attribute description content is inputted in frame, recommendation search entry can be presented in the combobox of input frame, wrap according in recommendation search entry The keyword included can be obtained therefrom and the matched search entry of attribute description.Can also in question and answer website, such as know, hundred Degree, which is known, waits websites to search and the matched search entry of attribute description.It is not limited in the embodiment of the present invention.
In order to make it easy to understand, being introduced below by example way to the matched search entry of attribute description is searched.
For attribute description content " soil above, following two soil ", it is inputted the search input frame of search engine In, occur " what figure following two soil above reads ", " what word one above native following two soil is " etc. in combobox Search entry, can using one of them as with the matched search entry of attribute description, to scan for.It is appreciated that with The quantity of the matched search entry of attribute description can be one, or multiple.When matched search entry quantity is more When a, in order to improve the hit rate of target character, it can be carried out the more entry of searching times as matched search entry Search.
It in some cases, is not in the search entry of recommendation in the combobox of search box, it can be directly by input Attribute description in frame is scanned for as with the matched search entry of attribute description.For example, being directed to the attribute description, " left side one A list founds people, one, the right wood ", it can directly be scanned for as with the matched search entry of attribute description.
S603: target character is determined according to the corresponding search result of described search entry.
It is got in S602 with after the matched search entry of attribute description, is scanned for, can be obtained with the search entry To at least one search result, according to the search result, available target character corresponding with attribute description.One kind can Can implementation in, can be using the encyclopaedia page of search entry as search result corresponding with search entry, or can be with By number of clicks in the search behavior based on the search entry or the most page of number is opened, it is corresponding as search entry Search result by number of clicks in the search behavior based on the search entry or can also open number more than or equal to default threshold The page of value is as the corresponding search result of search entry.Include target character in mentioned above searching results, also may include and mesh Marking-up accords with unrelated character, can be by carrying out semantic analysis to search result, to extract target character.
In order to make it easy to understand, illustrating.
For the search behavior of attribute description " one above native following two soil ", available multiple search results can It include more the encyclopaedia page as search result corresponding with search entry, is opened the encyclopaedia page, in the encyclopaedia page Character, such as " Yao (yao), ancient times are the same as " Yao ".Mean the high appearance ... in mountain " in included character, can be to the encyclopaedia page Semantic analysis is carried out, extracting the corresponding target character of attribute description " one above native following two soil " is " Yao ", pronunciation For yao, look like in ancient times identical with Yao.
Above-described embodiment is to be converted to attribute description content by the pinyin string for describing characterization attributes, and retouch according to attribute The search entry for stating content acquisition and attribute description content matching, determines target character according to the search result of search entry.? In some possible implementations, it can also directly be scanned for the pinyin string that characterization attributes describe, obtain search result, into And determine target character.
By characterization attributes description pinyin string " sangetu " for, in the input frame of search engine input sangetu into Row search, sangetu can be identified as three soil automatically according to search history data etc. by search engine, and show three soil Corresponding search result, according to the available target character of the search result " Yao ".Fig. 7, which is shown, directly to be searched with pinyin string The search result schematic diagram of rope.In this example, it can also still be scanned for using sangetu as search entry.
The above are a kind of search behaviors Internet-based provided in an embodiment of the present invention to determine the method for target character Specific implementation.By obtaining the corresponding attribute description of character string, searched from internet and the attribute description is matched searches Rope entry, according to the corresponding search result acquisition target character of search entry, the mass data as present in internet, thus This method can identify that application range is more extensive to more character, can be used as the side that character is determined based on attribute library The supplement of method.
When above-described embodiment is mainly the pinyin string of attribute description from character string, the angle pair of target character how is inputted Input method provided in an embodiment of the present invention is introduced.In some possible implementations, character string can also be attribute How the English string of description inputs target character and is introduced with reference to the accompanying drawing to when character string is English string.
Target character is defeated when Fig. 8 show the English string that a kind of character string provided in an embodiment of the present invention is attribute description Enter the flow chart of method.Fig. 8 is please referred to, this method comprises:
S801: English is gone here and there and carries out semantic analysis.
Semantic analysis is carried out to English string and namely analyzes the expressed meaning of English string.Since English string can characterize To the attribute description of target character, it can determine that above-mentioned attribute description is that the attribute of which kind is retouched by semantic analysis result It states, for example, for the description of spelling of target character, the description of pronunciation or the description for expressing meaning etc..
In this step, going here and there progress semantic analysis to English can be there are many implementation.In a kind of possible realization side In formula, English can be gone here and there based on neural network model and carry out semantic analysis.May be used also before carrying out semantic analysis to English string First to be segmented to English string, preferably to carry out semantic analysis.
This step is illustrated below with reference to specific example.
The English string for describing target character attribute is " theoppositeofstrange ", carries out semanteme going here and there to the English First English string can be segmented before analysis, word segmentation result is following " the/opposite/of/strange ", is based on this point Word result carries out semantic analysis.The process of semantic analysis can pass through nerve for word segmentation result is input to neural network model The deep learning of network model, can the expression meaning to the word segmentation result of input predict, can using prediction result as The semantic analysis result of above-mentioned English string.In this example, semantic analysis result can be " strange antonym ", it is possible to understand that For the description of the expression meaning for target character.
S802: the target character with English string semantic matches is obtained.
Semantic analysis is carried out going here and there to English, after obtaining semantic analysis result, available and English string semantic matches Target character.Obtaining can be there are many implementation with the target character of English string semantic matches.In a kind of possible realization In mode, the search entry to match with semantic analysis result can be inquired in internet, obtain that search entry is corresponding to search Rope is as a result, to obtain target character.It, can be according to the English pre-established in attribute library as the extension of the embodiment of the present invention The corresponding relationship of string and target character searches target character corresponding with English string in attribute library.
Below with reference to specific example, it is introduced to obtaining with the process of the target character of English string semantic matches.
After obtaining semantic analysis result " strange antonym ", it is strange anti-can to determine that target character can be expressed as Adopted word, that is to say, that the meaning that target character indicates be it is familiar, accordingly, target character can be determined for " familiar ".Wherein, The process for obtaining familiar may be that the corresponding target character of English string is searched from preset attribute library.Attribute library can be with It pre-establishes, may include the corresponding relationship of the English string and character of attribute description.It in this example, can be in attribute library Including the corresponding relationship between English string " theoppositeofstrange " and character " familiar ", closed according to the correspondence System, can determine target character " familiar ".
Fig. 9 is referred to using the effect picture of input method provided in an embodiment of the present invention.Fig. 9 show a kind of character string The effect diagram of target character input when being gone here and there for the English of attribute description.As shown in figure 9, in input English string After " theoppositeofstrange ", which can be gone here and there and carry out semantic analysis, obtain the mesh with English string semantic matches Marking-up accords with " familiar ", and target character " familiar " is shown as candidate item, so that user selects the candidate item, Realize the input of target character " familiar ".
The above are a kind of characters when character string is English string, described according to characterization attributes provided in an embodiment of the present invention String determines the specific implementation of target character.When user's intention input English word, but the specific spelling of English word is not known When WriteMode, it can use natural language and attribute description carried out to the English word, obtain the English string for characterizing the attribute description, Since English string and English word both correspond to same based on natural language attribute description obtained, English string being capable of table The attribute description is levied, which can characterize the English word, thus can go here and there according to the English and determine that the English is single Word.Semantic analysis is carried out by going here and there to English, according to semantic analysis result, the available and English semantic English to match of string Word namely target character, show target character as candidate item, and user can be made to directly select candidate item reality Now to the input of target character, is operated without additional search etc., improve user experience.
The specific implementation of input method based on the embodiment of the present invention, the embodiment of the invention also provides one kind Input unit is below illustrated input unit provided by the embodiments of the present application from the angle of function modoularization.
Figure 10 show the structural schematic diagram of input unit provided in an embodiment of the present invention, please refers to Figure 10, the device packet It includes:
Acquiring unit 1001, for obtaining the character string of input, character string is used to characterize the description to target character；
Determination unit 1002, for determining target character according to character string；
Display unit 1003, for being shown target character as candidate item.
Optionally, determination unit 1002 is specifically used for:
From preset attribute library, the corresponding target character of character string is searched, preset attribute library includes characterization character The character string of attribute description and the corresponding relationship of character.
Optionally, determination unit 1002 includes:
Attribute description content determines subelement, for determining attribute description according to character string when character string is pinyin string Content；
Semantic analysis subelement determines target character for carrying out semantic analysis to attribute description content.
Optionally, semantic analysis subelement includes:
Charcter topology obtains subelement, for carrying out semantic analysis to attribute description content, obtains the character of target character Structure；
Target character obtains subelement, for obtaining the corresponding target character of charcter topology.
Optionally, semantic analysis subelement includes:
Character name acquiring subelement obtains the character of target character for carrying out semantic analysis to attribute description content Title；
Target character obtains subelement, for obtaining the corresponding target character of character title.
Optionally, determination unit 1002 includes:
Attribute description content determines subelement, for determining attribute description according to character string when character string is pinyin string Content；
Subelement is searched, for searching and the search entry of attribute description content matching；
Target character determines subelement, for determining target character according to the corresponding search result of search entry.
Optionally, display unit 1003 includes:
Phonetic obtains subelement, for obtaining the phonetic of Chinese character when target character is Chinese character；
Subelement is shown, for showing the phonetic of Chinese character and Chinese character as the same candidate item.
Optionally, determination unit 1002 includes:
Semantic analysis subelement, for being gone here and there to English and carrying out semantic analysis when character string is English string；
Target character obtains subelement, for obtaining the target character with English string semantic matches.
Device shown in Fig. 10 is and device corresponding to method shown in FIG. 1 specific implementation and side shown in FIG. 1 Method is similar, and with reference to the description in method shown in FIG. 1, which is not described herein again.
The input unit provided through the embodiment of the present invention, when using input method input, when user is not knowing target In the case where the accurate representation of character, target character can be described, obtain the character string for characterizing the description, so Afterwards, target character is determined according to the character string, and is shown target character as candidate item.Wherein, which can To characterize the description to target character, therefore, the character string and target character both correspond to the same description, and character string being capable of table The description is levied, and the description can characterize the target character, it therefore, should even if user is not known when being inputted using input method The accurate representation method of target character, input method can automatically determine the target character according to the character string, realize to the mesh The input of marking-up symbol, improves user experience.
Above-described embodiment is described input unit provided in an embodiment of the present invention from the angle of function modoularization, under Face is illustrated input transposition provided in an embodiment of the present invention from the angle of hardware.
Figure 11 show a kind of hardware structural diagram of input unit provided in an embodiment of the present invention, please refers to Figure 11, Input unit 1100 may include following one or more components: processing component 1102, memory 1104, power supply module 1106, Multimedia component 1108, audio component 1110, the interface 1112 of input/output (I/O), sensor module 1114, and communication Component 1116.
The integrated operation of the usual control device 1100 of processing component 1102, such as with display, telephone call, data communication, Camera operation and record operate associated operation.Processing component 1102 may include one or more processors 1120 to execute Instruction, to perform all or part of the steps of the methods described above.In addition, processing component 1102 may include one or more moulds Block, convenient for the interaction between processing component 1102 and other assemblies.For example, processing component 1102 may include multi-media module, To facilitate the interaction between multimedia component 1108 and processing component 1102.
Memory 1104 is configured as storing various types of data to support the operation in equipment 1100.These data Example includes the instruction of any application or method for operating on device 1100, contact data, telephone book data, Message, picture, video etc..Memory 1104 can by any kind of volatibility or non-volatile memory device or they Combination is realized, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), it is erasable can Program read-only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash memory Reservoir, disk or CD.
Power supply module 1106 provides electric power for the various assemblies of device 1100.Power supply module 1106 may include power management System, one or more power supplys and other with for device 1100 generate, manage, and distribute the associated component of electric power.
Multimedia component 1108 includes the screen of one output interface of offer between described device 1100 and user.? In some embodiments, screen may include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, Screen may be implemented as touch screen, to receive input signal from the user.Touch panel includes that one or more touch passes Sensor is to sense the gesture on touch, slide, and touch panel.The touch sensor can not only sense touch or sliding is dynamic The boundary of work, and further acknowledge duration and pressure associated with the touch or slide operation.In some embodiments, more Media component 1108 includes a front camera and/or rear camera.When equipment 1100 is in operation mode, as shot mould When formula or video mode, front camera and/or rear camera can receive external multi-medium data.Each preposition camera shooting Head and rear camera can be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio component 1110 is configured as output and/or input audio signal.For example, audio component 1110 includes a wheat Gram wind (MIC), when device 1100 is in operation mode, when such as call mode, recording mode, and voice recognition mode, microphone quilt It is configured to receive external audio signal.The received audio signal can be further stored in memory 1104 or via communication Component 1116 is sent.In some embodiments, audio component 1110 further includes a loudspeaker, is used for output audio signal.
I/O interface 1112 provides interface, above-mentioned peripheral interface module between processing component 1102 and peripheral interface module It can be keyboard, click wheel, button etc..These buttons may include, but are not limited to: home button, volume button, start button and Locking press button.
Sensor module 1114 includes one or more sensors, and the state for providing various aspects for device 1100 is commented Estimate.Such as the state that opens/closes of equipment 1100, the relative positioning of component, such as institute can be confirmed in sensor module 1114 The display and keypad that component is device 1100 are stated, sensor module 1114 can also confirm that device 1100 or device 1,100 1 The position change of a component, the existence or non-existence that user contacts with device 1100,1100 orientation of device or acceleration/deceleration and dress Set 1100 temperature change.Sensor module 1114 may include proximity sensor, be configured in not any physics Presence of nearby objects are confirmed when contact.Sensor module 1114 can also include optical sensor, as CMOS or ccd image are sensed Device, for being used in imaging applications.In some embodiments, which can also include acceleration sensing Device, gyro sensor, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 1116 is configured to facilitate the communication of wired or wireless way between device 1100 and other equipment.Dress The wireless network based on communication standard, such as WiFi can be accessed by setting 1100,2G or 3G or their combination.It is exemplary at one In embodiment, communication component 1116 receives broadcast singal or broadcast correlation from external broadcasting management system via broadcast channel Information.In one exemplary embodiment, the communication component 1116 further includes near-field communication (NFC) module, to promote short distance Communication.For example, radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band can be based in NFC module (UWB) technology, bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, device 1100 can be by one or more application specific integrated circuit (ASIC), number Signal processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for executing the above method.
Specifically, the embodiment of the invention provides a kind of electronic equipment, which can be specially device 1100, include Perhaps more than one program one of them or more than one program is stored in memory by memory 1104 and one In 1104, and be configured to by one or more than one processor 1120 execute the one or more programs include Instruction for performing the following operation:
The character string of input is obtained, character string is used to characterize the description to target character；
Target character is determined according to character string；
It is shown target character as candidate item.
Optionally, determine that target character includes: according to character string
From preset attribute library, the corresponding target character of character string is searched, preset attribute library includes characterization character The character string of attribute description and the corresponding relationship of character.
Optionally, character string is the pinyin string of attribute description, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Semantic analysis is carried out to attribute description content and determines target character.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the charcter topology of target character；
Obtain the corresponding target character of charcter topology.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the character title of target character；
Obtain the corresponding target character of character title.
Optionally, character string is the pinyin string of attribute description, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Search the search entry with attribute description content matching；
Target character is determined according to the corresponding search result of search entry.
Optionally, target character is carried out display as candidate item includes:
When target character is Chinese character, the phonetic of Chinese character is obtained；
It is shown the phonetic of Chinese character and Chinese character as the same candidate item.
Optionally, character string includes English string, determines that target character includes: according to character string
English is gone here and there and carries out semantic analysis；
Obtain the target character with English string semantic matches.
The embodiment of the invention also provides a kind of non-transitorycomputer readable storage medium including instruction, for example including The memory 1104 of instruction, above-metioned instruction can be executed by the processor 1120 of device 1100 to complete the above method.For example, described Non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk and light Data storage device etc..
A kind of non-transitorycomputer readable storage medium, when the instruction in the storage medium is by the processing of electronic equipment When device executes, so that electronic equipment is able to carry out a kind of input method, which comprises
The character string of input is obtained, character string is used to characterize the description to target character；
Target character is determined according to character string；
It is shown target character as candidate item.
Optionally, determine that target character includes: according to character string
From preset attribute library, the corresponding target character of character string is searched, preset attribute library includes characterization character The character string of attribute description and the corresponding relationship of character.
Optionally, character string includes pinyin string, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Semantic analysis is carried out to attribute description content and determines target character.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the charcter topology of target character；
Obtain the corresponding target character of charcter topology.
Optionally, semantic analysis is carried out to attribute description content and determines that target character includes:
Semantic analysis is carried out to attribute description content, obtains the character title of target character；
Obtain the corresponding target character of character title.
Optionally, character string includes pinyin string, determines that target character includes: according to character string
Attribute description content is determined according to character string；
Search the search entry with attribute description content matching；
Target character is determined according to the corresponding search result of search entry.
Optionally, target character is carried out display as candidate item includes:
When target character is Chinese character, the phonetic of Chinese character is obtained；
It is shown the phonetic of Chinese character and Chinese character as the same candidate item.
Optionally, character string includes English string, determines that target character includes: according to character string
English is gone here and there and carries out semantic analysis；
Obtain the target character with English string semantic matches.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment it Between same and similar part may refer to each other, each embodiment focuses on the differences from other embodiments. For equipment and system embodiment, since it is substantially similar to the method embodiment, so describe fairly simple, The relevent part can refer to the partial explaination of embodiments of method.Equipment and system embodiment described above is only schematic , wherein unit may or may not be physically separated as illustrated by the separation member, it is shown as a unit Component may or may not be physical unit, it can and it is in one place, or may be distributed over multiple networks On unit.Some or all of the modules therein can be selected to achieve the purpose of the solution of this embodiment according to the actual needs. Those of ordinary skill in the art can understand and implement without creative efforts.
The above, a kind of specific embodiment only of the invention, but scope of protection of the present invention is not limited thereto, In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by anyone skilled in the art, It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with scope of protection of the claims Subject to.

Claims (11)

1. a kind of input method, which is characterized in that the described method includes:
The character string of input is obtained, the character string is used to characterize the description to target character；
The target character is determined according to the character string；
It is shown using the target character as candidate item.
2. the method according to claim 1, wherein described determine the target character packet according to the character string It includes:
From preset attribute library, the corresponding target character of the character string is searched, the preset attribute library includes characterization word The character string of the attribute description of symbol and the corresponding relationship of character.
3. described according to the word the method according to claim 1, wherein the character string includes pinyin string Symbol string determines that the target character includes:
Attribute description content is determined according to the character string；
Semantic analysis is carried out to the attribute description content and determines the target character.
4. according to the method described in claim 3, it is characterized in that, described true to attribute description content progress semantic analysis Determining the target character includes:
Semantic analysis is carried out to the attribute description content, obtains the charcter topology of the target character；
Obtain the corresponding target character of the charcter topology.
5. according to the method described in claim 3, it is characterized in that, described true to attribute description content progress semantic analysis Determining the target character includes:
Semantic analysis is carried out to the attribute description content, obtains the character title of the target character；
Obtain the corresponding target character of the character title.
6. described according to the word the method according to claim 1, wherein the character string includes pinyin string Symbol string determines that the target character includes:
Attribute description content is determined according to the character string；
Search the search entry with the attribute description content matching；
The target character is determined according to the corresponding search result of described search entry.
7. method described in -6 any one according to claim 1, which is characterized in that described using the target character as candidate Item carries out display
When the target character is Chinese character, the phonetic of the Chinese character is obtained；
It is shown the phonetic of the Chinese character and the Chinese character as the same candidate item.
8. the method according to claim 1, wherein the character string include English string, it is described according to the word Symbol string determines that the target character includes:
Semantic analysis is carried out to the English string；
Obtain the target character with the English string semantic matches.
9. a kind of input unit, which is characterized in that described device includes:
Acquiring unit, for obtaining the character string of input, the character string is used to characterize the description to target character；
Determination unit, for determining the target character according to the character string；
Display unit, for being shown using the target character as candidate item.
10. a kind of electronic equipment, which is characterized in that include memory and one or more than one program, wherein one A perhaps more than one program is stored in memory and is configured to execute described one by one or more than one processor A or more than one program includes the instruction for performing the following operation:
The character string of input is obtained, the character string is used to characterize the description to target character；
The target character is determined according to the character string；
It is shown using the target character as candidate item.
11. a kind of non-transitorycomputer readable storage medium, which is characterized in that when the instruction in the storage medium is by electronics When the processor of equipment executes, so that a kind of method that electronic equipment is able to carry out keyboard operation, which comprises
The character string of input is obtained, the character string is used to characterize the description using to target character；
The target character is determined according to the character string；
It is shown using the target character as candidate item.
CN201711463446.8A 2017-12-28 2017-12-28 Input method and input device Active CN109976548B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711463446.8A CN109976548B (en) 2017-12-28 2017-12-28 Input method and input device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711463446.8A CN109976548B (en) 2017-12-28 2017-12-28 Input method and input device

Publications (2)

Publication Number Publication Date
CN109976548A true CN109976548A (en) 2019-07-05
CN109976548B CN109976548B (en) 2022-07-19

Family

ID=67075109

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711463446.8A Active CN109976548B (en) 2017-12-28 2017-12-28 Input method and input device

Country Status (1)

Country Link
CN (1) CN109976548B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111753175A (en) * 2020-06-28 2020-10-09 恩亿科(北京)数据科技有限公司 Entry display method, entry display device and readable storage medium

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1949148A (en) * 2006-07-25 2007-04-18 网之易信息技术（北京）有限公司 Chinese characters inputting method and device
CN101615084A (en) * 2009-08-11 2009-12-30 腾讯科技（深圳）有限公司 Chinese character input method and device
CN101702100A (en) * 2009-10-28 2010-05-05 卓望数码技术（深圳）有限公司 Word input method and word input device
CN101727193A (en) * 2008-10-28 2010-06-09 英业达股份有限公司 System for inputting Cantonese characters by combining Chinese characters and method thereof
CN103076892A (en) * 2012-12-31 2013-05-01 百度在线网络技术（北京）有限公司 Method and equipment for providing input candidate items corresponding to input character string
WO2014000263A1 (en) * 2012-06-29 2014-01-03 Microsoft Corporation Semantic lexicon-based input method editor
CN104166462A (en) * 2013-05-17 2014-11-26 北京搜狗科技发展有限公司 Input method and system for characters
CN107247519A (en) * 2017-08-16 2017-10-13 北京搜狗科技发展有限公司 A kind of input method and device
CN107315487A (en) * 2016-04-27 2017-11-03 北京搜狗科技发展有限公司 A kind of input processing method, device and electronic equipment

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1949148A (en) * 2006-07-25 2007-04-18 网之易信息技术（北京）有限公司 Chinese characters inputting method and device
CN101727193A (en) * 2008-10-28 2010-06-09 英业达股份有限公司 System for inputting Cantonese characters by combining Chinese characters and method thereof
CN101615084A (en) * 2009-08-11 2009-12-30 腾讯科技（深圳）有限公司 Chinese character input method and device
CN101702100A (en) * 2009-10-28 2010-05-05 卓望数码技术（深圳）有限公司 Word input method and word input device
WO2014000263A1 (en) * 2012-06-29 2014-01-03 Microsoft Corporation Semantic lexicon-based input method editor
CN103076892A (en) * 2012-12-31 2013-05-01 百度在线网络技术（北京）有限公司 Method and equipment for providing input candidate items corresponding to input character string
CN104166462A (en) * 2013-05-17 2014-11-26 北京搜狗科技发展有限公司 Input method and system for characters
CN107315487A (en) * 2016-04-27 2017-11-03 北京搜狗科技发展有限公司 A kind of input processing method, device and electronic equipment
CN107247519A (en) * 2017-08-16 2017-10-13 北京搜狗科技发展有限公司 A kind of input method and device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111753175A (en) * 2020-06-28 2020-10-09 恩亿科(北京)数据科技有限公司 Entry display method, entry display device and readable storage medium
CN111753175B (en) * 2020-06-28 2024-06-07 恩亿科(北京)数据科技有限公司 Entry display method, entry display device and readable storage medium

Also Published As

Publication number Publication date
CN109976548B (en) 2022-07-19

Similar Documents

Publication Publication Date Title
KR102603466B1 (en) Method for training a voiceprint extraction model and method for voiceprint recognition, and device and medium thereof, program
CN104836720B (en) Method and device for information recommendation in interactive communication
US20170154104A1 (en) Real-time recommendation of reference documents
WO2020107813A1 (en) Method and apparatus for positioning descriptive statement in image, electronic device and storage medium
CN109145213B (en) Historical information based query recommendation method and device
CN110781305A (en) Text classification method and device based on classification model and model training method
CN108008832A (en) A kind of input method and device, a kind of device for being used to input
CN116166843B (en) Text video cross-modal retrieval method and device based on fine granularity perception
WO2019109663A1 (en) Cross-language search method and apparatus, and apparatus for cross-language search
CN109002184A (en) A kind of association method and device of input method candidate word
CN110019675A (en) A kind of method and device of keyword extraction
KR102580017B1 (en) Voice packet recommendation methods, devices, facilities and storage media
CN109471919A (en) Empty anaphora resolution method and device
CN111984749A (en) Method and device for ordering interest points
CN108345625A (en) A kind of information mining method and device, a kind of device for information excavating
CN111538830B (en) French searching method, device, computer equipment and storage medium
CN108334526A (en) The methods of exhibiting and device of search result items
CN109408652B (en) Picture searching method, device and equipment
CN109977390A (en) A kind of method and device generating text
CN113033163A (en) Data processing method and device and electronic equipment
CN108073293B (en) Method and device for determining target phrase
JP7208968B2 (en) Information processing method, device and storage medium
US20230326369A1 (en) Method and apparatus for generating sign language video, computer device, and storage medium
CN109976548A (en) A kind of input method and input unit
CN113157966B (en) Display method and device and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant